<!DOCTYPE html>

<html dir="ltr" lang="en"><head>
<!--
  ══ v57 · THE THREE PAGE DEFECTS THE PROCESS AUDIT FOUND ════════════
  Strings are unchanged. These are all markup and script.

    1. PER-LANGUAGE FONT STACKS. v56 had ZERO html[lang] rules. The only
       lang-scoped font rules were on .lm-body, the lead-magnet modal, so
       ja ko zh-Hans zh-Hant ar th rendered the whole page in the Latin
       fallback. Added at html[lang], so they also apply to the static
       per-language pages the split produces. Thai also gets extra
       line-height, or tone marks collide with the line above.

    2. <html lang> NOW WINS OVER navigator.language. v56's init() read
       only the browser. A visitor with an English browser on the German
       page got an English modal, and applyPageLang overwrote the baked
       German CTA text from the inline STRINGS dictionary. The browser is
       now only a hint, and only on the English root.

    3. THE BARE NUMERALS ARE KEYED. v55 declared 4.8, 189, 1,800+ and 21
       "not language". That was wrong: thousands separators and the
       decimal mark ARE language. The German page showed 1,800+ in the
       stats strip and 1.800+ in a keyed string 400px away. New keys:
       hero_rating, stat2_num, stat3_num, stat4_num.
  ════════════════════════════════════════════════════════════════════
-->
<!--
  ══ v56 · JEFF'S FIXES, 2026-08-06 ══════════════════════════════════
  Four changes on top of v55. This is the version that goes up Monday.
  There is no v57 planned.

    1. FAQ LINK. "Read all FAQs →" pointed at https://tv.eslpod.com/p/faqs/,
       a Teachable page. Now https://www.eslpod.com/faqs/. The other three
       tv.eslpod.com links (Contact, Privacy, Terms) are untouched, because
       those pages really do live on Teachable. Flagging in case they should
       move too.
    2. ONE FREE-LESSON WORDING. The footer link said "Get a Free Lesson!"
       while the hero button said "Get a free lesson", both on the single key
       cta_trigger. It is now "Get a free lesson" in both places, and the
       inline STRINGS.cta_trigger dictionary was relevelled in all 17
       languages to drop the exclamation mark and the Spanish opening ¡, so
       the register matches the English.
    3. LGM_BASE SWAPPED to https://www.eslpod.com/lgm/. The obsolete
       DO-NOT-SWAP comment above it is gone, and so is the console guard at
       the bottom of <body> that watched for the interim path.
    4. Thai testimonial slots were rebalanced. That is a change to the
       locale map, not to this file.
  ════════════════════════════════════════════════════════════════════
-->
<!--
  ══ v55 · FULL data-i18n COVERAGE ═══════════════════════════════════
  v55 adds data-i18n to every remaining unit of visible English on the
  page. It changes NO copy, NO CSS, and NO script. Every edit is either
  a new attribute or a wrapper span carrying text that already rendered
  in the same position, so v55 renders pixel-identical to v54 in English.

    GROUP 1  curriculum topics grid — 16 headings + 16 example lists
    GROUP 2  nine testimonial cards — quote, flag, name, location
    GROUP 3  hero and stat chrome — "1 billion", "15M", "yrs",
             and the portrait tag Ph.D. line
    GROUP 4  seven aria-labels and four image alt strings

  THE "·" SEPARATOR MOVED OUT OF THE TRANSLATABLE STRINGS. It used to
  sit at the front of the testimonial location text and at the front of
  the portrait Ph.D. line, which is the same shape as the
  ue_price_billed leading space that has already caused trouble, and it
  would have been repeated across 16 locales x 9 cards. It now lives in
  markup: one aria-hidden .ep-portrait-sep span in the hero, and one
  .ep-tcard-loc::before rule for the cards. The cards use CSS rather
  than a span because .ep-tcard-meta is a flex row with gap:8px, so an
  extra element there would have added 8px of gap. That CSS rule is the
  ONLY style change in v55. Rendering is byte-identical to v54.

  RENDERER GAPS TO FIX BEFORE THE SPLIT. Two, both pre-existing:
    a. render_multilang_pages.py sets el.string = S[k], which flattens
       any inline tag inside a keyed element. Keys carrying inline HTML
       (hero_h1, stat4, testiN_quote) will lose or escape their markup.
       It needs to write innerHTML, not string.
    b. That renderer only walks [data-i18n]. It never applies
       -placeholder, -aria, -alt or -option variants. Group 4 above and
       the existing popup placeholder keys are inert until it does.

  Stars (★★★★★), bare numerals (189, 1,800+, 21, 4.8) and UI glyphs
  (☰ ▾ ‹ › 🔒 🛡️) are deliberately NOT keyed. They carry no language.
  ════════════════════════════════════════════════════════════════════
-->
<!--
  ══ TRIAL LAUNCH CHECKLIST ══════════════════════════════════════════
  The seven-day free trial is BUILT and SWITCHED ON as of v45. The
  structured data was merged into the @graph in v46, so the old
  two-step comment flip no longer exists.

  STILL OPEN, and the reason this file is not uploaded:

    1. /unlimited-english still 302s to tv.eslpod.com, and every trial
       CTA on this page points at /unlimited-english#ep-pricing. Remove
       that RewriteRule and ship a sales page before this goes live.
    2. CLOSED in v56. LGM_BASE now reads https://www.eslpod.com/lgm/.
       Jeff moves the card-gated thank-you page into /lgm/ at the same
       time this file goes up. The interim /lgm/temp/ path and the
       console guard that watched for it are both gone.

  CTA ARCHITECTURE, set in v47. One ask per surface:
    HERO            free lesson, always, and it is the only hero button
    NAV PILL        trial while data-trial="on", free lesson while "off"
    MOBILE STICKY   mirrors the nav pill
    UE CARD         "See plans & pricing" button plus one browse link
  The hero trial button and the UE card trial link were DELETED, not
  hidden. Switching the trial off no longer brings them back, and
  switching it on no longer creates a second hero button.

  TO SWITCH THE TRIAL BACK OFF, two edits:
    a. On <div id="ep-page">, change data-trial="on" to "off"
    b. Delete the "How does the 7-day free trial work?" Question object
       from the FAQPage in the @graph. Leaving it advertises a trial the
       page no longer shows.

  Terms wired, taken from LandingPages_UE-Sales-Page_Trial_v15.html:
  card-gated, cancel before Day 7, charged Day 8 at $29/month or $290/year,
  reminder email 2 days before. The only unconfirmed value is the CTA
  destination, currently /unlimited-english#ep-pricing.
  ════════════════════════════════════════════════════════════════════
-->
<meta charset="utf-8"/>
<meta content="width=device-width, initial-scale=1" name="viewport"/>
<title>ESLPod · Speak Better English with Dr. Jeff McQuillan</title>
<meta content="Understandable English lessons for adults who want to actually speak. By Dr. Jeff McQuillan, Krashen co-author. 15M learners since 2005." name="description"/>
<meta content="ESLPod · Speak Better English with Dr. Jeff McQuillan" property="og:title"/>
<meta content="Understandable English lessons for adults who want to actually speak. By Dr. Jeff McQuillan, Krashen co-author. 15M learners since 2005." property="og:description"/>
<meta content="https://www.eslpod.com/images/eslpod-og.png" property="og:image"/>
<meta content="1200" property="og:image:width"/>
<meta content="630" property="og:image:height"/>
<meta content="https://www.eslpod.com/" property="og:url"/>
<meta content="website" property="og:type"/>
<meta content="ESLPod" property="og:site_name"/>
<meta content="en_US" property="og:locale"/>
<meta content="summary_large_image" name="twitter:card"/>
<meta content="ESLPod · Speak Better English with Dr. Jeff McQuillan" name="twitter:title"/>
<meta content="Understandable English lessons for adults who want to actually speak. By Dr. Jeff McQuillan, Krashen co-author. 15M learners since 2005." name="twitter:description"/>
<meta content="https://www.eslpod.com/images/eslpod-og.png" name="twitter:image"/>
<meta content="@eslpod" name="twitter:site"/>
<link href="https://www.eslpod.com/" rel="canonical"/>
<script>
/* v63: browser auto-detect, ported from the FAQ (render_multilang_pages.py),
   which was the only page family that had one, because it is the only family
   that script builds. English page only, first-time visitors only, and ?nolang
   escapes it so nobody is trapped. Crawlers ignore this and read the static
   HTML that already exists at every language's URL.

   In the head on purpose: running it after the picker would redirect after
   first paint and flash the English page.

   Reads BOTH storage keys on purpose: this family stores eslpod_ue_lang, the
   FAQ stores eslpod_lang, and without both a visitor who already chose a
   language on the FAQ would be redirected here anyway. */
/* ESLPod experiment framework v1 — assignment, stickiness, exposure.
 *
 * Ships inside the <head> of every Home page, immediately replacing the v63 browser-language
 * auto-redirect block. It is written to be pasted whole, and it is TESTED whole by
 * Analytics_Experiment-Framework_v1_tests_v1.mjs against a stubbed browser.
 *
 * ─────────────────────────────────────────────────────────────────────────────────────
 * WHAT IT DOES, AND THE ONE THING IT MUST NEVER DO
 *
 * It decides, for a first-time visitor whose browser language matches one of our sixteen
 * locales, whether to send them to the localised page (as today) or leave them on English —
 * and records which. That is EXP-009: does localisation actually lift lead-magnet signups?
 *
 * The thing it must never do is change who is eligible. Eligibility is computed by exactly
 * the same rules the v63 block used — English page only, first-time visitors only, ?nolang
 * escapes, same sixteen codes, same zh-Hant/zh-Hans handling. If eligibility drifted, the
 * two arms would be drawn from different populations and the result would mean nothing.
 *
 * ─────────────────────────────────────────────────────────────────────────────────────
 * WHY THE EXPOSURE EVENT FIRES ON THE LANDING PAGE, NOT AT THE MOMENT OF ASSIGNMENT
 *
 * The control arm redirects immediately. An `experiment_view` pushed to the dataLayer
 * microseconds before location.replace() would frequently never reach GA4 — the page is
 * torn down first. Worse, it would fail ASYMMETRICALLY: the treatment arm stays put and
 * always sends, the control arm races and sometimes does not, so the control arm would
 * appear to have fewer sessions than it really has and every rate computed against it would
 * be inflated. That is not a bug that shows up as an error; it shows up as a winning result.
 *
 * So assignment is recorded in a cookie, and BOTH arms fire `experiment_view` on the page
 * they land on — control on /xx/, treatment on /. Same code path, same timing, once per
 * session, deduped through sessionStorage.
 *
 * ─────────────────────────────────────────────────────────────────────────────────────
 * QUERY STRING SURVIVAL. The v63 redirect dropped the entire query string, destroying
 * utm_id, fbclid and gclid for every auto-redirected visitor — see
 * Analytics_Signup-Path-Audit_v1.md. This block carries them across, and appends
 * lang_auto=1 so the auto-redirected population is finally distinguishable from people who
 * searched in Spanish and landed on /es/ directly. Both fixes are prerequisites for the
 * experiment, not extras: without lang_auto we cannot even measure how big the eligible
 * pool is, and without the utm passthrough a tagged ad click loses its campaign id.
 */
(function () {
  var PAGE_LANG = 'en';                    // rewritten per file by the build
  var EXP = 'EXP-009';
  var COOKIE = 'esl_exp';
  var COOKIE_DAYS = 180;
  var SUP = ['en', 'ar', 'zh-Hans', 'zh-Hant', 'fr', 'de', 'id', 'it', 'ja', 'ko',
             'pl', 'pt', 'es', 'ru', 'th', 'tr', 'vi'];

  /* ---------------------------------------------------------------- tiny helpers */

  function readCookie(name) {
    try {
      var m = document.cookie.match(new RegExp('(?:^|;\\s*)' + name + '=([^;]*)'));
      return m ? decodeURIComponent(m[1]) : '';
    } catch (e) { return ''; }
  }

  function writeCookie(name, value, days) {
    try {
      var d = new Date();
      d.setTime(d.getTime() + days * 864e5);
      // Lax, not None: this cookie is only ever read first-party. Secure because the site
      // is https-only. Domain left unset so it stays on the exact host.
      document.cookie = name + '=' + encodeURIComponent(value) +
        ';expires=' + d.toUTCString() + ';path=/;SameSite=Lax;Secure';
      return true;
    } catch (e) { return false; }
  }

  /* The cookie holds every experiment the visitor is in, so a second live test does not
     evict the first:  "EXP-009:A|EXP-001:B"  */
  function getArm(exp) {
    var raw = readCookie(COOKIE);
    if (!raw) return '';
    var parts = raw.split('|');
    for (var i = 0; i < parts.length; i++) {
      var kv = parts[i].split(':');
      if (kv[0] === exp) return (kv[1] === 'A' || kv[1] === 'B') ? kv[1] : '';
    }
    return '';
  }

  function setArm(exp, arm) {
    var raw = readCookie(COOKIE);
    var parts = raw ? raw.split('|') : [];
    var out = [];
    for (var i = 0; i < parts.length; i++) {
      if (parts[i].split(':')[0] !== exp) out.push(parts[i]);
    }
    out.push(exp + ':' + arm);
    return writeCookie(COOKIE, out.join('|'), COOKIE_DAYS);
  }

  /* 50/50. Math.random is fine here: this is a coin flip per visitor, not a hash that has
     to be reproducible anywhere else. */
  function assign() { return Math.random() < 0.5 ? 'A' : 'B'; }

  /* Once per session, both arms, identical code path. sessionStorage rather than a page
     flag because the control arm's exposure happens on the NEXT page load. */
  function fireExposure(exp, arm) {
    var key = 'esl_exp_fired_' + exp;
    try { if (sessionStorage.getItem(key)) return false; } catch (e) { /* private mode */ }
    try { sessionStorage.setItem(key, '1'); } catch (e) { /* fall through and still fire */ }
    try {
      (window.dataLayer = window.dataLayer || []).push({
        event: 'experiment_view', experiment_id: exp, variant: arm
      });
      return true;
    } catch (e) { return false; }
  }

  /* Browser language -> one of our codes, or null. Lifted verbatim from v63 so eligibility
     cannot drift between the redirect and the experiment. */
  function detect() {
    var n = '';
    try {
      n = (navigator.language || (navigator.languages && navigator.languages[0]) || '').toLowerCase();
    } catch (e) { return null; }
    if (!n) return null;
    if (n.indexOf('zh-hant') === 0 || n.indexOf('zh-tw') === 0 ||
        n.indexOf('zh-hk') === 0 || n.indexOf('zh-mo') === 0) return 'zh-Hant';
    if (n.indexOf('zh') === 0) return 'zh-Hans';
    var t = n.split('-')[0];
    if (t === 'in') t = 'id';                       // legacy ISO code for Indonesian
    for (var k = 0; k < SUP.length; k++) {
      if (SUP[k].toLowerCase() === n || SUP[k].toLowerCase() === t) return SUP[k];
    }
    return null;
  }

  function storedLang() {
    var s = null;
    try { s = localStorage.getItem('eslpod_ue_lang') || localStorage.getItem('eslpod_lang'); } catch (e) {}
    if (!s) {
      var m = document.cookie.match(/(?:^|;\s*)eslpod_(?:ue_)?lang=([^;]*)/);
      if (m) { try { s = decodeURIComponent(m[1]); } catch (e) { s = m[1]; } }
    }
    return s;
  }

  /* ---------------------------------------------------------------- the two entry points */

  /* A LOCALISED PAGE. The redirect never ran here. If this visitor carries an assignment,
     this IS the control arm's landing page, so fire the exposure and stop. */
  if (PAGE_LANG !== 'en') {
    var armHere = getArm(EXP);
    if (armHere) fireExposure(EXP, armHere);
    return;
  }

  /* THE ENGLISH ROOT. */

  /* Already assigned and still here => treatment arm returning, or control arm who came
     back to English. Fire exposure for whichever arm they are in and do not re-randomise. */
  var existing = getArm(EXP);
  if (existing) {
    if (existing === 'B') fireExposure(EXP, existing);
    // An 'A' visitor sitting on English has navigated back deliberately; their exposure
    // already fired on the localised page. Re-firing here would count them twice.
    return;
  }

  /* Eligibility — identical to v63. */
  if (storedLang()) return;                                  // has chosen a language before
  if (/[?&]nolang(=|&|$)/.test(location.search)) return;     // explicit escape hatch
  var d = detect();
  if (!d || d === 'en') return;                              // English speakers are not eligible

  /* Eligible and unassigned: this is the randomisation point. */
  var arm = assign();
  setArm(EXP, arm);

  if (arm === 'B') {
    // TREATMENT: stay on English. Nothing to navigate; record the exposure here.
    fireExposure(EXP, arm);
    return;
  }

  /* CONTROL: the current behaviour, with the query string carried across and the
     auto-redirect marked. Exposure fires when the localised page loads. */
  try {
    var q = new URLSearchParams(location.search);
    q.set('lang_auto', '1');
    location.replace('https://www.eslpod.com/' + d + '/?' + q.toString() + location.hash);
  } catch (e) {
    // URLSearchParams exists everywhere this page supports, but a redirect must never be
    // the thing that fails. The fallback keeps the visitor moving and keeps the marker;
    // only the original parameters are lost, which is v68 behaviour.
    location.replace('https://www.eslpod.com/' + d + '/?lang_auto=1');
  }
})();
</script>
<link href="https://www.eslpod.com/" hreflang="en" rel="alternate"/><link href="https://www.eslpod.com/ar/" hreflang="ar" rel="alternate"/><link href="https://www.eslpod.com/zh-Hans/" hreflang="zh-Hans" rel="alternate"/><link href="https://www.eslpod.com/zh-Hant/" hreflang="zh-Hant" rel="alternate"/><link href="https://www.eslpod.com/fr/" hreflang="fr" rel="alternate"/><link href="https://www.eslpod.com/de/" hreflang="de" rel="alternate"/><link href="https://www.eslpod.com/id/" hreflang="id" rel="alternate"/><link href="https://www.eslpod.com/it/" hreflang="it" rel="alternate"/><link href="https://www.eslpod.com/ja/" hreflang="ja" rel="alternate"/><link href="https://www.eslpod.com/ko/" hreflang="ko" rel="alternate"/><link href="https://www.eslpod.com/pl/" hreflang="pl" rel="alternate"/><link href="https://www.eslpod.com/pt/" hreflang="pt" rel="alternate"/><link href="https://www.eslpod.com/es/" hreflang="es" rel="alternate"/><link href="https://www.eslpod.com/ru/" hreflang="ru" rel="alternate"/><link href="https://www.eslpod.com/th/" hreflang="th" rel="alternate"/><link href="https://www.eslpod.com/tr/" hreflang="tr" rel="alternate"/><link href="https://www.eslpod.com/vi/" hreflang="vi" rel="alternate"/><link href="https://www.eslpod.com/" hreflang="x-default" rel="alternate"/>
<!-- Favicon (v14: generated from /assets/ESLPodLogo.jpg; v17: fallback switched to .png) -->
<!-- v52: favicon standard. New ESLPod badge art, cache-buster bumped to
     20260806. Favicons cache past a normal reload, so the query string MUST
     change whenever /assets/favicon.png is overwritten or returning visitors
     keep the old icon. apple-touch-icon.png is the flush-square build (iOS
     applies its own corner rounding); favicon.png and icon-192.png keep the
     transparent rounded corners for browser tabs. -->
<link href="https://eslpod.com/assets/favicon.png?v=20260901" rel="icon" sizes="32x32" type="image/png"/>
<link href="https://eslpod.com/assets/icon-192.png?v=20260901" rel="icon" sizes="192x192" type="image/png"/>
<link href="https://eslpod.com/assets/favicon.png?v=20260901" rel="shortcut icon" type="image/png"/>
<link href="https://eslpod.com/assets/apple-touch-icon.png?v=20260901" rel="apple-touch-icon" sizes="180x180"/>
<!-- INDEXABLE — no noindex (Home is the SEO root) -->
<!-- Start Cookie Policy Code -->
<!-- <script type="text/javascript" src="/65c993535b2f09f71f31350df425666e.js"></script> -->
<!-- End Cookie Policy Code -->
<!-- ===== JSON-LD structured data (v9) ===== -->
<script type="application/ld+json">{
  "@context": "https://schema.org",
  "@graph": [
    {
      "@type": "Organization",
      "@id": "https://www.eslpod.com/#organization",
      "name": "Center for Educational Development",
      "alternateName": "ESLPod",
      "url": "https://www.eslpod.com/",
      "logo": "https://www.eslpod.com/assets/ESLPodLogo.png",
      "foundingDate": "2005",
      "founder": [
        {
          "@type": "Person",
          "@id": "https://www.eslpod.com/#jeff-mcquillan",
          "name": "Jeff McQuillan",
          "jobTitle": "Co-founder and Host, ESLPod",
          "alumniOf": {
            "@type": "CollegeOrUniversity",
            "name": "University of Southern California"
          }
        },
        {
          "@type": "Person",
          "@id": "https://www.eslpod.com/#lucy-tse",
          "name": "Lucy Tse",
          "jobTitle": "Co-founder and Writer, ESLPod",
          "alumniOf": {
            "@type": "CollegeOrUniversity",
            "name": "University of Southern California"
          }
        }
      ],
      "sameAs": [
        "https://podcasts.apple.com/us/podcast/speak-english-with-eslpod-com-3-new-lessons-a-week/id75908431",
        "https://www.youtube.com/user/eslpod",
        "https://www.facebook.com/eslpod/",
        "https://www.x.com/eslpod",
        "https://www.instagram.com/eslpod/"
      ],
      "inLanguage": "en"
    },
    {
      "@type": "Course",
      "name": "Unlimited English",
      "description": "Understandable English lessons for adults who want to actually speak. By Dr. Jeff McQuillan, Krashen co-author. 15M learners since 2005.",
      "provider": {
        "@id": "https://www.eslpod.com/#organization"
      },
      "instructor": [
        {
          "@id": "https://www.eslpod.com/#jeff-mcquillan"
        },
        {
          "@id": "https://www.eslpod.com/#lucy-tse"
        }
      ],
      "inLanguage": "en"
    },
    {
      "@type": "FAQPage",
      "mainEntity": [
        {
          "@type": "Question",
          "name": "What level of English do I need to start?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Lessons are designed for adult intermediate learners: people who can read a lot of English, understand most of it, but still find it hard to speak easily or understand native speakers. If you've tried apps like Duolingo or Babbel and feel you're not making progress, you're in the right place."
          }
        },
        {
          "@type": "Question",
          "name": "Are all of your lessons 100% in English?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Yes! All of our lessons are designed for intermediate and advanced English users. We do not use any translation. All lessons and instructions are completely in English to give you complete immersion in the language."
          }
        },
        {
          "@type": "Question",
          "name": "How does the 7-day free trial work?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "You get 7 days to try all of our 1,800+ lessons for free. Cancel anytime before Day 7. On Day 8, your credit card is charged either $29 (for the monthly plan) or $290 (for the annual plan), depending on which one you signed up for. We will email you 2 days before we charge you to remind you."
          }
        },
        {
          "@type": "Question",
          "name": "Does the membership auto-renew? Can I cancel anytime?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Yes. Monthly memberships renew every month, annual memberships renew every year, until you cancel. You can cancel from your account easily. Every course and membership has a 30-day money-back guarantee."
          }
        },
        {
          "@type": "Question",
          "name": "How is this different from Duolingo, Babbel, or online tutors?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Most apps and tutors focus on either rote memorization or \"output\" practice, forcing you to speak before you're ready. They're based on now-disproven 19th century theories of how we learn languages. Unlimited English is for adults who enjoy practical English in lessons that are funny and interesting. We use current scientific research to create real, interesting English at a level you can actually understand. The result is the ability to listen to fast English and to speak more fluently."
          }
        },
        {
          "@type": "Question",
          "name": "Is this only for American English?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "The hosts are American, so the accent is American English. But the vocabulary, the grammar, and the method work for any English you'll need, whether it is British, Australian, or international business English. The way clear English works is the same everywhere."
          }
        },
        {
          "@type": "Question",
          "name": "Can I download lessons for offline use?",
          "acceptedAnswer": {
            "@type": "Answer",
            "text": "Yes. Every lesson can be downloaded to our iOS or Android phone app so you can listen on a plane, on the subway, or anywhere your phone has no signal."
          }
        }
      ],
      "url": "https://www.eslpod.com/",
      "inLanguage": "en"
    }
  ]
}</script>
<!-- ===== End JSON-LD ===== -->
<!-- TRIAL FAQ SCHEMA. v46: the trial question now lives INSIDE the FAQPage in
     the @graph above, between "Are all of your lessons 100% in English?" and "Does the
     membership auto-renew?". There is no second FAQPage on this page any more.
     TO SWITCH THE TRIAL OFF: set data-trial="off" on #ep-page AND delete that
     one Question object from the @graph, or the structured data will advertise
     a trial the page no longer shows. -->
<!--
  ESLPod Home Page · v49
  URL: eslpod.com/
  Per LandingPages_DesignSpec_v3_01_Home.md + v3_00_Foundation.md.
  Design system tokens lifted from Magnet TY v1 (Foundation §3).
  Lead-magnet popup (modal markup, CSS, runtime) embedded verbatim from index-1.html.

  ── CHANGELOG ─────────────────────────────────────────────────────
  v49 (2026-08-05) - applied to file v48.html. Two one-line corrections,
  no copy moved, no markup restructured, nothing else touched.

  1. FOOTER FREE-LESSON LINK. It shipped as href="#", so with the script
     blocked or the link opened in a new tab it resolved to whatever URL
     the visitor was already on. On /podcast/ that reads as a link back
     to /podcast/, which is how Jeff found it. It now points at
     https://www.eslpod.com/free, which redirects to the live lead-magnet
     offer. .popup-trigger is unchanged, so a normal click still opens
     the modal and the href is only ever the no-JavaScript fallback.
     Do not put the bare "#" back.

  2. HERO RATING. 4.8 became 4.6. The live US Apple Podcasts page read
     4.6 from 2,437 ratings on 2026-08-05, and the Apple subscription
     page at /podcast/ states 4.6. The two pages agree now.

  v48 (build session, 2026-08-04) - applied to file v47.html
  CTA COLOUR SYSTEM. Jeff's calls, after advice. No copy or markup moved.
  THE PROBLEM, measured before changing anything
  - Ten CTA elements on the page. Nine were orange and one was green, so the
    only green button was "See plans & pricing". There was no rule, just one
    right instinct applied once.
  THE RULE
  - Orange means the free lesson. Green means anything that leads to paying.
    Colour now tracks the offer rather than the position, which is what makes it
    survive the trial toggle: the free-lesson twins in the nav and sticky bar
    stay orange on their own when data-trial="off".
  WHAT MOVED
  - The trial pill and the mobile sticky trial went from orange to green.
  - "See plans & pricing" went from var(--ep-green) to the same green, so the two
    paid buttons match. Jeff's call, and it also fixes a contrast failure that
    was already live.
  - Nothing else changed colour. Every free-lesson button was already orange.
  CONTRAST, measured not eyeballed
  - var(--ep-green) #159b97 with white text is 2.39:1 and fails WCAG AA at every
    size. It was on the pricing button. The new --ep-green-cta #159b97 is
    5.43:1 and passes.
  - FLAGGED, NOT FIXED: var(--ep-orange) #f05a22 with white text is 3.39:1. That
    clears AA for large text, which WCAG defines as 18.66px bold or larger, and
    these labels run 14.7px to 16.8px bold. So the orange is under the bar at the
    sizes actually used, on the free-lesson buttons and everywhere else orange
    appears sitewide. Out of scope for this build and worth a decision later.
  - The .ep-textlink green underline is still var(--ep-green). It is a 2px rule
    on a dark-text link, not a text-on-fill pair, so the ratio above does not
    apply to it. Left alone deliberately.
  SCOPE
  - Home only, on Jeff's instruction. The rule travels with the logo rollout to
    the other pages afterwards, and it is not yet written into the brand guide.
  - Also tidied: the UE card's two links lost their line break in v47 when the
    trial link was deleted. Markup only, no rendered change.

  v47 (build session, 2026-08-04) - applied to file v46.html
  CTA REDUCTION. Jeff's calls, after advice. No copy rewritten.
  THE PROBLEM, measured before changing anything
  - With the trial on, the page carried nine CTAs, five trial and four free
    lesson. Five of them resolved to the same URL, /unlimited-english#ep-pricing:
    the nav pill, the hero trial button, the mobile sticky, "See plans &
    pricing", and the UE card trial link. The UE card was offering one
    destination three times, not three choices.
  HERO
  - The trial button and its terms line are DELETED. The free lesson button is
    now the only button in the hero.
  - Reasoning Jeff accepted: the home page rubric scores the hero button on
    whether it matches the dominant visitor goal, and this page takes cold
    podcast and organic traffic. The CRO brief's "lead with the trial" is
    Decision 1, written for the sales landing page, and having both pages lead
    with the same button stops the home page routing. Third, the button said
    "Start My Free 7-Day Trial" and delivered a pricing section on another page,
    which is the link-promise fault v20 already fixed once elsewhere.
  - hero_freecourse_micro lost its .ep-notrial-only class and is now
    unconditional, since there is no trial button above it to disambiguate.
    Its trial twin, hero_freecourse_micro_trial, is deleted.
  UE CARD
  - The "Start My Free 7-Day Trial" text link is DELETED. The card keeps "See
    plans & pricing" as its button and "See everything inside Unlimited English"
    as its one text link. Those two are the only pair on the card with different
    destinations, the anchor and the page top.
  NAV AND MOBILE STICKY
  - Unchanged, and that is the decision, not an oversight. Both already swap to
    the trial while data-trial="on", so the trial keeps two persistent surfaces
    and the sticky bar mirrors the nav, which is what Jeff chose.
  CONSEQUENCES
  - Trial CTAs on Home drop from five to two, both in persistent chrome. If
    trial starts look thin, the lever is the sales page hero, not this one.
  - Orphaned markup-only i18n keys, left in no element: hero_trial_cta,
    hero_trial_micro, hero_freecourse_micro_trial, ue_cta_trial. They are not in
    STRINGS, so nothing breaks. Any localised Home built from this file should
    drop them from its key list.
  - .ep-hero-trial-cta and .ep-hero-trial-micro CSS rules are now unused. Left in
    place deliberately, so restoring a hero trial button is markup only.

  v46 (build session, 2026-08-04) - applied to file v45.html
  Three of Jeff's calls, none of them copy.
  - STRUCTURED DATA MERGED. The standalone FAQPage block is gone. Its one
    question now sits inside the FAQPage in the @graph, between "Are all of your
    lessons 100% in English?" and "Does the membership auto-renew?", which is
    the third slot, matching where it sits in the visible FAQ. One FAQPage per page, which is what Google asks
    for. The cost is that switching the trial off is now two edits, not one, and
    the note in <head> spells both out.
  - GUARD REWRITTEN to match. It used to test for the standalone
    script[data-trial-schema] element, which v46 deletes, so it would have
    false-positived on every load. It now checks the @graph text for the trial
    question itself. Second guard false-positive found in two versions, both
    caused by the guard testing for the mechanism instead of the outcome.
  - STEP 3 REVERTED. LGM_BASE is back to https://www.eslpod.com/lgm/temp/ on
    Jeff's confirmation that the card-gated thank-you page has NOT replaced the
    old page at /lgm/ yet. The console guard now warns, correctly, and will keep
    warning until that page is swapped and the line is changed.
  - HEAD CHECKLIST REWRITTEN. It described a launch procedure that no longer
    exists. It now lists what is actually open, and the two edits to switch the
    trial back off.
  NOT UPLOADED, on Jeff's instruction
  - Home is held until the sales page ships. /unlimited-english still 302s to
    tv.eslpod.com, so every trial CTA here would send visitors off the site.
    v46 sits on disk ready for the day that redirect comes out.

  v45 (build session, 2026-08-04) - applied to file v44.html
  THE TRIAL IS ON. All three checklist steps done, plus one casing fix.
  - Step 1. #ep-page data-trial="off" became "on". That reveals all eight
    .ep-trial-only elements and hides the four .ep-notrial-only ones.
  - Step 2. The TRIAL FAQ SCHEMA block is uncommented and live. The two markers
    are kept as real comments so it can be switched back off the same way.
  - Step 3. LGM_BASE went from https://www.eslpod.com/lgm/temp/ to
    https://www.eslpod.com/lgm/. THIS ASSUMES the card-gated thank-you page has
    replaced the old "How to Start a Conversation" page at /lgm/. If it has not,
    this line is wrong and the popup sends people to the old page.
  - GUARD BUG FOUND AND FIXED, and it is the exact trap the guard exists to
    prevent. The guard string-matches the popup script element for the staging
    path. A v35 code comment two lines above LGM_BASE quoted that path
    literally, so the warning kept firing after step 3 was correctly done. The
    comment is rewritten without the literal. Verified: warning gone, and the
    schema check still fires when the block is re-commented.
  - Casing. The hero terms line said "Cancel anytime before day 7" while the
    FAQ, the JSON-LD, the sales page hero and timeline, and Trial_v24 all say
    "Day 7". Jeff's call: capital D. One instance changed, in hero_trial_micro.
    The lowercase instances that remain are inside changelog comments, quoting
    what Trial_v15 said, and are not visible copy.
  BLOCKING, NOT FIXED HERE
  - Every trial CTA points at /unlimited-english#ep-pricing. Per the 2026-08-03
    handoff, that URL still 302s to tv.eslpod.com and neither sales variant is
    live. Until that redirect is removed and a sales page ships, this page sends
    trial clicks off the site.
  - Two FAQPage entities now exist on this page: the one inside the @graph and
    the standalone trial block. Raised with Jeff rather than resolved here.
  - The eight trial i18n keys are markup-only and English. Fine for this page,
    which is English, and still outstanding for any localised Home.

  v44 (build session, 2026-08-04) - applied to file v43.html
  One sentence in founder_p1. Nothing else in the file changed.
  - "Today, it is among the most-respected and popular English learning programs
    in the world." became "Today, it is one of the most respected and popular
    English learning programs anywhere."
  - Why: "in the world" closed both of the first two sentences. Jeff picked this
    from fifteen rewrites. The hyphen in "most-respected" went too, since a
    superlative adverb before an adjective does not take one.
  - The first sentence is untouched, so the 2005 first-in-the-world claim still
    stands exactly as Jeff wrote it.

  v43 (build session, 2026-08-04) - applied to file v42.html
  Two changes on Jeff's instruction: the Your Professors copy and the new logo.

  COPY — Your Professors
  - founder_p1 through founder_p4 replaced with Jeff's supplied text. The
    i18n keys, the eyebrow, the H2 and the cred3 line are untouched. These four
    keys are markup-only and are not in STRINGS, so nothing on this page needs
    retranslating.
  - Kept from v42, because Jeff's pasted copy carries no markup: the ESLPod
    registered mark as a <sup> in p1, <em> on Why Don't They Learn English? and
    on Harvard Educational Review, and the link to eslpod.com/research/.
  - The research link moved anchor. v42 hung it on "researching" in a sentence
    that no longer exists, so it now sits on "research" in the new p4. Same URL,
    same inline style.
  - Wording changes inside Jeff's copy, noted so they are not read as drift:
    "first English teaching podcast" became "first English learning podcast",
    the USC parenthetical was dropped from p2, Dr. Tse's "doctorate" became
    "Ph.D.", and p2 gained the CNN line.
  UNVERIFIED CLAIMS, flagged not blocked
  - "among the most-respected and popular English learning programs in the
    world" is a new superlative and is not sourced. Jeff's claim, Jeff's call.
  - "has appeared on CNN as a leading language expert" is new to this page.
    Supplied by Jeff, not verified against a broadcast record here.
  - The 15 million learners, one billion downloads, 60 books and 40 years
    figures all carry over from v42 unchanged.
  - "in the world" now closes two consecutive sentences in p1. Left verbatim
    because it is Jeff's copy, and raised with him rather than silently edited.

  LOGO
  - New asset. /assets/ESLPodLogo.png is replaced by the green badge, trimmed
    from the supplied 940x788 file to 867x306. Filename kept on Jeff's call, so
    the src does not change and every page picks the new mark up on upload.
  - .ep-logo img height went from 72px to 84px. Padding unchanged at
    13px 0 11px 2.5px, so the badge renders 60px and the header bar goes from
    96px to 109px. Chosen by Jeff from a six-way render, delivered as
    09_Briefs-and-Research/LandingPages_Logo-Comparison_v1.png.
  - Shadow trimmed off. It is invisible on the charcoal bar and only added
    dead space.
  - SCOPE. Home only. The site-wide swap is queued behind Jeff seeing this live,
    and it now travels with the 48px rollout rather than being pure CSS, because
    every other page inherits the new asset the moment it is uploaded.

  v42 (build session, 2026-08-03) - applied to file v41.html
  One correction, three places. No copy, layout, logo or tracking changes.
  - POPUP-BLOCKER FALLBACK moved from https://www.eslpod.com/lgm/signup.html to
    https://www.eslpod.com/signup.html, which is the root page.
  - How the error happened, so it is not repeated: v41 took /lgm/signup.html
    from Jeff's answer to a question I asked, and neither of us checked it.
    His .htaccess, seen later the same day, carries
    RewriteRule ^signup/?$ /signup.html [L] at the site root and nothing under
    /lgm/. Jeff then confirmed the root page exists and there is no signup
    folder. So v41 pointed the fallback at a page that does not exist, and
    because this only fires when a popup blocker suppresses the modal, it would
    have failed silently for exactly the visitors it exists to rescue.
  - Changed in all three places, the same three v40 unified: the hoisted
    FALLBACK_URL constant, the inline default in the blocker script, and the
    applyLang selector that appends ?lang= to a matching anchor.
  - The ?lang= append still matters. Jeff confirms signup.html is localised,
    though it has no per-language break-out pages the way the FAQ does, so one
    URL carrying a language parameter is the right shape for it.

  v41 (build session, 2026-08-03) - applied to file v40.html
  Jeff's three corrections to v40. No logo change, no structural change.
  - Curriculum pill: "What you get" became "What You Get", title case, per Jeff.
  - curr_h2: v40 had it as "What you get with Unlimited English". Jeff's wording
    is "What's Inside Unlimited English", so the H2 keeps the "inside" framing
    and the pill carries "What You Get" above it. The v34 mismatch is closed on
    Jeff's wording, not on mine.
  - Straight apostrophe in "What's", matching the 60-odd contractions already in
    this page's reader copy. The four &rsquo; entities in the file are all in
    non-reader strings and were left alone.
  - The internal label, data-screen-label="04 What is inside", and the CSS
    section comment still say "What is inside". Neither renders.
  - POPUP-BLOCKER FALLBACK destination resolved by Jeff:
    https://www.eslpod.com/lgm/signup.html, replacing the unverified
    https://eslpod.com/signup.html. Changed in all three places that carried it,
    which is the hoisted constant, the inline default in the blocker script, and
    the applyLang selector that appends ?lang= to any matching anchor.
  - That applyLang selector is still dead code, because no anchor in the markup
    points at the fallback. It was updated rather than left pointing at the old
    URL, so it does the right thing if such a link is ever added.
  - The VERIFY comment is gone from the constant, since the destination is now
    Jeff's stated answer rather than a guess.

  v40 (build session, 2026-08-03) - applied to file v39.html
  Four flags from the v39 changelog, closed on Jeff's instruction. No logo
  change: the padded 48px pattern at line ~1003 is untouched.
  - curr_h2: "What is inside Unlimited English" became "What you get with
    Unlimited English", so the H2 now agrees with the curriculum pill two lines
    above it, which v34 changed to "What you get" and left mismatched. The v34
    flag is closed. curr_h2 is a markup-only key and is not in STRINGS, so the
    page translation pass owns the 16 non-English versions of this line.
  - The internal label on the section, data-screen-label="04 What is inside",
    and the CSS section comment above #ep-curriculum both still say "What is
    inside". Neither one renders, and renaming them would churn the design-tool
    screen map for no reader benefit.
  - STRINGS.cta_trigger: 11 languages said "course" where the reviewer-approved
    STRINGS.cta key for the same offer says "lesson". Fixed by reusing the noun
    already approved in cta: es lección, pt lição, fr leçon, de Lektion, it
    lezione, ja レッスン, ko 레슨, ru урок, tr ders, vi bài học, ar درس.
  - zh-hans and zh-hant were NOT changed. Both keys already say 课程 / 課程, and
    the reviewer-approved cta uses that same word, so they were never part of
    the defect. The v35 note counted 11 languages and it is 11, not 13.
  - id, pl and th were already correct in cta_trigger and are untouched.
  - STRINGS.fallback_link in id, pl and th: reviewed and kept as written. "Ada
    masalah? Klik di sini." and "Masz problem? Kliknij tutaj." are both
    idiomatic and match the English sense. The Thai line carries no question
    mark, which is correct Thai punctuation and not an omission.
  - THE LARGER FINDING ON THAT KEY. STRINGS.fallback_link is dead data on this
    page. Its consumer is $$('.fallback-link'), and no element in the markup
    carries that class, because v20 deleted the hero "Having trouble? Click
    here." link on NN/g broken-promise grounds. All 17 values are translated and
    none of them render. Nothing was removed here, since the strings cost
    nothing and the link may come back, but the v35 flag is closed as moot
    rather than as fixed. A native Thai pass on this key buys nothing until an
    element actually uses it.
  - A REAL DEFECT FOUND NEXT DOOR, and fixed. v35 recorded that the
    popup-blocker fallback URL had been "hoisted into a named constant so it is
    a one-line change". It had not. FALLBACK_URL in the popup runtime was
    declared and never read by anything, while the blocker-fallback script near
    </body> carried its own hard-coded second copy of the same URL in a separate
    IIFE. Changing the hoisted constant would have changed nothing, and the two
    copies could drift apart without any warning.
  - Fixed by publishing the constant as window.ESLPOD_FALLBACK_URL and having
    the blocker script read it, with the literal kept only as an inline default
    if the runtime never ran. The destination is now genuinely one line.
  - The destination itself is STILL UNVERIFIED and still marked VERIFY:
    https://eslpod.com/signup.html. It was not confirmed against the live server,
    because WebFetch on eslpod.com returns ROBOTS_DISALLOWED and working around
    that is off limits. Deployment playbook decision 3 removed the fallback LINK
    on the thank-you page, which is a different thing from this blocker
    fallback, so it does not settle this. Jeff needs to name the destination.
  - Nav CTA and mobile sticky CTA now swap with the trial state, which closes
    the v32 "STILL OPEN" item. The free-lesson control in each of the three
    places gained .ep-notrial-only, and a trial twin carrying .ep-trial-only and
    data-track="cta_click_trial" sits beside it. Nothing new is needed at launch
    beyond step 1 of the checklist, because the existing toggle rules already
    drive both classes.
  - Desktop nav and mobile drawer read "Start free trial", which is short enough
    for the pill. The sticky bar reads "Start My Free 7-Day Trial", matching the
    hero button word for word.
  - Both trial twins are <a> elements, not popup triggers, and both point at
    /unlimited-english#ep-pricing, the same destination the hero trial button
    and the UE card trial link already use.
  - CRO consequence, stated rather than hidden: with the trial on, the nav and
    the sticky bar stop offering the free lesson. That path stays reachable from
    the hero button, the #ep-magnet section and the footer link, all three of
    which are unconditional.
  - New markup-only i18n keys: nav_trial_cta, mobile_sticky_trial.
  - Changelog backfilled below for v37, v38 and v39, which had no entries.

  v39 (2026-07-31) - applied to file v38.html
  LOGO ONLY. One declaration changed, plus its comment. Nothing else in the file
  differs from v38.
  - .ep-logo img padding went from 21px 0 19px 2.5px to 13px 0 11px 2.5px, so
    the wordmark went from 32px to 48px inside the same 72px box.
  - Jeff's call after seeing the swap live. v38 reproduced the old size exactly
    and read too small once the asset was clean, so 48px is a deliberate size
    increase and not a restoration.
  - Rendered and measured at 48px: logo box 127.44 x 72, header bar unchanged at
    97 / 91 / 96 desktop and 97 / 95 / 96 at 390px.

  v38 (2026-07-31) - applied to file v36.html
  LOGO ONLY, and the branch that v39 and v40 descend from. v37 is not an
  ancestor of this file.
  - Established the padded pattern. .ep-logo img kept height: 72px and gained
    padding: 21px 0 19px 2.5px, which reproduced the old asset's transparent
    margins as CSS at the same 72/2463 scale: 718 to 21px top, 651 to 19px
    bottom, 88 to 2.5px left.
  - The rule that outlived this version: size the wordmark with the padding and
    never with the height, because the two padding numbers must always sum with
    the content to 72 or the header bar moves.
  - Wordmark measured 81.25 x 29.75 before and 80.75 x 29.50 after in Chromium
    at 4x, with the header bar unchanged at 1440 / 1024 / 768 / 390.
  - Recorded the trap that still stands: never upload ESLPodLogo_v2-forUpload.png,
    which is padded differently and needs different numbers.

  v37 (2026-07-30) - applied to file v36.html - DEAD BRANCH, do not build on it
  - Set .ep-logo img to a bare height: 32px as part of a 37-file site-wide
    standardization, before anyone had seen the trimmed asset in place.
  - Superseded the next day. The bare-height approach was replaced by the padded
    pattern in v38, and the 32px size was rejected by Jeff in v39.
  - Left on disk deliberately, because the 34 files still carrying that 32px
    height are the open logo-audit item.

  v36 (build session, 2026-07-30) - applied to file v35.html
  JEFF'S BUG REPORT. The popup submitted with an empty name and email.
  NOT a v35 regression. Present since v26, and present in the canonical
  Analytics_Popup_Production_v10.html too, which has no validation at all.
  Only the home page is fixed here, per Jeff.
  - The <form> carried novalidate, which switches off the browser's own
    required-field checking. Removed. The three required fields now behave
    natively, and the browser renders its message in the visitor's own
    language, which suits a 17-language popup.
  - The submit handler tested the language dropdown and nothing else, so once
    a language was picked, any click submitted. It now calls form.checkValidity()
    first and returns on failure. The language test is kept underneath it as a
    belt-and-braces guard, because the <select> is populated by script and an
    empty value is a different failure from an invalid one.
  - Name and email are trimmed before validation. A field holding only spaces
    satisfies `required` and would otherwise have passed.
  - Order matters and was chosen deliberately: validation runs BEFORE
    btn.disabled=true. In v35 the button was disabled first, so had validation
    existed there it would have locked the visitor out of their own correction.
  - Verified in a headless render, five cases. Language only, name with no
    email, whitespace in both fields, and the malformed address "notanemail"
    are all blocked with no navigation and no Worker POST. A fully valid
    submission still posts to the v6 Worker and redirects to /lgm/temp/?lang=.
  - No copy, layout, translation, tracking or redirect changes.

  v35 (build session, 2026-07-30) - applied to file v34.html
  JEFF'S QUEUE - popup wiring only. No copy, layout or section changes.
  - Bovard photo: /assets/Bovard_USC.jpg became .png, per Jeff. NOT verified
    against the live server (the fetch tool will not read image bytes). Confirm
    the .png is uploaded before this page goes live, or the USC photo 404s.
  - THE BUG THAT MATTERED. WORKER_BASE pointed at the LEGACY v5 Worker,
    eslpod-analytics.titleman.workers.dev, with the v5 secret. That Worker does
    not carry AC_FIXED_TAGS, so no home-page signup ever received lm-5-phrases,
    UEM-SL-NonApp, 5Phrases or Popup, and the Active Campaign automation that
    triggers on "Tag lm-5-phrases is added" could never fire from this page.
    Now posts to eslpod-analytics-v6.titleman.workers.dev with the v6 secret,
    per LandingPages_PopupV6-Deployment_v5.md section 3. Tags are applied by
    the Worker, not by this page, so nothing else here needed to change.
  - Redirect rebuilt. THANKYOU_BASE was https://eslpod.com/trial and the page
    built a path at /trial/<lang>/trial-<lang>-a.html. Those per-language trial
    pages are retired. Renamed to LGM_BASE and set to
    https://www.eslpod.com/lgm/temp/, Jeff's interim destination, which serves
    LandingPages_LGM-ThankYou-Page_v2.html. The old "How to Start a
    Conversation" page with the $1 trial is still live at /lgm/ itself, which
    is why the interim path exists.
  - buildRedirectURL now sends lang as the first query param. It never sent it
    at all before, and ?lang= is the one param the thank-you page reads. The
    utm set, fbclid, gclid and ga_client_id still follow.
  - Meta Pixel Lead fired content_name "Lead Magnet - How to Start a
    Conversation in English", the wrong magnet, with an em dash in it. Now
    "5 Phrases That Make You Easier to Understand in English".
  - STRINGS.title: 13 of the 17 languages still carried the old "How to Start a
    Conversation" headline while English read 5 Phrases. Replaced from
    Analytics_Popup_Production_v10.html, the reviewer-approved translation pass.
    English, id, pl and th were already correct and are untouched.
  - STRINGS.cta: same problem, wider than queued. 11 languages said "course"
    where English says "lesson". All replaced from popup v10. Queued as Spanish
    and Portuguese only; the other nine had the identical defect and fixing two
    of eleven would have shipped a worse inconsistency than either state.
  - Hidden-field hydrate list had 'gclid' twice. Deduped. Cosmetic.
  - Popup-blocker fallback URL hoisted into a named constant. Destination
    unchanged at eslpod.com/signup.html and still marked VERIFY, because no
    confirmed replacement exists. Deployment playbook decision 3 removed the
    fallback LINK on the thank-you page, which is a different thing from this
    blocker fallback, so it was not treated as governing here.
  - TRIAL LAUNCH CHECKLIST is now three steps, not two. Step 3 is the LGM_BASE
    swap from /lgm/temp/ to /lgm/. The bottom-of-body guard script warns in the
    console if the trial is switched on while the redirect still says /lgm/temp/.
  NOT DONE, deliberately
  - STRINGS.cta_trigger still says "course" in the same 11 languages, and
    fallback_link is unreviewed in id, pl and th. Those are page copy, not popup
    wiring, and the translation pass is on hold pending Jeff's review.
  - The 5 Phrases thank-you page itself. Jeff's call: interim page now, the
    July-16 card-gated page after the trial launches.

  v34 (build session, 2026-07-28) - applied to file v33.html
  - Curriculum pill: "What is inside" became "What you get". FLAG: the H2 two
    lines below it still reads "What is inside Unlimited English", because Jeff
    named the pill only.
  - Your Professors section rewritten to Jeff's copy. The eyebrow is now plural,
    "Your professors". The old two paragraphs became four: ESLPod's history,
    then Dr. McQuillan, then Dr. Tse, then the shared method.
  - H2 changed from "Dr. Jeff McQuillan" to "Dr. Jeff McQuillan & Dr. Lucy Tse".
    JEFF DID NOT SPECIFY THIS. The section now introduces two people and the
    eyebrow is plural, so a single name in the heading contradicted both. Change
    it if you wanted something else.
  - cred1, "Author of more than 60 books and scientific articles", REMOVED. The
    same claim is now in the Dr. McQuillan paragraph, three lines above, and
    saying it twice reads careless. Restore it in one word if you disagree.
    cred3 kept.
  - The "researching" link to eslpod.com/research/ was kept on the method
    paragraph. Jeff's pasted copy had no link this time, but the link was
    already live and he did not ask to remove it.
  - founder_p3 and founder_p4 are new markup-only i18n keys.
  - Portrait fallback text changed from "Dr. Jeff McQuillan" to "University of
    Southern California", because the image is the Bovard building now, not a
    portrait. The fallback only shows if the image fails to load.
  UNVERIFIED CLAIMS IN THIS COPY, flagged not blocked
  - "the first English teaching podcast in the world" is a superlative, and it
    is the kind a competitor challenges. Jeff's claim, Jeff's call.
  - Dr. Tse's book being best-selling, the dozens of research articles, and the
    Harvard Educational Review contribution are all unverified here.

  v33 (build session, 2026-07-28) - applied to file v32.html
  Makes the trial FAQ a two-line flip instead of a retyping job.
  - TRIAL FAQ SCHEMA block added right after the live FAQPage JSON-LD, fully
    written and wrapped in an HTML comment. Publishing it means deleting the two
    comment lines. The main FAQPage stays valid the whole time it is switched
    off, which is why the trial question is not commented out INSIDE the main
    block: a comment inside JSON-LD makes the whole block unparseable.
    Tradeoff accepted: this ships as a second FAQPage block once live. Google
    merges multiple JSON-LD blocks. The alternative was maintaining two full
    copies of the FAQ schema, which drift the first time any other answer is
    edited.
  - TRIAL LAUNCH CHECKLIST comment added at the top of <head>, listing the two
    edits in order.
  - Guard script added before </body>. If data-trial is "on" and the trial
    schema is still commented out, it prints one console warning. It renders
    nothing, sends nothing, and fires nothing in production.
  - The two toggle CSS rules were rewritten from [data-trial="off"] and
    [data-trial="on"] to #ep-page:not([data-trial="on"]) and
    #ep-page[data-trial="on"]. Now the literal string data-trial="off" appears
    exactly once in live markup, on #ep-page, so a find-and-replace cannot
    silently break a CSS selector instead of flipping the page.
  - Verified all three states in a render. Off: one JSON-LD block, trial hidden,
    no warning. On with the schema still commented: trial visible, warning
    fires. On with the schema uncommented: two JSON-LD blocks, both parse, no
    warning.
  LOGO
  - No CSS change in this file. The logo is fixed in the asset instead. The
    original /assets/ESLPodLogo.png was 2938x2463 with the mark occupying only
    the middle 2849x1094, so 44 percent of the canvas height was transparent
    padding and a 72px CSS box rendered a 32px mark. The replacement file is
    cropped and then given back a deliberate 8.3 percent of even vertical
    padding, so the existing 72px box renders a 60px mark. Every page that
    references that filename gets the fix with no CSS edit anywhere.

  v32 (build session, 2026-07-28) - applied to file v31.html
  JEFF'S QUEUE
  - Hero H1: the bullet is gone. The two halves are now block-level spans, so
    the break between them is structural instead of depending on where the text
    runs out. Still 2 lines from 600px to 1920px, 4 lines on a phone. With the
    bullet removed the halves are shorter, so #ep-hero .ep-h1 went back to the
    global 4.6vw curve and the 861-1140px band is full size again.
  - founder_p2 research link: /research.html became /research/.
  - Footer About: "The Method" added at https://www.eslpod.com/method, directly
    above The Science.
  - Top nav: "The Method" added between Pricing and The Science, same URL. Also
    added to the mobile drawer, which mirrors the desktop nav.
  - Press logos stay white. The six files in /assets/ are *_white.png, which is
    white artwork, not colour art with a filter over it. Colour needs colour
    source files uploaded. See the note to Jeff.
  ANALYTICS FIXES, against Analytics_Instrumentation-Spec_v1.md
  - Meta Pixel <noscript> tracking image was missing. The head carried an empty
    <noscript></noscript> where it belonged. Added.
  - GTM <noscript> iframe was roughly 54,000 characters into the body instead of
    directly after <body>. Moved to the top of <body>.
  - gclid was never captured or forwarded, while the five utm_* params and
    fbclid were. Added a hidden gclid field and put it in all three parameter
    lists, so Google Ads clicks now reach the Worker and the /lgm/ redirect.
  - The click listener matched only [data-track="cta_click_trial"], so the two
    other data-track attributes on the page, cta_click_freecourse_nav and
    cta_click_pricing, were decorative and fired nothing. The listener now
    matches any [data-track] and pushes that value as the GA4 event name.
    Verified in a render: clicking the pricing link and the nav CTA now push
    cta_click_pricing and cta_click_freecourse_nav.
  SEO / PERFORMANCE
  - loading="lazy" and decoding="async" added to the nine below-the-fold images
    (USC photo, both UE mockups, six press logos). The nav logo and the hero
    portrait stay eager because they are above the fold.
  - Not changed and not needed for an English-only launch: no hreflang cluster
    yet, since there are no translated URLs to point at. That belongs to the
    translation pass and to eslpod-seo-structure.
  - No robots meta, which is correct. Absent means index, follow.
  STILL OPEN, deliberately not built
  - The nav CTA and the mobile sticky CTA still say "Get a free lesson" in both
    trial states. Whether they should swap to the trial when data-trial="on" is
    a CRO decision for Jeff, not a mechanical fix.
  - The trial FAQ is still absent from the FAQPage JSON-LD, by design.
  - New markup-only i18n keys this version: nav_themethod, foot_themethod.

  v31 (build session, 2026-07-28) - applied to file v30.html
  Hero H1 was rendering on three lines in Chrome. Now two, from 600px up to
  1920px. Three changes, all in the hero:
  - .ep-hero-grid columns went from "1.1fr 1fr" to "minmax(0, 1fr) 220px". The
    old split was sized for the pre-v28 480px portrait. The portrait is 240px
    now, so that track was holding roughly 100px of empty space the headline
    needed. The portrait still centres inside its track.
  - .ep-hero-grid column-gap went from clamp(32px, 5vw, 64px) to
    clamp(24px, 3vw, 40px), which returns another 24px at desktop width.
  - The H1 is now two inline-block spans (.ep-h1-part), so each half wraps as a
    unit and the break always lands on the bullet rather than wherever the text
    runs out. hero_h1 is a markup-only key and is not in STRINGS, so the popup
    loader cannot overwrite the spans.
  - #ep-hero .ep-h1 scales on 4.2vw instead of the global 4.6vw, so the first
    half never outruns its column between 861px and about 1140px. Full 3rem
    returns above ~1143px, so 1280px and wider are unchanged at 48px.
  - Measured: 2 lines at 600, 768, 860, 900, 1000, 1024, 1100, 1200, 1280, 1366,
    1440, 1600 and 1920. Phones at 430px and below run 4 lines, which is the
    headline simply being longer than a phone is wide.
  - The ESLPod logo size complaint is NOT in this version. Jeff is having a
    bigger icon made and asked to hold.

  v30 (build session, 2026-07-28) - applied to file v29.html
  - curr_foot: "Every lesson: audio + 10-page Learning Guide. Listen anywhere.
    Cancel anytime." became "Every lesson: audio + 10-page Learning Guide -
    Listen anywhere - Cancel anytime" with bullet separators and no closing
    period. The two periods were splitting the line into fragments, which the
    house rule forbids.
  - faq_trial_a: the cancel deadline now reads "Cancel anytime before Day 7"
    instead of "before Day 8", approved by Jeff. The charge date stays Day 8,
    so the page tells people to cancel a full day before any money moves. This
    matches the hero trial microcopy, which already said day 7, and it removes
    the day 7 / Day 8 mismatch this page inherited from Trial_v15. Jeff is
    fixing Trial_v15 to match. Not legal advice.
  - The duplicated 1,800+ lessons line stays in both places (pillar2_mech_body
    and ue_b1), per Jeff. Closed, not a defect.

  v29 (build session, 2026-07-28) - applied to file v28.html
  Trial terms filled in from Jeff's own material rather than left as
  placeholders. Sources: LandingPages_UE-Sales-Page_Trial_v15.html in
  01_Sales-Pages, uemsales11.html, and UE-Pipeline_Brief_v3.md. All three state
  a 7-day CARD-GATED trial. Trial_v15 also confirms the Teachable checkout
  product IDs (6776709 monthly, 6776708 annual, 6776707 skip-the-trial) and that
  every trial CTA on that page points at its own #ep-pricing block.
  - Hero trial button: label "Start My Free 7-Day Trial", matching the sales
    page. Destination changed from the placeholder /trial to
    /unlimited-english#ep-pricing, because the plan must be chosen before the
    trial starts and the home page carries no pricing block.
  - Hero trial microcopy now states the full terms: "7 days free, then $290 a
    year or $29 a month - Cancel anytime before day 8 - We will email you 2 days
    before charging you". Both plan totals appear at equal legibility, per the
    FTC / CA-ARL prominence principle. Not legal advice.
  - The "No credit card needed" line is now two lines behind the toggle.
    .ep-notrial-only keeps the current wording. .ep-trial-only reads "No credit
    card needed for the free lesson", so the no-card claim cannot be read as
    applying to the card-gated trial sitting under it.
  - UE card trial link relabelled and repointed to the same pricing anchor.
  - Trial FAQ added, hidden behind the toggle, wording lifted from the live
    sales page FAQ. It is deliberately NOT in the FAQPage JSON-LD, because CSS
    cannot toggle schema and the page must not advertise a switched-off trial.
    When the trial goes live, add that one Q and A to the JSON-LD by hand.
  - DISCREPANCY FOR JEFF, reproduced rather than resolved. Trial_v15 says
    "Cancel anytime before day 7" in its hero and sticky bar, and "cancel your
    trial anytime before Day 8. On Day 8, your credit card is charged" in its
    FAQ. Its own "How your free trial works" block labels the charge step Day 8
    and then says "You can cancel anytime online before day 7." The home page
    now carries both of Jeff's lines verbatim, so the same day 7 / Day 8
    mismatch is on this page too. Fixing it means changing Trial_v15 first, then
    matching here. It is a negative-option disclosure detail, so it is worth
    getting right before the trial launches. Not legal advice.
  - New markup-only i18n keys, all defaulting to inline English: hero_trial_cta,
    hero_trial_micro, hero_freecourse_micro_trial, ue_cta_trial, faq_trial_q,
    faq_trial_a. They need translation before the trial launches in a
    non-English market.
  - Prices unchanged and still reconcile: $24/mo billed annually ($290/year),
    $29/month, $290/year.

  v28 (build session, 2026-07-28) - applied to file v27.html
  HERO
  - H1: "Finally understand real English &bull; Then speak it with confidence"
    with no periods, per Jeff.
  - Subhead: "Built on 40 years of research into how adults actually become
    fluent, delivered in 20-minute lessons you can do anywhere" with no period.
    "acquire fluency" became "become fluent" and "daily" was dropped.
  - Portrait halved. .ep-hero-portrait went from max-width 480px with
    align-self:stretch to max-width 240px with a fixed 4/5 ratio, centred on
    both axes. Mobile 360px became 220px, stacked variant 380px became 240px.
    The name tag was scaled to fit (.8rem to .64rem, full-width and centred
    inside the card).
  - Microcopy under the primary button: "No credit card - Audio + text" became
    "No credit card needed".
  PROOF BAR
  - #ep-press padding went from 36px to 72px top and bottom, so the whole band
    is taller. The gap between the stats row and the logo row is untouched, as
    requested (.ep-stats margin-bottom, divider margins and eyebrow margins are
    all unchanged).
  - Stat numbers went from #fff to var(--ep-green), #159b97.
  BODY COPY
  - "1,800+ lessons on daily life, business, travel, idioms, culture, and more"
    applied to BOTH pillar2_mech_body and ue_b1, which said nearly the same
    thing in two different ways. FLAG: Jeff named one line, not two. Revert
    whichever one he did not mean.
  - how_step3: "Understand better, speak more fluently" became "i+1 Targeting
    takes you to the next level so you can speak more fluently".
  - magnet_sub AND the popup body key both became "A 10-minute audio lesson with
    the five exact phrases you can use when someone doesn't understand what you
    said in English" with no period. Per Jeff, the 17 non-English translations
    of the body key are deliberately NOT retranslated.
  - Sample lesson transcript: "goes to a store" became "goes to the store",
    "this 50 dollar package" became "this 50-dollar package", and "Finally, you
    hear the line again" became "Finally, you will hear the line again". FLAG:
    two earlier lines in the same block still read "you hear", so the three are
    now inconsistent. Jeff is checking the audio.
  - Learning Guide page count: "10+ page" and "8-10 page" became "10-page" in
    all three places (pillar2_body, curr_lead, curr_foot). Hyphenated because
    the phrase modifies "Learning Guide" in every one of those three lines.
  YOUR PROFESSOR
  - Photo swapped from /assets/lecturer-052526.jpg to
    https://www.eslpod.com/assets/Bovard_USC.png, with alt text for the Bovard
    Administration Building at USC.
  - founder_p1 already matched Jeff's requested bio word for word, so it is
    unchanged.
  - founder_p2 replaced with Jeff's new paragraph. "USC" is now spelled out as
    "the University of Southern California", and the "researching" link to
    eslpod.com/research.html is preserved with its existing underline styling.
  FOOTER
  - foot_tagline became "Bringing science to language learning since 2005" and
    the old foot_krashen line was removed as a duplicate. The byline follows on
    the next line. "Clear English lessons that help adults actually speak" is
    gone.
  TRIAL SCAFFOLDING
  - #ep-page now carries data-trial="off". Two rules in <style
    id="trial-toggle-v28"> hide .ep-trial-only while off and hide
    .ep-notrial-only while on. Flip that one word to launch the trial.
  - Two trial elements are wired and hidden: a hero button plus its terms line,
    and a text link in the Unlimited English card. Both carry
    data-track="cta_click_trial", which the page's existing GA4 listener already
    fires on, so trial click tracking works the moment the toggle flips.
  - UNVERIFIED and blocking a launch: the /trial destination is a placeholder,
    and the hero terms line does not yet state whether a card is required or
    what happens on day 8. Both need Jeff's confirmed trial terms. An HTML
    comment sits directly above the hero block saying so. Not legal advice.
  POPUP LANGUAGES
  - Indonesian, Polish and Thai added to the popup dropdown and to all 15
    STRINGS keys. Thirteen keys took their values from
    LandingPages_Home-IndexTest_v10.html. cta_trigger and fallback_link had no
    id/pl/th values anywhere, so those six strings are Claude's translations and
    are UNREVIEWED.
  - Three imported values were corrected: the Indonesian and Thai
    secure_microcopy used em dashes, which are banned, and the Indonesian,
    Polish and Thai lang_placeholder values were missing the dash brackets the
    other 14 languages use.
  - Prices unchanged and still reconcile: $24/mo billed annually ($290/year).

  v27 (merge session, 2026-07-27) - applied to file v26.html
  Merge of LandingPages_Home-IndexTest_v10.html copy + mobile CSS onto the v26
  standalone base. v10 had newer copy and a mobile hero fix, but its lead-magnet
  popup runtime was an older build that had lost the Meta Pixel Lead event, the
  GA4 lead_magnet_open / lead_magnet_submit / cta_click_trial events, the fbp /
  fbc / event_id hidden fields, and the browser-language detection and RTL
  handling. v26's tracking and language machinery are kept intact here.
  - New FAQ item (faq_english_q / faq_english_a): "Are all of your lessons 100%
    in English?" Added to the visible FAQ list and to the FAQPage JSON-LD.
    These two keys are markup-only, like every other FAQ key, so no STRINGS
    entry is needed.
  - Popup badge: "Trusted by 15,000,000 learners in 189 countries" became
    "Trusted by 15 million students in 189 countries" (en only; the non-English
    badge strings still read 15,000,000, which says the same thing).
  - Popup body: "A 10-minute audio lesson with five exact phrases..." became
    "This 10-minute lesson gives you 5 exact phrases to use when someone does
    not understand your English." All 14 translations updated from v10.
  - secure_microcopy: "Secure form." became "This form is secure." (fragment
    fix). All translations updated from v10, except that v10's Arabic, Simplified
    Chinese and Traditional Chinese values used em dashes, which are banned, so
    those three keep a period.
  - privacy: "Your email stays private. No spam, unsubscribe anytime." became
    "Your email stays private - No spam" (bullet separator). All translations
    updated from v10. NOTE: this drops "unsubscribe anytime" from the form.
  - Added the popup-hero-cta-fix-v4 style block from v10: below 700px the hero
    button goes full-width, centered, and forced orange.
  - NOT merged from v10: the three added popup languages (Indonesian, Polish,
    Thai). v10's STRINGS is missing cta_trigger and fallback_link entirely, so
    adding those three languages here would leave two keys falling back to
    English. Held as a separate task.
  - Prices are unchanged and reconcile: $24/mo billed annually ($290/year).

  v26 (build session, 2026-06-01) — applied to file v25.html
  • UE price-line compliance fix (Option B from pricing review). Changed the
    standalone "From $24/month" — which could be misread as a $24 monthly-billing
    option (actual monthly plan is $29; $24 is the annual-plan equivalent,
    $290/yr ÷ 12 = $24.17) — to "From $24/mo, billed annually ($290/year)". The
    billing total now sits adjacent and equally legible to the per-month figure,
    per the FTC/CA-ARL "total price at least as prominent" principle. Kept the
    round $24 anchor (per page-structure guidance: price transparency / anchoring
    is a documented conversion lever; hidden price = top bounce cause).
  • New i18n string ue_price_billed (", billed annually ($290/year)"). NOTE:
    translations not yet updated — translate ue_price_billed before localizing.
  • "/month" → "/mo" on ue_price_per to match the new shorter line.

  v25 (build session, 2026-06-01) — applied to file v24.html
  • UE Main Course card layout bugfix (the actual cause of the "content pushed
    left, blank right column" report — NOT the bullets). .ep-ue-card is a 2-col
    grid (body + mockup). The mobile single-column override in the ≤860px media
    query was written as ".ep-ue-card" (specificity 0,1,0), which is LESS
    specific than "[data-ue-layout=mockup-right] .ep-ue-card" (0,2,0) at the top
    of the stylesheet. Once data-ue-layout was hardcoded onto #ep-page (v22 dev-
    scaffolding strip), that 2-col rule won even on mobile; below 640px the
    mockup column is display:none, so the body sat in one track and the empty
    mockup track showed as a blank right column. Fix: bumped the responsive
    rule to "#ep-ue .ep-ue-card" (1,1,0) so it wins. Card now spans full width
    and the v24 centered content/bullets actually center across the whole card.
    Pre-existing since the 2-col card + tweaks default; surfaced now.

  v24 (build session, 2026-06-01) — applied to file v23.html
  • UE Main Course card bullets (mobile ≤640px): third pass. v22 (center each
    bullet around its checkmark) and v23 (center the whole left-aligned block)
    both read lopsided because a multi-line bullet list naturally fills the
    card width, leaving no slack to center into. v24 switches the bullets to a
    fully centered vertical stack — green check badge centered ON TOP of each
    bullet, bold heading + description centered below. li becomes a flex column
    (align-items:center); the ✓ glyph (li::after) is re-centered over the now-
    centered circle (left:50% + translateX(-50%)). Heading/lead/price unchanged
    (still centered). Desktop (>640px) layout untouched.

  v23 (build session, 2026-06-01) — applied to file v22.html
  • FAQ5 answer (faq5_a — "How is this different from Duolingo, Babbel, or
    online tutors?"): opening sentence swapped from "Duolingo and Babbel drill
    you in grammar, give you lots of output 'exercises,' and try to motivate
    you with prizes and 'streaks.'" → "Most apps and tutors focus on either
    rote memorization or 'output' practice, forcing you to speak before you're
    ready." Rest of the answer unchanged. Drifts faq5_a from non-EN locales.
  • UE Main Course card bullets (mobile ≤640px): replaced v22's per-bullet
    "center around the checkmark" (read lopsided — checkmark pinned to card
    edge, text floating center) with a centered block: .ep-ue-bullets shrinks
    to content width (fit-content + margin auto) and centers under the card,
    while each li keeps checkmark-left + text-left (grid 24px auto, text-align
    left). Heading/lead/price stay centered. Verify balance in a narrow browser.

  v22 (build session, 2026-06-01) — applied to file v21.html
  Built per LandingPages_SessionBrief_v24.md. Home-page items only; item 1
  (UE Sales page) and items 14-15 (English-only disclosure, Teachable
  checkout accordion) deferred per Jeff. No new assets supplied — nav logo,
  favicon, and press-logo items (5, 8, 9) NOT addressed this build; references
  unchanged and still pending re-cropped/retina/SVG sources from Jeff.
  • Item 4 (Safari nav CTA pill): root cause found — .ep-nav-cta is applied to
    a <button> but never reset -webkit-appearance, so Safari painted its native
    button background/border behind the rounded orange pill. Added
    -webkit-appearance:none; appearance:none; border:0; background-clip:
    padding-box to .ep-nav-cta. Confident fix (button + missing appearance
    reset is established Safari behavior); verify on a real Safari device.
  • Item 2 (UE Main Course mobile center-align): inside @media (max-width:640px),
    added text-align:center to #ep-ue .ep-h2/.ep-ue-lead/.ep-ue-price-line and
    centered each bullet's checkmark+text cluster (grid-template-columns:auto
    auto; justify-content:center). Bullet treatment = "centered around the
    checkmark" — chosen default; flag vs. plain centered stack for Jeff. NOTE:
    brief said ".ep-price-line"; actual class is ".ep-ue-price-line".
  • Item 3 (magnet CTA harmonization): "Get the 5 Phrases!" → "Get a free
    lesson" (magnet_cta). Magnet H2 headline left unchanged.
  • Item 6 (Curriculum DE Business pill): deleted "Arriving for an Appointment"
    example. Re-verify two-line wrap at column width in browser.
  • Item 7 (pillar2_body): replaced with the Learning Guide copy.
  • Item 10 (strip dev scaffolding): removed the Tweaks panel (TWEAK_DEFAULTS,
    applyTweaks, buildPanel, postMessage/editmode host hooks) and the dead
    carousel auto-rotate timer. The seven layout toggles are now hardcoded as
    static data-attributes on #ep-page so the render is unchanged. Manual
    carousel prev/next + dots retained.
  • Item 11 (ue_lead): dropped trailing period.
  • Item 12 (curr_lead): dropped trailing periods; internal period → " • ".
  • Item 13 (how_step2 — brief's "step 1"): "Listen to the Slow-Explain-Fast
    audio" → "Listen to the audio while you are on the train, in your car, on
    your walk, or drinking your coffee". (Target string is step 2 in markup;
    matched by string, not position.)
  Translation: all changed strings (pillar2_body, curr_lead, magnet_cta,
  ue_lead, how_step2, curr_de Business example) now drift from non-EN locales —
  adds to the resweep owed per brief.

  v20 (Jeff edits, 2026-05-27) — applied to file v17.html
  • Hero subhead rewritten (no terminal period per Jeff's spec):
    "20 minutes a day of real conversations, with lessons built on the only
    method proven to produce fluent adult speakers." →
    "Built on 40 years of research into how adults actually acquire fluency,
    delivered in daily 20-minute lessons you can do anywhere"
  • Big idea lead p: "Most apps and tutors teach the wrong things." → "Apps,
    AI chatbots, and live tutors focus on all the wrong things." Sharper
    competitive framing — names the actual category competitors (AI chatbots
    explicit) and "all the wrong things" hits harder than "the wrong things".
  • Magnet section bg reworked (Option A — white card on dark shelf):
    #ep-magnet bg moved from var(--ep-dark) to lighter warm charcoal #2b3a3a.
    Card flipped from transparent-on-dark to WHITE with strong dual-layer
    shadow (0 30px 60px rgba(0,0,0,.35) + 0 8px 20px rgba(0,0,0,.18)). All
    inner text colors reverted from white-on-dark back to standard dark-on-
    white (H2 dark, sub dark, body #444, microcopy dark). Orange CTA shadow
    pulled back to .35 alpha (no longer needs the extra punch since it sits
    on a white card). Treats the magnet as a featured CTA card lifted off a
    dark shelf — strongest contrast pattern for capture sections.
  • Logo .jpg → .png: nav <img>, JSON-LD Organization "logo" field, and
    favicon fallback all switched from /assets/ESLPodLogo.jpg to .png. The
    JPG version has black letterbox baked in; the PNG (used on
    /research_v9.html and similar pages) renders the logo cleanly without
    padding bars above and below.

  v19 (Jeff edits, 2026-05-27) — applied to file v16.html
  • Proof bar logos: removed filter: brightness(0) invert(1) — was stripping all
    brand color, making CNN/El País/Macworld/Deutschlandfunk/O Diário/Techwd into
    pure white silhouettes that defeated the trust-signal purpose. Restored color
    rendering at opacity .85 on dark bg.
  • #ep-magnet background: cream var(--ep-bg) → dark var(--ep-dark). Restores
    strict alternation (UE cream → Magnet dark → Products white → Founder cream)
    AND gives the LGM conversion moment its own visual stage. Card now sits as
    a transparent panel with rgba white border on the dark bg. All inner text
    converted to light: H2 #fff, eyebrow rgba(255,255,255,.55), sub/microcopy
    rgba(255,255,255,.78). Orange CTA button shadow intensified to .45 alpha
    for visibility against the dark backdrop.
  • UE feature card restructured: H2 "Unlimited English: 1,800+ lessons from
    $24/month" split into a clean hierarchy — eyebrow (THE MAIN COURSE) → H2
    (Unlimited English) → lead p (1,800+ lessons in clear, real English.) →
    dedicated green price line (From $24/month) → bullets → CTA. Price now has
    its own visual weight as a green emphasized line instead of being buried
    in the H2 wrap. New data-i18n keys: ue_lead, ue_price_from, ue_price_per.
    ue_h2 string changed from "Unlimited English: 1,800+ lessons from $24/month"
    to just "Unlimited English". Inline clamp() font-size on H2 removed
    (was forcing a 1.85rem floor that broke mobile layout).
  • UE card mobile (<640px) fixes: H2 font-size capped at 1.55rem (no more
    7-line wrap on narrow screens), price line shrunk to 1.7rem, bullet bold
    lead-in moved to display:block (so "1,800+ lessons" reads as a clean
    heading and the description flows under it, not a mid-phrase break),
    mockup illustrations hidden (decorative aria-hidden anyway, were cramping
    layout with broken overlap at narrow widths), card padding tightened to
    22px 18px from clamp 28-48px.
  • UE CTA "See plans & pricing →" gets white-space:nowrap so it stops
    wrapping onto 3 lines on narrow viewports.
  • Mobile drawer nav: added second .ep-mobile-section heading ("MORE") above
    Pricing + The Science, so those top-level nav items no longer visually
    nest under the LESSONS group. New data-i18n key: nav_more.

  v18 (Jeff edits, 2026-05-27) — applied to file v15.html
  • Hero subhead rewritten: "Built on 40 years of Dr. Stephen Krashen's research into how adults actually acquire fluency." → "20 minutes a day of real conversations, with lessons built on the only method proven to produce fluent adult speakers." Per Page-Structure brief: leads with the HOW (time + mechanism) since the eyebrow pill already carries the audience clarifier.
  • Background colors (Path A): #ep-press proof bar moved from cream var(--ep-bg) → dark var(--ep-dark). Stats numbers now #fff, labels rgba(255,255,255,.62), eyebrow rgba(255,255,255,.55), divider rgba(255,255,255,.12). Press logos inverted to white via filter: brightness(0) invert(1) at opacity .72 (hover 1.0). Fixes the cream-on-cream merge with #ep-audio below and gives the trust stats a stage.
  • Background colors (Path A): #ep-magnet moved from white → cream var(--ep-bg) so it breaks the white-on-white stack with #ep-products below.
  • 15M+ → 15M cleanup: founder credlist (cred3 string) and meta description both dropped the trailing + for consistency with the proof bar.
  • Footer Learn column reordered + Listen Free added + $7 7-Day Trial deleted. New order: Unlimited English / Apple Podcasts / Get a Free Course! / Listen Free (→ /listen/) / Special Courses. New data-i18n key: foot_listen.
  • Footer About column: "The Method" → "The Science" (foot_method i18n key kept; link still points at /research/).

  v17 (Jeff edits, 2026-05-26) — 19 queued items applied to file v14
  • Hero portrait pill: "Ph.D., Applied Linguistics + Education" → "Ph.D. in Applied Linguistics + Education".
  • Hero subhead: deleted the fragment after "fluency" per anti-AI guide. New: "Built on 40 years of Dr. Stephen Krashen's research into how adults actually acquire fluency."
  • Proof bar (#ep-press): border-bottom removed so it blends into cream #ep-audio band below. Logos 32px → 22px, opacity .9 → .65, hairline divider added between stats and logos. Stats: "15M+" → "15M", "20+ yrs" → "21 yrs", "1,800+" kept.
  • Audio intro: deleted "Notice how" prefix. New: "Every difficult word is explained in clear English you already understand."
  • Audio after-text: "Hear another 60 minutes." → "Want to hear more?"
  • Audio scrubber: play button now toggles between play ▶ and pause ❚❚ icons via play/pause event listeners on audioEl. Click-to-seek upgraded to support mouse drag and touch drag. If audio was paused when seek completes, playback auto-resumes.
  • Nav restructure: "Courses" → "Lessons" (Listen Free folded into the dropdown alongside UE + Apple Podcasts). "How It Works" → "The Science". Added "Pricing" → /unlimited-english#ep-pricing. About removed from primary nav (stays in footer column). Mobile drawer mirrors the new structure.
  • Big idea pillar 2 H3: "Get large amounts of input." → "Get large amounts of input" (no trailing period).
  • Favicon: added <link rel="icon"> and <link rel="apple-touch-icon"> pointing at /assets/favicon.png and /assets/apple-touch-icon.png. Existing JPG kept as fallback.
  • UE H2: "from $24.17/m" → "from $24/month".
  • UE bullet 4: "Listen anywhere. Phone, browser, iOS or Android app" → "Listen anywhere on your phone, browser, and iOS or Android app".
  • Founder credlist: "20+ years of ESLPod.com" → "21 years of ESLPod.com".
  • Testimonials (#ep-celebrate) background: var(--ep-bg) cream → #f0ebe1 warm sand/beige (breaks identical-cream stack with founder section above).
  • Testimonials reorder: 1. Jake (was "cooljake", renamed) — 2. Dmitry — 3. Yuzo (NEW, Japan, TOEIC 570→840) — 4. Farrie — 5. Raphael — 6. Jhonatan — 7. Tsolmon — 8. Marco — 9. Txema. Tony deleted. Yuzo's semicolon scrubbed to a period per voice rule.
  • Footer Learn column: "$1 7-Day Trial" → "$7 7-Day Trial".
  • #19: Pricing nav item points at /unlimited-english#ep-pricing anchor — no new pricing page needed.

  v16 (Jeff edits, 2026-05-18)
  • Hero trust strip: "15M+ learners" → "1 billion downloads" (kills duplication with eyebrow "Join 15 million people").
  • UE feature card secondary link: "Or get a free course first →" (modal) → "See everything inside Unlimited English →" (/unlimited-english, top of sales page). Restores in-body path to the UE sales-page narrative for visitors who want the case made, not just pricing. Free course remains exposed via dedicated #ep-magnet section directly below, mobile sticky bar, nav button, and footer — no loss of free-course visibility.

  v15 (Jeff edits, 2026-05-18)
  • Hero microcopy: "1 hour of audio" → "Audio + text".
  • Hero: deleted ghost "Or try Unlimited English for $1 →" button entirely. Free course is now the sole hero CTA. Trial path still available via mobile sticky bar's small text link and via footer.
  • Audio sample meta: "120 seconds · From the Daily English library" → "Complete lessons have 20-25 minutes of audio + 30-40 minutes of reading material" (reframes from "tiny sample" to "tiny sample of much bigger lesson").
  • UE feature card "See plans & pricing" CTA: destination changed from /trial to /unlimited-english#ep-pricing (deep-links to the pricing section anchor on the UE sales page per NN/g link-promise rule). Tracking event renamed cta_click_trial → cta_click_pricing.
  • FAQ swap: deleted "Is the $1 trial really $1?" and "What happens after the 7-day trial ends?". Added "Is this only for American English?" and "Can I download lessons for offline?".
  • Trademark: added ® superscript to every visible "ESLPod" instance in body content (founder paragraph, stats strip, all 5 testimonial quotes that mention the brand). Left untouched: meta tags, title, JSON-LD, alt attributes, aria-labels, HTML comments, and any "ESLPod.com" lowercase-domain instances (footer logo brand-mark, copyright line, founder credlist "20+ years of ESLPod.com").

  v14 (CTA architecture pass, 2026-05-18)
  • Top-nav (desktop + mobile drawer): "Try for $1" → "Get a free course" (popup-trigger). Persistent nav CTA now mirrors LGM.
  • Hero: trial button demoted to ghost style; copy switched to "Or try Unlimited English for $1 →" (explicit downgrade prefix). Equal-weight competition with free course resolved.
  • Hero: deleted "Having trouble? Click here." link (broken-promise per NN/g; popup-blocker fallback JS at lines 2172-2187 handles edge cases silently). Replaced with doubt-remover microcopy: "No credit card · 1 hour of audio · Yours forever".
  • UE feature card: "Start with $1 (7-day trial)" → "See plans & pricing →" (Highrise/Going.com pattern, softer navigational verb). Sibling text link replaced with popup-trigger "Or get a free course first →" — explicit LGM downgrade path mid-page.
  • Free course capture section: deleted second "Having trouble? Click here." link (same broken-promise issue).
  • Audio sample after-text: "Get a free course →" now triggers modal (consistency — was direct link to signup.html, now matches every other free-course entry point).
  • Mobile sticky bar: added quiet text link "or try Unlimited English for $1 →" beneath primary free-course CTA. Restores trial path for deep-scroll mobile readers without making the bar two-CTA. Body padding-bottom bumped 84→108px.
  • Footer Product column: "Get a Free Course!" moved from position 5 to position 1.

  v2 (Jeff edits, 2026-05-11)
  • Header: dropped English Collections dropdown + TV link (mobile + desktop).
  • Hero: subhead rewritten to full sentences; founder-photo placeholder text removed.
  • Method: em dashes scrubbed; "fastest way possible" rewritten to "the way 40 years of research says..."; Step 3 closing line rewritten; "Research-driven" kept (per Jeff).
  • Audio sample: rebuilt as "Competing in Business" (120s), de984.mp3, UE-Sales-style transcript with Yoko / freebies / gift-with-purchase. Below-card CTA switched to popup-trigger "Get a free English course".
  • UE card: full-sentence subhead; lesson screenshots wired to eslpod.com/assets/{web,app}screenshot.png.
  • Free course: trimmed "No tricks, no credit card needed"; under-CTA microcopy becomes "No credit card needed" (no trailing period in v11); em dash scrubbed in modal body (EN + all 14 i18n strings).
  • Other products grid: section deleted entirely.
  • Founder: Lucy removed; "Dr. Jeff McQuillan" H2 (no period); body bio adds "downloaded more than 1 billion times"; Krashen em dashes -> commas; Lucy credential line deleted.
  • Stats strip: "Of teaching" -> "Of ESLPod". 189 Countries retained per Jeff.
  • Testimonials: "Real stories. Real countries. Real progress." H2 deleted. 10 fabricated/Lorem slots replaced with Best-18 picks (Yuzo, Raphael, Sarah, Luan, Tony, Jake, Txema, Ayse, Marco, Liu).
  • FAQ: kicker eyebrow removed; H2 = "Common Questions"; Q1 (free podcast vs UE) deleted; trial answer em dash -> period; Cancel answer rewritten with full membership auto-renew disclosure + Jeff's "cancel from your account easily"; "Different." -> "They're different products."; Duolingo em dashes -> colon + period.
  • Footer: drops Lucy from byline; replaces Krashen line with "Bringing science to language learning on the internet."; Product column: English Collections -> Special Courses (#), TV Courses deleted; Company column deleted; Connect column rebuilt with Podcasts (was Apple Podcasts), YouTube, X (Twitter), Facebook, Instagram, RSS, Member login, Contact; legal links drop Refund Policy + Contact (kept Privacy + Terms); trust-strip emojis + "every course" wording retained per Jeff.
  • UE bullet em dashes (2 & 4) scrubbed.
  • FAQ Q3 em dash scrubbed.
  • Meta description: Lucy removed.

  v1 (initial)
  • Sticky full top menu (Foundation §4.1).
  • Hero: Jeff founder portrait placeholder + dual CTA (orange popup-trigger primary, green Try-for-$1 secondary).
  • Big idea band: 3 columns w/ Slow-Explain-Fast Technique™ + i + 1 Targeting™ nested.
  • AudioSample (.tl-audio-card): 60-sec Daily English placeholder lesson "Asking for the Check at a Restaurant".
  • UE feature card: stylized phone+desktop SVG faux-screenshots.
  • Free course capture section: orange popup-trigger pill, no embedded form.
  • Other products grid: Apple Podcasts / English Collections / TV.
  • Founder credibility: Jeff portrait placeholder + Krashen co-author lineage (Lucy mentioned, no photo per user).
  • Testimonials carousel: Yuzo (Japan), Hélio (Brazil) pulled from Trial TY HTML + 8 Lorem ipsum.
  • FAQ snippet (5 questions).
  • Full 5-column footer (Foundation §5.1).
  • Lead-magnet popup: modal + runtime embedded VERBATIM from index-1.html.
  • Sticky bottom mobile CTA: Try for $1 (green) per spec default.
  • Tweaks: hero layout, big-idea style, testimonial carousel, mobile CTA, primary hero CTA color.
-->
<!-- Page language, handed to GTM. Added 2026-09-04.
     MUST stay above the GTM snippet: the Data Layer Variable reads the merged model,
     so pushing here means the very first page_view carries lang. Below it, the most
     important event of the session goes out without one.
     The GTM forwarder half comes AFTER these pages are live; wiring it first sends
     empty values on every event. -->
<script>window.dataLayer=window.dataLayer||[];window.dataLayer.push({lang:"en"});</script>
<!-- ===== Google Tag Manager (loads GA4 G-1X3C30PB35) ===== -->
<script>(function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':
new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],
j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src=
'https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);
})(window,document,'script','dataLayer','GTM-P8GJZSFN');</script>
<!-- ===== Meta Pixel ===== -->
<script>
!function(f,b,e,v,n,t,s){if(f.fbq)return;n=f.fbq=function(){n.callMethod?
n.callMethod.apply(n,arguments):n.queue.push(arguments)};if(!f._fbq)f._fbq=n;
n.push=n;n.loaded=!0;n.version='2.0';n.queue=[];t=b.createElement(e);t.async=!0;
t.src=v;s=b.getElementsByTagName(e)[0];s.parentNode.insertBefore(t,s)}(window,
document,'script','https://connect.facebook.net/en_US/fbevents.js');
fbq('init', '1619730374980621');
fbq('track', 'PageView');
</script>
<noscript><img alt="" height="1" src="https://www.facebook.com/tr?id=1619730374980621&ev=PageView&noscript=1" style="display:none" width="1"/></noscript><style id="trial-toggle-v28">
/* v28: seven-day free trial scaffolding.
   #ep-page carries data-trial="off" (launch state) or data-trial="on".
   Flip that one word to reveal every .ep-trial-only element at once and hide
   every .ep-notrial-only element. Nothing else needs to change. */
#ep-page:not([data-trial="on"]) .ep-trial-only   { display: none !important; }
#ep-page[data-trial="on"]       .ep-notrial-only { display: none !important; }
#ep-hero .ep-hero-trial-cta {
  display: block; width: 100%; max-width: 460px;
  margin: 12px auto 0; text-align: center;
}
#ep-hero .ep-hero-trial-micro {
  margin-top: 8px; font-size: 12px; line-height: 1.5; text-align: center;
  color: rgba(0,0,0,.55);
}
</style>
<style id="popup-hero-cta-fix-v4">
@media (max-width: 700px){
  #ep-hero .ep-hero-cta-row{ justify-content:center; }
  #ep-hero .ep-hero-cta-col{ width:100%; }
  #ep-hero .ep-btn-large{ background:var(--ep-orange)!important; display:block; width:100%; max-width:460px; margin-left:auto; margin-right:auto; text-align:center; }
}
</style>
<style id="ep-palette-baked">
:root{--ep-dark:#1e1c26;--ep-darker:#15131b;--ep-green:#413b66;--ep-green2:#d9d3ee;--ep-pill-bg:#1e1c26;--ep-pill-text:#d9d3ee;--ep-orange:#6a3fb5;--ep-green-cta:#6a3fb5;--ep-blue:#413b66;--ep-red:#a8392f;--ep-muted:#666070;--ep-bg:#f6f5f9;--ep-border:#e3e0ea;}
.ep-btn,.ep-btn-green,.ep-btn-large,.ep-nav-cta,.ep-mobile-stickycta a{
  background:var(--ep-green-cta) !important;color:#fff !important;
  box-shadow:0 8px 22px color-mix(in srgb, var(--ep-green-cta) 42%, transparent) !important;
  transition:filter .15s ease, transform .12s ease, box-shadow .15s ease !important;}
.ep-btn:hover,.ep-btn-green:hover,.ep-btn-large:hover,.ep-nav-cta:hover,.ep-mobile-stickycta a:hover{
  filter:brightness(1.06) !important;transform:translateY(-2px) !important;
  box-shadow:0 12px 30px color-mix(in srgb, var(--ep-green-cta) 55%, transparent) !important;}
.ep-btn-ghost{background:transparent !important;box-shadow:none !important;color:var(--ep-dark) !important;}
#ep-hero .ep-btn-ghost,.ep-hero .ep-btn-ghost,[data-bg="dark"] .ep-btn-ghost{color:#fff !important;}
.ep-btn-ghost:hover{box-shadow:none !important;}
.ep-pc-badge{box-shadow:0 6px 16px color-mix(in srgb, var(--ep-green) 38%, transparent) !important;}
.ep-guarantee-badge{box-shadow:0 4px 14px color-mix(in srgb, var(--ep-green) 38%, transparent) !important;}
.ep-pc-bonus{background:color-mix(in srgb, var(--ep-green) 10%, #fff) !important;
  border-color:color-mix(in srgb, var(--ep-green) 40%, transparent) !important;}
.ep-case-tag{background:color-mix(in srgb, var(--ep-blue) 12%, #fff) !important;}


.ep-topics-rail--daily .ep-topics-chip{background:color-mix(in srgb, var(--ep-green) 8%, #fff) !important;
  border-color:color-mix(in srgb, var(--ep-green) 35%, #fff) !important;}
.tl-audio-tag{background:color-mix(in srgb, var(--ep-orange) 55%, transparent) !important;
  border-color:color-mix(in srgb, var(--ep-orange) 65%, transparent) !important;color:#fff !important;}
.tl-audio-card::after{background:radial-gradient(circle, color-mix(in srgb, var(--ep-orange) 20%, transparent), transparent 70%) !important;}
.tl-audio-play{background:var(--ep-orange) !important;
  box-shadow:0 4px 14px color-mix(in srgb, var(--ep-orange) 42%, transparent) !important;}
.tl-audio-play:hover{background:var(--ep-orange) !important;filter:brightness(1.08) !important;}
.tl-audio-bar-fill{background:var(--ep-orange) !important;}
.tl-audio-line{border-left-color:color-mix(in srgb, var(--ep-green) 55%, transparent) !important;}
.ep-topics summary{background:var(--ep-orange) !important;
  box-shadow:0 6px 18px color-mix(in srgb, var(--ep-orange) 32%, transparent) !important;}
.ep-topics summary:hover{background:var(--ep-orange) !important;filter:brightness(1.08) !important;}
.ep-topics summary:focus-visible{outline-color:var(--ep-orange) !important;}
.ep-topics-rail--cultural .ep-topics-chip{background:color-mix(in srgb, var(--ep-orange) 8%, #fff) !important;
  border-color:color-mix(in srgb, var(--ep-orange) 35%, #fff) !important;}
.ep-final-icon{box-shadow:0 8px 22px color-mix(in srgb, var(--ep-orange) 45%, transparent) !important;}
.ep-method-step .icon-wrap,.ep-inside-card .icon-wrap{
  background:color-mix(in srgb, var(--ep-green) 12%, #fff) !important;color:var(--ep-green) !important;}
</style>
</head><body>
<noscript><iframe height="0" src="https://www.googletagmanager.com/ns.html?id=GTM-P8GJZSFN" style="display:none;visibility:hidden" width="0"></iframe></noscript>
<!-- v32: GTM noscript moved here, directly after <body>, per Analytics_Instrumentation-Spec_v1.md -->
<style>
/* ─── Brand tokens (Foundation §3.1) ───────────────────────── */
:root {
  --ep-dark:    #1e1c26;
  --ep-darker:  #15131b;
  --ep-green:   #413b66;
  --ep-green2:  #d9d3ee;
  --ep-pill-bg: #1e1c26;
  --ep-pill-text: #d9d3ee;
  --ep-orange:  #6a3fb5;
  --ep-muted:   #666070;
  --ep-bg:      #f6f5f9;
  --ep-border:  #e3e0ea;
  --ep-r:       12px;
  --ep-blue:    #413b66;
  --ep-red:     #a8392f;
}

html, body { margin: 0; padding: 0; }
body {
  font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, Helvetica, Arial, sans-serif;
  color: var(--ep-dark); background: #fff;
  -webkit-font-smoothing: antialiased;
}

/* ── v57: PER-LANGUAGE FONT STACKS ────────────────────────────────────
   v56 had none. The only lang-scoped rules were on .lm-body, the modal, so
   the six non-Latin locales rendered the whole page in the Latin fallback:
   CJK fell back to system Han with wrong regional glyph forms, Arabic to
   whatever the OS chose, Thai to a face with no proper tone-mark stacking.
   These sit on html[lang] so they apply to the static per-language pages
   the eslpod-seo-structure split produces, and they inherit into .lm-body.  */
html[lang="ja"], html[lang="ja"] body {
  font-family: -apple-system, BlinkMacSystemFont, "Hiragino Sans", "Hiragino Kaku Gothic ProN",
    "Yu Gothic", "Noto Sans JP", Meiryo, sans-serif;
}
html[lang="ko"], html[lang="ko"] body {
  font-family: -apple-system, BlinkMacSystemFont, "Apple SD Gothic Neo", "Malgun Gothic",
    "Noto Sans KR", "Nanum Gothic", sans-serif;
}
html[lang="zh-Hans"], html[lang="zh-Hans"] body {
  font-family: -apple-system, BlinkMacSystemFont, "PingFang SC", "Hiragino Sans GB",
    "Microsoft YaHei", "Noto Sans SC", sans-serif;
}
html[lang="zh-Hant"], html[lang="zh-Hant"] body {
  font-family: -apple-system, BlinkMacSystemFont, "PingFang TC", "Microsoft JhengHei",
    "Noto Sans TC", sans-serif;
}
html[lang="ar"], html[lang="ar"] body {
  font-family: -apple-system, BlinkMacSystemFont, "Geeza Pro", "SF Arabic", "Segoe UI",
    "Noto Naskh Arabic", "Traditional Arabic", Tahoma, sans-serif;
}
html[lang="th"], html[lang="th"] body {
  font-family: -apple-system, BlinkMacSystemFont, "Thonburi", "Noto Sans Thai",
    "Leelawadee UI", Sarabun, Tahoma, sans-serif;
}
/* Thai needs the extra leading or tone marks collide with the line above. */
html[lang="th"] body { line-height: 1.75; }
/* ─────────────────────────────────────────────────────────────────────── */

#ep-page * { box-sizing: border-box; }

/* ─── Type ─────────────────────────────────────────────────── */
.ep-h1 { font-size: clamp(2rem,4.6vw,3rem); font-weight: 900; line-height: 1.12; color: var(--ep-dark); margin: 0 0 .55em; letter-spacing: -.015em; text-wrap: pretty; }
.ep-h1-light { color: #fff; }
.ep-h2 { font-size: clamp(1.5rem,3.2vw,2.2rem); font-weight: 900; line-height: 1.18; color: var(--ep-dark); margin: 0 0 .5em; letter-spacing: -.012em; text-wrap: pretty; }
.ep-h2-light { color: #fff; }
.ep-h3 { font-size: clamp(1.05rem,2vw,1.25rem); font-weight: 800; line-height: 1.3; color: var(--ep-dark); margin: 0 0 .4em; }
.ep-lead { font-size: clamp(1rem,2vw,1.12rem); line-height: 1.65; margin: 0 0 1em; color: #333; text-wrap: pretty; }
.ep-lead-light { color: rgba(255,255,255,.86); }
.ep-eyebrow {
  display: inline-block;
  background: var(--ep-pill-bg);
  color: var(--ep-pill-text);
  padding: 9px 20px;
  border-radius: 999px;
  font-size: .95rem;
  font-weight: 800;
  letter-spacing: .12em;
  text-transform: uppercase;
  margin: 0 0 1.1em;
  line-height: 1.1;
}
/* Translucent pill for dark hero bg */
#ep-hero .ep-eyebrow {
  background: rgba(197, 232, 138, 0.10);
  border: 1px solid rgba(197, 232, 138, 0.35);
  color: var(--ep-pill-text);
}
[data-hero-bg="light"] #ep-hero .ep-eyebrow {
  background: var(--ep-pill-bg);
  border: 0;
  color: var(--ep-pill-text);
}
/* v13: Founder is now on light bg, use default solid pill */
#ep-founder .ep-eyebrow {
  background: var(--ep-pill-bg);
  color: var(--ep-pill-text);
  border: 0;
}
/* Legacy muted variant — kept so existing markup still renders, no-op now */
.ep-eyebrow-muted { /* override the old muted text-only variant */ }
.ep-center { text-align: center; }

.ep-wrap      { max-width: 900px;  margin: 0 auto; padding: 0 20px; }
.ep-wrap-wide { max-width: 1100px; margin: 0 auto; padding: 0 20px; }
.ep-sec       { padding: clamp(48px, 7vw, 80px) 0; }

/* ─── Buttons ──────────────────────────────────────────────── */
.ep-btn {
  display: inline-block; box-sizing: border-box;
  background: var(--ep-orange); color: #fff !important;
  font-size: 1.02rem; font-weight: 800;
  padding: 16px 26px; border: 0; border-radius: 32px;
  text-align: center; cursor: pointer; line-height: 1.3;
  text-decoration: none;
  box-shadow: 0 8px 24px color-mix(in srgb, var(--ep-orange) 35%, transparent);
  transition: filter .15s ease, transform .1s ease;
  font-family: inherit;
}
.ep-btn:hover { filter: brightness(.94); transform: translateY(-2px); }
.ep-btn-green { background: var(--ep-green) !important; box-shadow: none; }
/* Flat hero CTA — no glow, no fuzz */
#ep-hero .ep-btn { box-shadow: none !important; text-shadow: none; }
.ep-btn-large { font-size: 1.12rem; padding: 20px 32px; }
.ep-btn-block { display: block; width: 100%; max-width: 460px; margin: 0 auto; }
.ep-btn-ghost { background: transparent !important; color: #fff !important; box-shadow: none; border: 2px solid rgba(255,255,255,.32); padding: 14px 24px; }
.ep-btn-ghost:hover { background: rgba(255,255,255,.06) !important; border-color: rgba(255,255,255,.5); }

.ep-textlink { color: var(--ep-dark); font-weight: 700; text-decoration: none; border-bottom: 2px solid var(--ep-green); padding-bottom: 1px; }
.ep-textlink:hover { color: var(--ep-darker); border-bottom-color: var(--ep-green2); }
.ep-textlink-light { color: #fff; border-bottom-color: var(--ep-green2); }
.ep-textlink-light:hover { color: #fff; border-bottom-color: #fff; }

.ep-fine { font-size: 13px; color: rgba(255,255,255,.7); text-align: center; line-height: 1.5; }
.ep-fine-dark { font-size: 13px; color: var(--ep-muted); text-align: center; line-height: 1.5; }

/* ─── Sticky top nav (Foundation §4) ───────────────────────── */
#ep-nav {
  position: sticky; top: 0; z-index: 90;
  background: var(--ep-dark);
  border-bottom: 1px solid rgba(255,255,255,.06);
}
.ep-nav-inner {
  max-width: 1280px; margin: 0 auto;
  display: flex; align-items: center; gap: 18px;
  padding: 12px 20px;
}
.ep-logo { display: inline-flex; align-items: center; text-decoration: none; flex-shrink: 0; }
.ep-logo svg { display: block; height: 34px; width: auto; }
/* LOGO — v43 (2026-08-04). Goes live WITH the asset swap, not before.
   /assets/ESLPodLogo.png is being replaced by the new green badge, trimmed to
   867x306. The supplied file was 940x788 with the badge occupying only
   867x306 of it, so about 61% of that canvas was empty and a 72px box would
   have rendered the badge at roughly 28px. The drop shadow was trimmed off on
   Jeff's call: it is a soft dark shadow on a #2b3336 bar, so it never showed,
   and it added 7 to 11px of dead space plus an off-centre placement.

   The badge renders 60px and the box is 84px. Every page here computes
   box-sizing: border-box, so 84 - 13 - 11 leaves exactly 60px of content
   height. The 2.5px on the left is inherited from the wordmark era and keeps
   the mark starting where it always did.

   WHAT CHANGED FROM v39. v39 kept the box at 72px and set the wordmark size
   with the padding, and its comment said the two numbers must always sum to
   72. That rule was written for a bare wordmark and does not survive a badge.
   The badge carries its own green field, so matching the old 48px letter
   height would need a 79px badge and a 128px header bar. Jeff chose 60px from
   a six-way render, which puts the letters near 36px and the header at 109px,
   up 13px from 96px. Change the badge size by changing the height and leaving
   the padding alone: 13 and 11 are now fixed, and the height is the dial.

   Inside the badge the letters plus the wifi arc are 61% of the badge height.
   That ratio is how you convert any future badge height into a letter height.

   Rendered and measured at 60px: logo box 173 x 84, header bar 109px desktop.

   Two traps. Do not read 84px here as the badge size. Do not upload
   ESLPodLogo_v2-forUpload.png, which is padded differently and needs
   different numbers. */
/* v50: handoff §4 nav standard — 40px logo, no padding, bar = 40 + 24 = 64px.
   Mobile drops to 32px so the bar lands at 56px, matching .ep-mobile-drawer's
   inset. Supersedes the v43 60px-badge sizing described above. */
.ep-logo img { display: block; height: 40px; width: auto; padding: 0; }
@media (max-width: 960px) { .ep-logo img { height: 32px; } }
.ep-nav-spacer { flex: 1 1 auto; }
.ep-nav-list {
  display: flex; align-items: center; gap: 4px; list-style: none; margin: 0; padding: 0;
}
.ep-nav-list > li { position: relative; }
.ep-nav-link {
  display: inline-flex; align-items: center; gap: 6px;
  color: #fff; font-size: .95rem; font-weight: 500;
  text-decoration: none; padding: 10px 12px; border-radius: 8px;
  cursor: pointer; background: transparent; border: 0; font-family: inherit;
}
.ep-nav-link:hover { font-weight: 700; background: rgba(255,255,255,.04); }
.ep-nav-caret { color: var(--ep-green2); font-size: .7rem; line-height: 1; transform: translateY(1px); }
.ep-nav-drop {
  position: absolute; top: calc(100% + 4px); left: 0;
  background: #353535; border-radius: 10px; padding: 6px;
  min-width: 220px;
  box-shadow: 0 16px 40px rgba(0,0,0,.32);
  display: none; flex-direction: column; gap: 1px;
  z-index: 100;
}
.ep-nav-drop.open { display: flex; }
.ep-nav-drop a {
  display: flex; align-items: center; gap: 8px;
  color: #fff; text-decoration: none; padding: 10px 12px;
  font-size: .92rem; border-radius: 6px;
}
.ep-nav-drop a:hover { background: rgba(255,255,255,.08); }
.ep-nav-drop a::before {
  content: "▸"; color: var(--ep-green2); font-size: .7rem; flex-shrink: 0;
}
.ep-nav-cta {
  display: inline-block; background: var(--ep-orange); color: #fff !important;
  font-weight: 800; font-size: .92rem;
  padding: 10px 18px; border-radius: 999px; text-decoration: none;
  margin-left: 6px;
  box-shadow: 0 4px 14px rgba(244,103,46,.35);
  /* v22 item 4: reset native button chrome — .ep-nav-cta is applied to a
     <button>, and without this reset Safari paints its default button
     background/border behind the rounded orange pill (the white bg/shadow
     ring artifact). border-box clip stops any background bleed at corners. */
  -webkit-appearance: none; appearance: none;
  border: 0; background-clip: padding-box;
}
.ep-nav-cta:hover { filter: brightness(.94); }
/* v13: login as plain text link, not a pill */
.ep-nav-loginlink {
  color: rgba(255,255,255,.78); font-size: .9rem; font-weight: 500;
  text-decoration: none; padding: 8px 10px; border-radius: 6px;
  margin-right: 6px;
}
.ep-nav-loginlink:hover { color: #fff; background: rgba(255,255,255,.04); }

/* Mobile nav */
.ep-nav-burger {
  display: none; background: transparent; border: 0; color: #fff;
  font-size: 1.4rem; cursor: pointer; padding: 8px 10px; border-radius: 8px;
}
.ep-nav-burger:hover { background: rgba(255,255,255,.06); }
.ep-mobile-drawer {
  display: none;
  position: fixed; inset: 56px 0 0 0;
  background: var(--ep-dark);
  padding: 24px 20px; overflow-y: auto;
  z-index: 89;
  flex-direction: column; gap: 4px;
}
.ep-mobile-drawer.open { display: flex; }
.ep-mobile-drawer a, .ep-mobile-drawer button {
  display: block; color: #fff; text-decoration: none;
  padding: 14px 12px; font-size: 1.05rem; font-weight: 600;
  border: 0; background: transparent; text-align: left; font-family: inherit;
  border-bottom: 1px solid rgba(255,255,255,.06);
  cursor: pointer; width: 100%;
}
.ep-mobile-drawer .ep-mobile-section {
  font-size: .76rem; font-weight: 800; letter-spacing: .12em; text-transform: uppercase;
  color: var(--ep-green2); padding: 18px 12px 6px; border: 0;
}
.ep-mobile-drawer .ep-nav-cta { margin: 16px 12px 0; text-align: center; }

@media (max-width: 960px) {
  .ep-nav-list, .ep-nav-cta-desk { display: none; }
  .ep-nav-burger { display: inline-flex; align-items: center; }
}
@media (min-width: 961px) {
  .ep-nav-burger { display: none; }
  .ep-mobile-drawer { display: none !important; }
}

/* Language pill (in nav) — matches Trial Thank You page */
.ep-lang-wrap { position: relative; flex-shrink: 0; }
.ep-lang-pill {
  display: inline-flex; align-items: center; gap: 6px;
  background: #fff; color: #1d5c1d;
  font-family: inherit;
  border: 1px solid #b2dcb2; border-radius: 20px;
  padding: 6px 12px; font-size: 13px; font-weight: 700;
  cursor: pointer;
  transition: background .15s ease, border-color .15s ease;
}
.ep-lang-pill:hover { background: #f4fbf3; border-color: var(--ep-green); }
.ep-lang-pill .ep-globe { color: var(--ep-green); display: inline-flex; }
.ep-lang-pill .ep-globe svg { width: 14px; height: 14px; display: block; }
.ep-lang-pill .ep-caret { font-size: .72rem; opacity: .65; }
.ep-lang-menu {
  position: absolute; top: calc(100% + 8px); right: 0;
  background: #fff; border: 1px solid var(--ep-border);
  border-radius: 12px; padding: 6px;
  box-shadow: 0 16px 44px rgba(0,0,0,.18);
  display: none; flex-direction: column; gap: 1px;
  min-width: 220px; max-height: 380px; overflow-y: auto;
  z-index: 100;
}
.ep-lang-menu.open { display: flex; }
html[dir="rtl"] .ep-lang-menu { right: auto; left: 0; }
@media (max-width: 560px) { .ep-lang-menu { right: 12px; left: auto; min-width: 240px; } }
/* v61: the options are anchors now, not buttons, so a crawler can follow them
   and a visitor can open one in a new tab. Both selectors are kept so the
   styling survives either markup. */
.ep-lang-menu button,
.ep-lang-menu a {
  display: flex; align-items: center; justify-content: space-between; gap: 12px;
  background: transparent; border: 0; padding: 10px 12px;
  font-family: inherit; font-size: 13px; color: var(--ep-dark);
  text-align: left; cursor: pointer; border-radius: 8px;
  width: 100%; font-weight: 600; text-decoration: none;
}
html[dir="rtl"] .ep-lang-menu button,
html[dir="rtl"] .ep-lang-menu a { text-align: right; }
.ep-lang-menu button:hover,
.ep-lang-menu a:hover { background: #f0f7f0; }
.ep-lang-menu button.active,
.ep-lang-menu a.active,
.ep-lang-menu button[aria-selected="true"],
.ep-lang-menu a[aria-selected="true"] { background: #eafbea; color: #1d5c1d; font-weight: 800; }
.ep-lang-opt-code {
  font-size: .72rem; font-weight: 800; letter-spacing: .06em;
  color: var(--ep-muted); text-transform: uppercase;
}

/* ─── Hero (#ep-hero) ──────────────────────────────────────── */
#ep-hero {
  background: var(--ep-dark);
  color: #fff;
  position: relative;
  overflow: hidden;
}
#ep-hero::before {
  content: ""; position: absolute;
  top: -30%; right: -10%; width: 60%; height: 100%;
  background: radial-gradient(ellipse at center, rgba(217,211,238,.10) 0%, transparent 60%);
  pointer-events: none;
}
.ep-hero-grid {
  /* v31: portrait track fixed at 300px instead of 1fr. The 1.1fr/1fr split was
     sized for the pre-v28 480px photo; the photo is 240px now, so that column
     was holding empty space the headline needed. */
  display: grid; grid-template-columns: minmax(0, 1fr) 220px;
  grid-template-areas:
    "eyebrow ."
    "copy    portrait";
  column-gap: clamp(24px, 3vw, 40px); /* v31: tightened so the headline gets the width */
  row-gap: 18px;
  align-items: stretch; /* v19: portrait matches h1-top → trust-bottom */
  padding: clamp(56px, 8vw, 88px) 0 clamp(44px, 6vw, 64px);
  position: relative;
}
/* v50: justify-self:start stops the grid item stretching the pill to the full
   column width. inline-block alone loses to the grid default of stretch. */
.ep-hero-grid > .ep-eyebrow { grid-area: eyebrow; align-self: start; justify-self: start; margin: 0; }
.ep-hero-copy { grid-area: copy; }
.ep-hero-portrait { grid-area: portrait; }
.ep-hero-copy { min-width: 0; }
.ep-hero-portrait {
  position: relative;
  width: 100%;
  justify-self: stretch;
  align-self: stretch;
  /* v18: padding removed so photo fills the rounded holder edge-to-edge */
}
.ep-hero-copy { min-width: 0; }
.ep-hero-portrait {
  position: relative;
  /* v28: halved (480px to 240px) and centred in its column. The v24 stretch made
     the photo fill the entire right column top to bottom. Fixed 4/5 ratio now,
     centred on both axes. */
  align-self: center;
  max-width: 240px; width: 100%;
  aspect-ratio: 4/5;
  margin-left: auto; margin-right: auto;
  border-radius: 18px; overflow: hidden;
  background: radial-gradient(ellipse at center, #1f2d4a 0%, #0a1426 90%); /* v21: matches photo backdrop */
  box-shadow: 0 30px 80px rgba(0,0,0,.4), 0 6px 16px rgba(0,0,0,.2);
}
.ep-hero-portrait img { position: absolute; inset: 0; width: 100%; height: 100%; object-fit: cover; object-position: center top; display: block; } /* v24: cover so new (matching-ratio) photo fills edge-to-edge */
.ep-hero-portrait-fallback {
  position: absolute; inset: 0;
  display: flex; flex-direction: column; align-items: center; justify-content: center;
  color: rgba(255,255,255,.45);
  font-family: ui-monospace, "SF Mono", Menlo, monospace;
  font-size: .82rem; letter-spacing: .04em; text-align: center; padding: 20px;
}
.ep-hero-portrait-fallback::before {
  content: ""; display: block; width: 96px; height: 96px;
  border-radius: 50%; background: rgba(255,255,255,.06);
  border: 2px dashed rgba(255,255,255,.18);
  margin-bottom: 18px;
}
.ep-hero-portrait-tag {
  /* v28: scaled down to fit the halved portrait */
  position: absolute; bottom: 10px; left: 10px; right: 10px;
  text-align: center;
  background: var(--ep-dark);
  color: #fff;
  color: #fff; font-size: .64rem; font-weight: 700; line-height: 1.35;
  padding: 6px 9px; border-radius: 14px;
  border: 1px solid rgba(255,255,255,.12);
}
.ep-hero-portrait-tag b { color: #fff; font-weight: 800; }
.ep-hero-cta-row { display: flex; flex-wrap: wrap; gap: 14px; margin: 28px 0 18px; align-items: flex-start; }
.ep-hero-cta-row .ep-btn { padding: 18px 28px; font-size: 1.05rem; }
/* v12: wrap orange CTA + "Having trouble?" link together so the link
   horizontally centers on the orange button only, not the whole row. */
.ep-hero-cta-col { display: flex; flex-direction: column; align-items: stretch; }
.ep-hero-cta-col .ep-hero-trouble { width: 100%; text-align: center; }
.ep-hero-trust {
  font-size: .92rem; color: rgba(255,255,255,.72);
  display: flex; flex-wrap: wrap; align-items: center; gap: 2px 6px;
  margin-top: 1px;
}
.ep-hero-trust b { color: #fff; font-weight: 700; }
.ep-hero-trust .ep-stars { color: var(--ep-orange); letter-spacing: .04em; }
.ep-hero-trust .ep-dot { color: rgba(255,255,255,.3); }

@media (max-width: 860px) {
  /* v9 fix: H1/subhead/CTA stack ABOVE portrait on mobile + small tablet.
     Copy block (.ep-hero-copy) is first in DOM; portrait (.ep-hero-portrait) is second.
     order:0 on copy + order:1 on portrait keeps source order on grid items below 860px. */
  .ep-hero-grid {
    grid-template-columns: 1fr;
    grid-template-areas:
      "eyebrow"
      "copy"
      "portrait";
    gap: 28px; padding: 44px 0;
  }
  .ep-hero-copy { order: 0; }
  .ep-hero-portrait { max-width: 220px; margin: 0 auto; aspect-ratio: 4/5; order: 1; }
  .ep-hero-cta-row .ep-btn { width: 100%; max-width: 420px; margin: 0 auto; }
  .ep-hero-cta-row { flex-direction: column; align-items: stretch; }
  /* v12: on mobile, the orange CTA + trouble link column matches the
     green button's width so they stack cleanly and the trouble link
     stays directly under the orange CTA. */
  .ep-hero-cta-col { width: 100%; max-width: 420px; margin: 0 auto; }
  .ep-hero-trust { justify-content: center; text-align: center; }
}

/* Hero bg tweak: light variant */
[data-hero-bg="light"] #ep-hero { background: #fff; color: var(--ep-dark); }
[data-hero-bg="light"] #ep-hero .ep-h1-light { color: var(--ep-dark); }
[data-hero-bg="light"] #ep-hero .ep-lead-light { color: #444; }
[data-hero-bg="light"] #ep-hero .ep-hero-trust { color: #555; }
[data-hero-bg="light"] #ep-hero .ep-hero-trust b { color: var(--ep-dark); }
[data-hero-bg="light"] #ep-hero .ep-hero-trust .ep-dot { color: rgba(0,0,0,.25); }
[data-hero-bg="light"] #ep-hero .ep-hero-microcopy { color: rgba(0,0,0,.55) !important; }
[data-hero-bg="light"] #ep-hero .ep-hero-portrait-tag { background: var(--ep-dark); color: #fff; border-color: rgba(255,255,255,.12); }
[data-hero-bg="light"] #ep-hero .ep-hero-portrait-tag b { color: #fff; }

/* Tweak: hero layout=stacked */
[data-hero-layout="stacked"] .ep-hero-grid { grid-template-columns: 1fr; max-width: 720px; margin: 0 auto; text-align: center; }
[data-hero-layout="stacked"] .ep-hero-copy { order: 0; }
[data-hero-layout="stacked"] .ep-hero-portrait { max-width: 240px; margin: 0 auto; order: 1; }
[data-hero-layout="stacked"] .ep-hero-cta-row { justify-content: center; }
[data-hero-layout="stacked"] .ep-hero-trust { justify-content: center; }

/* ─── Big idea band (#ep-bigidea) ──────────────────────────── */
#ep-bigidea { background: #fff; }
.ep-bigidea-intro { text-align: center; max-width: 760px; margin: 0 auto 48px; }
.ep-bigidea-method-tag {
  display: inline-block;
  background: #eafbea; color: #1d5c1d;
  font-size: .78rem; font-weight: 800;
  padding: 5px 12px; border-radius: 999px;
  margin-bottom: 14px;
  border: 1px solid #c8e3a3;
  letter-spacing: .02em;
}
.ep-pillars {
  display: grid; grid-template-columns: repeat(3, 1fr);
  gap: clamp(20px, 2.5vw, 32px);
}
.ep-pillar {
  background: var(--ep-bg);
  border: 1px solid var(--ep-border);
  border-radius: 16px;
  padding: clamp(24px, 3vw, 32px);
  display: flex; flex-direction: column; gap: 14px;
}
.ep-pillar-icon {
  width: 52px; height: 52px; border-radius: 14px;
  background: #fff; border: 1px solid var(--ep-border);
  display: inline-flex; align-items: center; justify-content: center;
  flex-shrink: 0;
}
.ep-pillar-icon svg { width: 28px; height: 28px; }
.ep-pillar-icon[data-tone="green"] { background: #eafbea; border-color: #c8e3a3; color: var(--ep-green); }
.ep-pillar-icon[data-tone="orange"] { background: #fff2e6; border-color: #f3c595; color: #b85f10; }
.ep-pillar-icon[data-tone="dark"] { background: var(--ep-dark); border-color: var(--ep-dark); color: var(--ep-green2); }
.ep-pillar h3 { font-size: 1.32rem; font-weight: 900; margin: 0; line-height: 1.25; }
.ep-pillar p { margin: 0; color: #333; line-height: 1.65; font-size: 1.05rem; }
.ep-pillar-mech {
  margin-top: auto;
  background: #fff; border: 1px solid var(--ep-border);
  border-radius: 10px; padding: 12px 14px;
  font-size: .85rem; line-height: 1.5;
}
.ep-pillar-mech b { display: block; color: var(--ep-dark); font-weight: 800; margin-bottom: 2px; font-size: .98rem; }
.ep-pillar-mech { font-size: .92rem; line-height: 1.55; }
.ep-pillar-mech span { color: var(--ep-muted); }

.ep-bigidea-cta { text-align: center; margin-top: 40px; }

@media (max-width: 860px) { .ep-pillars { grid-template-columns: 1fr; } }

/* Tweak: bigidea style=alternating */
[data-bigidea-style="alternating"] .ep-pillars { grid-template-columns: 1fr; gap: 18px; }
[data-bigidea-style="alternating"] .ep-pillar { display: grid; grid-template-columns: 80px 1fr; gap: 20px; align-items: start; }
[data-bigidea-style="alternating"] .ep-pillar-icon { width: 64px; height: 64px; }
[data-bigidea-style="alternating"] .ep-pillar-icon svg { width: 32px; height: 32px; }
[data-bigidea-style="alternating"] .ep-pillar-mech { grid-column: 2; }
[data-bigidea-style="alternating"] .ep-pillar h3,
[data-bigidea-style="alternating"] .ep-pillar p { grid-column: 2; }

/* ─── AudioSample (#ep-audio) ──────────────────────────────── */
#ep-audio { background: var(--ep-bg); }
/* v9 fix: tighten gap between audio sample and "The main course" (#ep-ue).
   Default .ep-sec padding is clamp(48px, 7vw, 80px) top + bottom.
   Reduce the audio section's bottom padding and the next section's top padding
   to ~half of default. Combined gap drops from ~96–160px to ~48–80px. */
#ep-audio.ep-sec { padding-bottom: clamp(22px, 3.2vw, 38px); }
#ep-ue.ep-sec    { padding-top:    clamp(22px, 3.2vw, 38px); }
.tl-audio-inner { max-width: 820px; margin: 0 auto; }
.tl-audio-intro { text-align: center; margin-bottom: 32px; }
.tl-audio-card {
  position: relative;
  background: var(--ep-dark);
  border-radius: 20px;
  padding: clamp(28px, 4vw, 40px);
  color: #fff;
  overflow: hidden;
  box-shadow: 0 24px 60px rgba(0,0,0,.18);
}
.tl-audio-card::before {
  content: none;
}
.tl-audio-tag {
  display: inline-block; position: relative;
  background: var(--ep-orange); color: #fff;
  font-size: .72rem; font-weight: 800; letter-spacing: .14em; text-transform: uppercase;
  padding: 5px 12px; border-radius: 20px;
}
.tl-audio-card h3 {
  position: relative;
  font-size: clamp(1.25rem, 2.5vw, 1.55rem);
  font-weight: 900; margin: 14px 0 6px; color: #fff; line-height: 1.25;
}
.tl-audio-meta { position: relative; font-size: .88rem; color: rgba(255,255,255,.65); margin: 0 0 22px; }
.tl-audio-player {
  position: relative;
  background: rgba(255,255,255,.06);
  border: 1px solid rgba(255,255,255,.12);
  border-radius: 14px; padding: 14px 16px;
  display: flex; align-items: center; gap: 16px;
}
.tl-audio-play {
  width: 56px; height: 56px; border-radius: 50%;
  background: var(--ep-orange); color: #fff;
  border: 0; cursor: pointer; flex-shrink: 0;
  display: inline-flex; align-items: center; justify-content: center;
  box-shadow: 0 8px 22px rgba(244,103,46,.5);
  transition: filter .15s ease, transform .1s ease;
}
.tl-audio-play:hover { filter: brightness(1.06); transform: scale(1.04); }
.tl-audio-play svg { width: 22px; height: 22px; }
/* v14: play/pause icon swap based on .playing class */
.tl-audio-play .ic-play  { display: inline-block; transform: translateX(2px); }
.tl-audio-play .ic-pause { display: none; }
.tl-audio-play.playing .ic-play  { display: none; }
.tl-audio-play.playing .ic-pause { display: inline-block; transform: translateX(0); }
.tl-audio-track { flex: 1; min-width: 0; }
.tl-audio-bar { height: 6px; background: rgba(255,255,255,.14); border-radius: 4px; overflow: hidden; }
.tl-audio-bar-fill { height: 100%; width: 0%; background: linear-gradient(90deg, var(--ep-orange), #ffc28b); border-radius: 4px; transition: width .12s linear; }
.tl-audio-time { display: flex; justify-content: space-between; font-size: .78rem; color: rgba(255,255,255,.6); margin-top: 6px; font-variant-numeric: tabular-nums; }
.tl-audio-transcript {
  position: relative;
  margin-top: 22px;
  padding: 18px 20px;
  background: rgba(255,255,255,.04);
  border: 1px solid rgba(255,255,255,.08);
  border-radius: 12px;
  font-size: .98rem; line-height: 1.7;
  color: rgba(255,255,255,.88);
}
.tl-audio-transcript b { color: #fff; font-weight: 800; background: color-mix(in srgb, var(--ep-orange) 18%, transparent); padding: 1px 5px; border-radius: 4px; }
.tl-audio-after { text-align: center; margin-top: 24px; font-size: .98rem; color: var(--ep-muted); }
.tl-audio-after a { color: var(--ep-dark); font-weight: 700; text-decoration: none; border-bottom: 2px solid var(--ep-green); padding-bottom: 1px; }
.tl-audio-after a:hover { border-bottom-color: var(--ep-green2); }

/* ─── UE feature card (#ep-ue) ─────────────────────────────── */
#ep-ue { background: #fff; }
.ep-ue-card {
  background: #fff;
  border: 1px solid var(--ep-border);
  border-radius: 20px;
  padding: clamp(28px, 4vw, 48px);
  display: grid;
  grid-template-columns: minmax(0, 1fr) minmax(0, 1.05fr);
  gap: clamp(28px, 5vw, 56px);
  align-items: center;
}
[data-ue-layout="mockup-right"] .ep-ue-card { grid-template-columns: minmax(0, 1.05fr) minmax(0, 1fr); }
[data-ue-layout="mockup-right"] .ep-ue-mockups { order: 2; }
[data-ue-layout="mockup-right"] .ep-ue-body { order: 1; }
@media (max-width: 860px) {
  /* v25 bugfix: must out-specify [data-ue-layout="mockup-right"] .ep-ue-card
     (0,2,0), which otherwise keeps the card a 2-column grid on mobile. With
     data-ue-layout hardcoded on #ep-page (v22) the plain .ep-ue-card (0,1,0)
     override lost, so the body sat in one column and the hidden mockup track
     left a blank column on the right. #ep-ue .ep-ue-card is (1,1,0) and wins. */
  #ep-ue .ep-ue-card { grid-template-columns: 1fr; gap: 28px; }
  .ep-ue-mockups { order: -1; }
}
.ep-ue-mockups {
  position: relative;
  aspect-ratio: 5 / 4;
  display: flex; align-items: center; justify-content: center;
}
.ep-ue-desktop {
  position: absolute; inset: 0 6% 18% 0;
  background: #fff; border-radius: 12px;
  border: 1px solid var(--ep-border);
  box-shadow: 0 20px 50px rgba(0,0,0,.12);
  overflow: hidden;
}
.ep-ue-desktop-bar {
  height: 26px; background: var(--ep-dark);
  display: flex; align-items: center; padding: 0 10px; gap: 6px;
}
.ep-ue-desktop-bar span { width: 9px; height: 9px; border-radius: 50%; background: rgba(255,255,255,.25); }
.ep-ue-desktop-body { padding: 14px 14px 0; }
.ep-ue-desktop-body .ep-ue-row { height: 8px; background: var(--ep-bg); border-radius: 4px; margin-bottom: 6px; }
.ep-ue-desktop-body .ep-ue-row:nth-child(1) { width: 40%; height: 14px; background: var(--ep-dark); margin-bottom: 12px; }
.ep-ue-desktop-body .ep-ue-row:nth-child(2) { width: 75%; }
.ep-ue-desktop-body .ep-ue-row:nth-child(3) { width: 92%; }
.ep-ue-desktop-body .ep-ue-row:nth-child(4) { width: 86%; }
.ep-ue-desktop-body .ep-ue-row:nth-child(5) { width: 60%; }
.ep-ue-desktop-player {
  margin-top: 14px; padding: 10px 12px;
  background: var(--ep-dark); border-radius: 8px;
  display: flex; align-items: center; gap: 10px;
}
.ep-ue-desktop-player::before {
  content: ""; width: 22px; height: 22px; border-radius: 50%;
  background: var(--ep-orange); flex-shrink: 0;
}
.ep-ue-desktop-player::after {
  content: ""; flex: 1; height: 4px; background: rgba(255,255,255,.16); border-radius: 2px;
  position: relative;
}
.ep-ue-phone {
  position: absolute; right: 0; bottom: 0;
  width: 38%; aspect-ratio: 9 / 18;
  background: #1a1a1a;
  border-radius: 22px;
  border: 8px solid #1a1a1a;
  box-shadow: 0 22px 50px rgba(0,0,0,.32), 0 0 0 1px rgba(0,0,0,.5);
  overflow: hidden;
  transform: rotate(4deg);
}
.ep-ue-phone-screen {
  width: 100%; height: 100%;
  background: var(--ep-dark);
  display: flex; flex-direction: column;
  padding: 8px 8px 12px;
  position: relative;
}
.ep-ue-phone-screen::before {
  content: ""; position: absolute;
  top: 0; left: 50%; transform: translateX(-50%);
  width: 36%; height: 14px;
  background: #000; border-radius: 0 0 10px 10px;
}
.ep-ue-phone-tag {
  margin-top: 18px;
  background: var(--ep-orange); color: #fff;
  font-size: .42rem; font-weight: 800; letter-spacing: .12em;
  padding: 2px 6px; border-radius: 10px; align-self: flex-start;
  text-transform: uppercase;
}
.ep-ue-phone-title {
  color: #fff; font-size: .58rem; font-weight: 800;
  margin-top: 6px; line-height: 1.2;
}
.ep-ue-phone-meta { color: rgba(255,255,255,.5); font-size: .42rem; margin-top: 2px; }
.ep-ue-phone-trans {
  flex: 1; margin-top: 8px;
  background: rgba(255,255,255,.06); border-radius: 4px;
  padding: 5px;
}
.ep-ue-phone-trans span {
  display: block; height: 2px; background: rgba(255,255,255,.18);
  border-radius: 1px; margin-bottom: 3px;
}
.ep-ue-phone-trans span:nth-child(2) { width: 80%; }
.ep-ue-phone-trans span:nth-child(3) { width: 92%; }
.ep-ue-phone-trans span:nth-child(4) { width: 70%; }
.ep-ue-phone-trans span:nth-child(5) { width: 88%; background: rgba(244,103,46,.6); }
.ep-ue-phone-trans span:nth-child(6) { width: 78%; }
.ep-ue-phone-trans span:nth-child(7) { width: 60%; }
.ep-ue-phone-player {
  margin-top: 4px; height: 16px;
  background: rgba(0,0,0,.3); border-radius: 4px;
  display: flex; align-items: center; padding: 0 5px; gap: 4px;
}
.ep-ue-phone-player::before {
  content: ""; width: 9px; height: 9px; border-radius: 50%; background: var(--ep-orange);
}
.ep-ue-phone-player::after {
  content: ""; flex: 1; height: 2px; background: rgba(255,255,255,.18); border-radius: 1px;
}
.ep-ue-mockup-label {
  position: absolute; bottom: -2px; left: 0;
  font-family: ui-monospace, "SF Mono", Menlo, monospace;
  font-size: .68rem; color: var(--ep-muted);
  letter-spacing: .04em;
}

.ep-ue-bullets { list-style: none; margin: 18px 0; padding: 0; display: flex; flex-direction: column; gap: 14px; }
.ep-ue-bullets li {
  display: grid; grid-template-columns: 24px 1fr; gap: 12px; align-items: start;
  font-size: 1rem; line-height: 1.55; color: #1d2e2e;
}
.ep-ue-bullets li::before {
  content: ""; width: 22px; height: 22px; border-radius: 50%;
  background: #eafbea; border: 1px solid #c8e3a3;
  position: relative; margin-top: 1px;
}
.ep-ue-bullets li svg { display: none; }
.ep-ue-bullets li b { font-weight: 700; }
.ep-ue-check { position: relative; }
.ep-ue-check::after {
  content: "✓"; color: var(--ep-green); font-weight: 900; font-size: .9rem;
  position: absolute; left: 6px; top: -1px;
}
.ep-ue-price { font-size: 1.6rem; font-weight: 900; color: var(--ep-dark); }
.ep-ue-price small { font-size: .82rem; font-weight: 600; color: var(--ep-muted); }
.ep-ue-price-or { color: var(--ep-muted); }
.ep-ue-cta-row { display: flex; flex-wrap: wrap; gap: 14px 22px; align-items: center; margin-top: 8px; }
.ep-ue-cta-row .ep-btn-green { white-space: nowrap; }

/* v16: UE feature card — split H2 + dedicated lead + dedicated price line.
   Mobile fixes: shrink H2, block-stack bullet bold lead-in, hide mockups,
   tighten card padding. */
.ep-ue-lead { font-size: 1.15rem; line-height: 1.5; color: var(--ep-muted); margin: 4px 0 14px; }
.ep-ue-price-line { font-size: 1.05rem; color: var(--ep-muted); margin: 0 0 20px; line-height: 1.3; }
.ep-ue-price-line b { color: var(--ep-green); font-size: 2rem; font-weight: 900; letter-spacing: -.01em; }
.ep-ue-price-line .ep-ue-price-per { font-size: 1rem; color: var(--ep-muted); font-weight: 600; letter-spacing: 0; }
.ep-ue-price-line .ep-ue-price-billed { font-size: 1rem; color: var(--ep-muted); font-weight: 600; letter-spacing: 0; white-space: nowrap; } /* v26: billing-total disclosure — kept at the same size/weight/color as the /mo text so the annual total is at least as legible as the per-month figure (FTC/CA-ARL prominence rule). */
@media (max-width: 640px) {
  #ep-ue .ep-h2, #ep-ue .ep-ue-lead { text-wrap: auto; }
  #ep-ue .ep-h2 { font-size: 1.55rem; line-height: 1.15; }
  .ep-ue-lead { font-size: 1rem; margin: 4px 0 10px; }
  .ep-ue-price-line b { font-size: 1.7rem; }
  .ep-ue-bullets li b { display: block; margin-bottom: 2px; }
  .ep-ue-bullets li span { display: block; width: 100%; }
  .ep-ue-mockups { display: none; }
  .ep-ue-card { padding: 22px 18px; }
  /* v24: UE Main Course card on mobile. Fully centered + symmetrical. Heading,
     lead, and price line center as text. Bullets switch from the grid row
     (checkmark-left + text-right) to a centered vertical stack: the green
     check badge sits centered ON TOP of each bullet, with the bold heading and
     description centered below it. This avoids the left-anchor imbalance of v22
     (center-around-checkmark) and v23 (centered block) — a multi-line bullet
     list can't be visually centered while staying a left-aligned row, so the
     check moves above the text instead. The ✓ glyph (li::after, normally
     absolute at left:6px over the left circle) is re-centered over the now-
     centered circle via left:50% + translateX(-50%). */
  #ep-ue .ep-h2,
  #ep-ue .ep-ue-lead,
  #ep-ue .ep-ue-price-line { text-align: center; }
  #ep-ue .ep-ue-bullets { width: auto; max-width: 100%; margin: 18px 0; }
  #ep-ue .ep-ue-bullets li {
    display: flex; flex-direction: column; align-items: center;
    text-align: center; gap: 4px;
  }
  #ep-ue .ep-ue-bullets li::before { margin: 0 0 4px; }
  #ep-ue .ep-ue-bullets li.ep-ue-check::after {
    left: 50%; top: 0; transform: translateX(-50%);
    height: 22px; line-height: 22px;
  }
  #ep-ue .ep-ue-bullets li span { width: auto; max-width: 100%; }
}

/* ─── Free-course capture (#ep-magnet) ─────────────────────── */
/* v17 Option A: lighter dark shelf (warm charcoal) + floating WHITE card with
   strong shadow. White card on dark = maximum contrast and the strongest CTA
   pattern for capture sections. Card text reverts to standard dark on white. */
#ep-magnet { background: var(--ep-dark); min-height: 640px; display: flex; align-items: center; }
#ep-magnet .ep-wrap { width: 100%; }
.ep-magnet-card {
  max-width: 720px; margin: 0 auto;
  background: #fff;
  border: 1px solid var(--ep-border);
  border-radius: 20px;
  padding: clamp(36px, 5vw, 56px);
  text-align: center;
  box-shadow: 0 30px 60px rgba(0,0,0,.35), 0 8px 20px rgba(0,0,0,.18);
}
.ep-magnet-card .ep-h2 { margin-bottom: 8px; }
.ep-magnet-card .ep-eyebrow { margin: 0 0 32px; }
.ep-magnet-card .ep-h2 { margin-top: 0; }
.ep-magnet-card .ep-magnet-sub { font-weight: 600; color: var(--ep-dark); font-size: 1.02rem; margin: 0 0 28px; }
.ep-magnet-card .ep-magnet-body { color: #444; font-size: .98rem; line-height: 1.65; max-width: 520px; margin: 0 auto 26px; }
.ep-magnet-card .popup-trigger {
  display: inline-flex; align-items: center; gap: .5rem;
  background: var(--ep-orange); color: #fff;
  border: 0; border-radius: 999px;
  padding: 18px 32px; font-weight: 800; font-size: 1.08rem;
  cursor: pointer; font-family: inherit;
  box-shadow: 0 10px 26px color-mix(in srgb, var(--ep-orange) 35%, transparent);
  transition: filter .15s ease, transform .1s ease;
}
.ep-magnet-card .popup-trigger:hover { filter: brightness(.94); transform: translateY(-2px); }
/* v50: equal optical spacing. sub->button and button->microcopy are both 28px,
   and nothing carries a bottom margin, so the card's own padding sets the top
   and bottom gaps equally and the block centres in the white box. */
.ep-magnet-microcopy { font-size: 1.05rem; font-weight: 700; color: var(--ep-dark); margin: 28px 0 0; }

/* ─── Other products grid (#ep-products) ───────────────────── */
#ep-products { background: #fff; }
.ep-products-head { text-align: center; margin-bottom: 36px; }
.ep-products-grid {
  display: grid; grid-template-columns: repeat(3, 1fr);
  gap: clamp(18px, 2.5vw, 28px);
}
@media (max-width: 860px) { .ep-products-grid { grid-template-columns: 1fr; } }
.ep-product {
  background: var(--ep-bg);
  border: 1px solid var(--ep-border);
  border-radius: 16px;
  padding: 28px;
  display: flex; flex-direction: column; gap: 12px;
  text-decoration: none; color: var(--ep-dark);
  transition: transform .15s ease, box-shadow .15s ease, border-color .15s ease;
}
.ep-product:hover { transform: translateY(-3px); box-shadow: 0 14px 34px rgba(0,0,0,.08); border-color: var(--ep-green2); }
.ep-product-icon {
  width: 48px; height: 48px; border-radius: 12px;
  background: #fff; border: 1px solid var(--ep-border);
  display: inline-flex; align-items: center; justify-content: center;
}
.ep-product h3 { font-size: 1.08rem; font-weight: 800; margin: 0; }
.ep-product p { font-size: .94rem; line-height: 1.55; color: #444; margin: 0; flex: 1; }
.ep-product-link {
  font-size: .9rem; font-weight: 800; color: var(--ep-dark);
  display: inline-flex; align-items: center; gap: 6px;
  border-bottom: 2px solid var(--ep-green);
  padding-bottom: 1px; align-self: flex-start;
}

/* ─── Founder credibility (#ep-founder) ────────────────────── */
/* v13: switched from dark to cream bg (HomePage-Analysis visual rec #1) */
#ep-founder { background: var(--ep-bg); color: var(--ep-dark); }
.ep-founder-grid {
  display: grid; grid-template-columns: minmax(0, 380px) 1fr;
  gap: clamp(32px, 5vw, 64px);
  align-items: center;
}
@media (max-width: 860px) {
  .ep-founder-grid { grid-template-columns: 1fr; gap: 28px; }
  .ep-founder-portrait { max-width: 320px; margin: 0 auto; }
}
.ep-founder-portrait {
  position: relative;
  aspect-ratio: 4 / 5;
  border-radius: 18px; overflow: hidden;
  background: linear-gradient(155deg, #2b3636 0%, #0f1a1a 100%);
  border: 1px solid rgba(255,255,255,.08);
  box-shadow: 0 22px 60px rgba(0,0,0,.32);
}
.ep-founder-portrait img { width: 100%; height: 100%; object-fit: cover; display: block; }
.ep-founder-portrait-fallback {
  position: absolute; inset: 0;
  display: flex; flex-direction: column; align-items: center; justify-content: center;
  color: rgba(255,255,255,.4);
  font-family: ui-monospace, "SF Mono", Menlo, monospace;
  font-size: .8rem; letter-spacing: .04em; text-align: center; padding: 20px;
}
.ep-founder-portrait-fallback::before {
  content: ""; display: block; width: 88px; height: 88px;
  border-radius: 50%; background: rgba(255,255,255,.06);
  border: 2px dashed rgba(255,255,255,.18);
  margin-bottom: 16px;
}
.ep-founder-body p { color: #333; font-size: 1.02rem; line-height: 1.7; margin: 0 0 1em; }
.ep-founder-body p:last-of-type { margin-bottom: 22px; }
.ep-founder-credlist {
  list-style: none; padding: 0; margin: 18px 0 24px;
  display: flex; flex-direction: column; gap: 10px;
}
.ep-founder-credlist li {
  display: grid; grid-template-columns: 22px 1fr; gap: 10px; align-items: start;
  font-size: .96rem; line-height: 1.55; color: #333;
}
.ep-founder-credlist li::before {
  content: ""; width: 8px; height: 8px; background: var(--ep-green2); border-radius: 50%;
  margin-top: 8px; margin-left: 7px;
}

/* ─── Testimonials (#ep-celebrate) ─────────────────────────── */
/* v14: warm sand/beige bg breaks identical-cream stack with founder section above */
#ep-celebrate { background: color-mix(in srgb, var(--ep-green) 6%, #f6f5f9); }
.ep-celebrate-head { text-align: center; margin-bottom: 32px; }
.ep-stats {
  display: flex; flex-wrap: wrap; justify-content: center; gap: 14px 38px;
  margin-bottom: 36px;
}
.ep-stat { text-align: center; }
.ep-stat b { display: block; font-size: clamp(1.5rem, 3vw, 2rem); font-weight: 900; color: var(--ep-dark); line-height: 1; }
.ep-stat span { display: block; font-size: .82rem; color: var(--ep-muted); margin-top: 6px; letter-spacing: .04em; }

/* v64: stat 4 nested a <span> inside its <b> to carry the number. The
   .ep-stat span rule directly above therefore hit that number too and made it a
   small, muted, BLOCK element — wrong size, wrong colour, and an extra line that
   pushed the whole fourth column below the other three. The number is now a
   plain <b>, exactly like stats 1-3, so it aligns and inherits the green.
   The unit and the suffix moved into the label; these two rules stop the nested
   spans taking the block/margin treatment a second time, and zero the sup's
   line-height so the ® cannot stretch the label's line box. */
.ep-stat .ep-stat-yrs span { display: block; font-size: inherit; color: inherit; margin-top: 0; letter-spacing: inherit; }
.ep-stat .ep-stat-yrs sup { font-size: .7em; line-height: 0; vertical-align: super; }

.ep-carousel { position: relative; }
.ep-carousel-track {
  display: grid; grid-auto-flow: column;
  grid-auto-columns: minmax(280px, 320px);
  gap: 16px; padding: 4px 4px 14px;
  overflow-x: auto;
  scroll-snap-type: x mandatory;
  -webkit-overflow-scrolling: touch;
}
.ep-carousel-track::-webkit-scrollbar { height: 6px; }
.ep-carousel-track::-webkit-scrollbar-thumb { background: rgba(0,0,0,.16); border-radius: 3px; }
.ep-tcard {
  background: #fff;
  border: 1px solid var(--ep-border);
  border-radius: 14px;
  padding: 22px;
  scroll-snap-align: start;
  display: flex; flex-direction: column; gap: 12px;
}
.ep-tcard-quote {
  font-family: Georgia, "Times New Roman", serif;
  font-style: italic; font-size: 1rem; line-height: 1.6;
  color: var(--ep-dark); margin: 0; flex: 1;
}
.ep-tcard-stars { color: var(--ep-orange); font-size: .9rem; letter-spacing: .04em; }
.ep-tcard-meta {
  font-size: .82rem; color: var(--ep-muted); font-weight: 600;
  display: flex; align-items: center; gap: 8px; flex-wrap: wrap;
}
.ep-tcard-meta b { color: var(--ep-dark); font-weight: 800; }
.ep-tcard-flag { font-size: 1.05rem; }
/* v55: the "· " separator moved out of the translatable location string
   and into generated content, so no locale carries a leading separator.
   white-space:pre keeps the trailing space the flex gap would otherwise eat. */
.ep-tcard-loc::before { content: "· "; white-space: pre; }
.ep-tcard.ep-tcard-placeholder .ep-tcard-quote { color: #888; }

.ep-carousel-controls {
  display: flex; justify-content: center; align-items: center; gap: 16px;
  margin-top: 14px;
}
.ep-carousel-btn {
  width: 40px; height: 40px; border-radius: 50%;
  background: #fff; border: 1px solid var(--ep-border);
  color: var(--ep-dark); cursor: pointer;
  display: inline-flex; align-items: center; justify-content: center;
  font-size: 1.1rem;
  transition: background .15s ease, border-color .15s ease;
}
.ep-carousel-btn:hover { background: var(--ep-bg); border-color: var(--ep-green); }
.ep-carousel-dots { display: flex; gap: 6px; }
.ep-carousel-dots span { width: 7px; height: 7px; border-radius: 50%; background: rgba(0,0,0,.18); }
.ep-carousel-dots span.active { background: var(--ep-green); width: 22px; border-radius: 4px; }

/* ─── FAQ (#ep-faq) ────────────────────────────────────────── */
#ep-faq { background: #fff; }
.ep-faq-list { max-width: 780px; margin: 0 auto; }
.ep-faq-item {
  border-top: 1px solid var(--ep-border);
  padding: 4px 0;
}
.ep-faq-item:last-child { border-bottom: 1px solid var(--ep-border); }
.ep-faq-item summary {
  list-style: none; cursor: pointer;
  padding: 22px 36px 22px 0;
  font-size: 1.05rem; font-weight: 700; color: var(--ep-dark);
  position: relative;
  line-height: 1.4;
}
.ep-faq-item summary::-webkit-details-marker { display: none; }
.ep-faq-item summary::after {
  content: "+"; position: absolute; right: 0; top: 50%; transform: translateY(-50%);
  font-size: 1.5rem; font-weight: 400; color: var(--ep-green); line-height: 1;
  transition: transform .15s ease;
}
.ep-faq-item[open] summary::after { content: "−"; }
.ep-faq-item[open] summary { color: var(--ep-darker); }
.ep-faq-item-body {
  padding: 0 0 22px;
  color: #444; line-height: 1.65; font-size: 1rem; max-width: 720px;
}
.ep-faq-cta { text-align: center; margin-top: 32px; }

/* ─── Footer (#ep-footer) ──────────────────────────────────── */
#ep-footer {
  background: var(--ep-darker);
  color: rgba(255,255,255,.72);
  padding: 64px 0 32px;
}
/* v10 fix: footer = 4 columns on desktop (Brand / Product / Learn / Connect). */
.ep-footer-grid {
  display: grid;
  grid-template-columns: 1.4fr repeat(3, 1fr);
  gap: clamp(24px, 3vw, 40px);
  margin-bottom: 48px;
}
@media (max-width: 960px) {
  .ep-footer-grid { grid-template-columns: 1fr 1fr; }
  .ep-footer-brand { grid-column: 1 / -1; }
}
@media (max-width: 540px) {
  .ep-footer-grid { grid-template-columns: 1fr; }
}
.ep-footer-brand { font-size: .9rem; line-height: 1.65; }
.ep-footer-brand .ep-logo svg { height: 30px; }
.ep-footer-brand p { margin: 14px 0; }
.ep-footer-brand .ep-footer-tagline { color: #fff; font-weight: 600; font-size: .98rem; }
.ep-footer-brand small { display: block; color: rgba(255,255,255,.54); font-size: .82rem; margin-top: 8px; line-height: 1.55; }
.ep-footer-col h4 {
  font-size: .82rem; font-weight: 800;
  text-transform: uppercase; letter-spacing: .08em;
  color: var(--ep-green2); margin: 0 0 16px;
}
.ep-footer-col ul { list-style: none; margin: 0; padding: 0; display: flex; flex-direction: column; gap: 10px; }
.ep-footer-col a, .ep-footer-col .lm-link {
  color: rgba(255,255,255,.72); text-decoration: none; font-size: .92rem;
  background: transparent; border: 0; padding: 0; font-family: inherit; cursor: pointer;
  text-align: left;
}
.ep-footer-col a:hover, .ep-footer-col .lm-link:hover { color: #fff; }

.ep-footer-trust {
  border-top: 1px solid rgba(255,255,255,.12);
  border-bottom: 1px solid rgba(255,255,255,.12);
  padding: 22px 0;
  display: flex; flex-wrap: wrap; gap: 14px 36px; justify-content: center;
  font-size: .9rem; color: rgba(255,255,255,.66);
}
.ep-footer-trust span { display: inline-flex; align-items: center; gap: 8px; }
.ep-footer-legal {
  display: flex; flex-wrap: wrap; align-items: center; justify-content: space-between;
  gap: 14px;
  padding-top: 24px;
  font-size: .82rem; color: rgba(255,255,255,.5);
}
.ep-footer-legal-links { display: flex; flex-wrap: wrap; gap: 14px; }
.ep-footer-legal a { color: rgba(255,255,255,.5); text-decoration: none; }
.ep-footer-legal a:hover { color: #fff; }

/* ─── Curriculum / What is inside UE (#ep-curriculum) ─── */
#ep-curriculum { background: var(--ep-bg); padding-top: clamp(28px, 4vw, 56px); padding-bottom: clamp(28px, 4vw, 56px); }
.ep-curriculum-intro { text-align: center; max-width: 760px; margin: 0 auto 40px; }
.ep-curriculum-grid { display: grid; grid-template-columns: 1fr 1fr; gap: 48px; max-width: 1080px; margin: 0 auto; }
.ep-curriculum-coltitle { font-size: 1.35rem; font-weight: 900; color: var(--ep-dark); margin: 0 0 4px; letter-spacing: -.01em; }
.ep-curriculum-colmeta { color: var(--ep-muted); font-size: .95rem; margin: 0 0 22px; font-weight: 600; }
.ep-curriculum-cats { list-style: none; margin: 0; padding: 0; display: flex; flex-direction: column; gap: 12px; }
.ep-curriculum-cats li { display: block; padding: 12px 14px; background: #fff; border-radius: 10px; border: 1px solid var(--ep-border); }
.ep-curriculum-cats li b { color: var(--ep-dark); font-weight: 800; font-size: 1.02rem; }
.ep-cat-count { color: var(--ep-green); font-weight: 700; font-size: .9rem; margin-left: 6px; }
.ep-cat-ex { display: block; color: #444; font-size: .9rem; line-height: 1.5; margin-top: 4px; }
.ep-curriculum-foot { text-align: center; color: var(--ep-muted); font-size: 1rem; margin-top: 36px; font-weight: 600; }
@media (max-width: 860px) {
  .ep-curriculum-grid { grid-template-columns: 1fr; gap: 32px; }
}
.ep-curriculum-more { margin: 14px 2px 0; font-size: .95rem; }
.ep-curriculum-more a { color: var(--ep-green); font-weight: 700; text-decoration: none; }
.ep-curriculum-more a:hover { text-decoration: underline; }

/* ─── How it works strip (#ep-howitworks) ─── */
#ep-howitworks { background: var(--ep-bg); padding-top: clamp(22px, 3vw, 38px); padding-bottom: clamp(22px, 3vw, 38px); }
.ep-howitworks-intro { margin: 0 auto 22px; }
.ep-howitworks-steps { list-style: none; padding: 0; display: grid; grid-template-columns: repeat(3, 1fr); gap: 18px; max-width: 980px; margin: 0 auto; }
.ep-howitworks-steps li { display: flex; flex-direction: column; align-items: center; text-align: center; padding: 22px 18px; background: #fff; border: 1px solid var(--ep-border); border-radius: 12px; }
.ep-howitworks-num { display: flex; align-items: center; justify-content: center; width: 36px; height: 36px; border-radius: 50%; background: var(--ep-green); color: #fff; font-weight: 900; font-size: 1.1rem; margin-bottom: 12px; }
.ep-howitworks-text { color: var(--ep-dark); font-size: 1.02rem; font-weight: 700; line-height: 1.35; }
@media (max-width: 720px) {
  .ep-howitworks-steps { grid-template-columns: 1fr; gap: 12px; }
}

/* ─── Sticky bottom mobile CTA ─────────────────────────────── */
.ep-mobile-stickycta {
  display: none;
  position: fixed; bottom: 0; left: 0; right: 0;
  background: #fff; border-top: 1px solid var(--ep-border);
  padding: 10px 16px calc(10px + env(safe-area-inset-bottom));
  z-index: 80;
  box-shadow: 0 -8px 24px rgba(0,0,0,.08);
  opacity: 0; pointer-events: none; transform: translateY(8px);
  transition: opacity .2s ease, transform .2s ease;
}
.ep-mobile-stickycta.ep-sticky-revealed { opacity: 1; pointer-events: auto; transform: none; }
.ep-mobile-stickycta a {
  display: block; width: 100%;
  background: var(--ep-green); color: #fff !important;
  font-weight: 800; font-size: 1.02rem; text-align: center;
  padding: 16px 20px; border-radius: 999px;
  text-decoration: none; box-shadow: 0 6px 16px rgba(21,155,151,.32);
}
[data-mobile-cta="orange"] .ep-mobile-stickycta a { background: var(--ep-orange); box-shadow: 0 6px 16px rgba(244,103,46,.32); }
[data-mobile-cta="off"] .ep-mobile-stickycta { display: none !important; }
@media (max-width: 720px) {
  .ep-mobile-stickycta { display: block; }
  body { padding-bottom: 108px; }
}

/* ─── Tweak: hero CTA color swap ──────────────────────────── */
[data-hero-cta="green-primary"] .ep-hero-cta-row .ep-btn:first-child { background: var(--ep-green); box-shadow: 0 8px 24px rgba(21,155,151,.32); }
[data-hero-cta="green-primary"] .ep-hero-cta-row .ep-btn-green { background: var(--ep-orange) !important; box-shadow: 0 8px 24px rgba(244,103,46,.35); }

/* ─── Tweak: testimonial style=grid ──────────────────────── */
[data-test-style="grid"] .ep-carousel-track { display: grid; grid-template-columns: repeat(3, 1fr); grid-auto-flow: row; grid-auto-columns: auto; overflow: visible; padding: 4px; gap: 20px; }
[data-test-style="grid"] .ep-carousel-controls { display: none; }
[data-test-style="grid"] .ep-tcard { scroll-snap-align: none; }
@media (max-width: 980px) {
  [data-test-style="grid"] .ep-carousel-track { grid-template-columns: repeat(2, 1fr); }
}
@media (max-width: 640px) {
  [data-test-style="grid"] .ep-carousel-track { grid-template-columns: 1fr; }
}
@media (max-width: 860px) {
  [data-test-style="grid"] .ep-carousel-track { grid-template-columns: 1fr; }
}

/* Prevent drawer scroll lock issues */
html.nav-locked, body.nav-locked { overflow: hidden; }

/* Reduced motion */
@media (prefers-reduced-motion: reduce) { * { animation-duration: 0s !important; transition-duration: 0s !important; } }

.tl-audio-line { font-style: italic; color: rgba(255,255,255,.72); margin: 0 0 .8em; font-size: .98rem; text-align: left; }
.tl-audio-quote { border-left: 3px solid var(--ep-green); padding: 4px 16px; margin: 0 0 1em; }
.tl-audio-quote p { margin: 0; color: #fff; }
.tl-audio-quote strong { color: #fff; font-weight: 700; }
.tl-audio-divider { border: 0; border-top: 1px solid rgba(255,255,255,.12); margin: 18px 0 18px; }
.tl-audio-toggle { margin: 0 0 1em; }
.tl-audio-toggle > summary { list-style: none; cursor: pointer; padding: 8px 14px; border: 1px solid rgba(255,255,255,.18); border-radius: 6px; display: inline-block; color: rgba(255,255,255,.9); font-weight: 600; font-size: .94rem; user-select: none; }
.tl-audio-toggle > summary::-webkit-details-marker { display: none; }
.tl-audio-toggle .tl-audio-hide { display: none; }
.tl-audio-toggle[open] .tl-audio-show { display: none; }
.tl-audio-toggle[open] .tl-audio-hide { display: inline; }
/* v15 Path A: proof bar on dark — stats and logos sit on var(--ep-dark) so they
   read as a trust strip continuing from the hero. Stats numbers go white,
   labels muted-white, logos inverted to white with reduced opacity. */
/* v31: each half of the H1 wraps as a unit, so the line always breaks at the
   bullet instead of wherever the text happens to run out. */
.ep-h1-part { display: block; }
/* v32: with the bullet gone the halves are shorter, so the hero H1 is back on
   the global 4.6vw curve. Each half is its own block, so the break is
   structural rather than dependent on where the text runs out. */
#ep-hero .ep-h1 { font-size: clamp(2rem, 4.6vw, 3rem); }

#ep-press { background: var(--ep-dark); padding: 72px 0; color: #fff; } /* v28: 36px to 72px top and bottom; the gap between the stats row and the logo row is deliberately untouched */
#ep-press .ep-stat b { color: var(--ep-green2); } /* v28: white to brand green */
#ep-press .ep-stat span { color: rgba(255,255,255,.62); }
.ep-press-divider { max-width: 480px; margin: 4px auto 20px; border: 0; border-top: 1px solid rgba(255,255,255,.12); }
.ep-press-eyebrow { text-align: center; font-size: .78rem; font-weight: 800; letter-spacing: .18em; text-transform: uppercase; color: rgba(255,255,255,.55); margin: 0 0 16px; }
.ep-press-row { display: flex; flex-wrap: wrap; align-items: center; justify-content: center; gap: clamp(18px, 3.4vw, 40px); }
/* v16: removed brightness(0) invert(1) — was killing brand color recognition
   on the press logos. On dark bg, color logos at .85 opacity read fine and
   preserve the trust signal of recognizable brand marks. */
.ep-press-row img { height: 22px; width: auto; max-width: 130px; object-fit: contain; opacity: .75; transition: opacity .15s ease; }
.ep-press-row img:hover { opacity: 1; }
@media (max-width: 700px) {
  .ep-press-row { gap: 14px 22px; }
  .ep-press-row img { height: 18px; max-width: 95px; }
}
/* UE feature card screenshots */
.ep-ue-desktop img, .ep-ue-phone img { width: 100%; height: 100%; object-fit: cover; display: block; }
.ep-ue-desktop { padding: 0 !important; }
.ep-ue-phone { padding: 0 !important; }
/* Carousel auto-rotate hint */
.ep-carousel-track.is-paused { /* visual cue not required */ }
</style>
<!-- ===== ESLPod Popup Styles (verbatim from index-1.html) ===== -->
<style id="eslpod-popup-css">
/* --- Modal shell --- */
.lm-overlay{position:fixed;inset:0;background:rgba(0,0,0,.55);display:none;z-index:9999;overflow-y:auto}
.lm-modal{position:relative;margin:4vh auto;background:#fff;max-width:880px;border-radius:20px;overflow:hidden;box-shadow:0 20px 60px rgba(0,0,0,.25)}
.lm-grid{display:block}
.lm-body{padding:28px 26px 22px}
.lm-x{position:absolute;top:10px;right:12px;border:none;background:#0000;font-size:28px;line-height:1;cursor:pointer;padding:6px;color:#666}
.lm-x:hover{color:#000}

/* --- Trust badge (no icon) --- */
.lm-badge{display:flex;align-items:center;justify-content:center;background:#eaf6da;color:var(--ep-blue);border:1px solid #c8e3a3;border-radius:999px;padding:6px 16px;font-size:13px;font-weight:600;margin:0 auto 14px;width:fit-content;max-width:95%;line-height:1.3;letter-spacing:.01em;text-align:center}

/* --- Typography & CTA (modal) --- */
.popup-title{font-size:32px;font-family:Arial,'Helvetica Neue',Helvetica,sans-serif;font-weight:700;text-align:center;margin:0 auto 14px;line-height:1.14;max-width:620px;display:block;box-sizing:border-box;padding:0 10px;color:#222}
.popup-body{font-size:18px;font-family:Arial,'Helvetica Neue',Helvetica,sans-serif;text-align:center;margin:0 auto 14px;line-height:1.34;max-width:520px;color:#222;padding:0 8px}
.popup-disclaimer{font-size:14px;color:#555;text-align:center;margin-top:10px;margin-bottom:0;font-family:Arial,'Helvetica Neue',Helvetica,sans-serif;max-width:95%;margin-left:auto;margin-right:auto;box-sizing:border-box;padding:0 6px;font-style:italic}

.lm-form{display:grid;grid-template-columns:1fr 1fr;gap:14px;margin-top:14px}
.lm-form .full{grid-column:1/-1}
.lm-form label{display:block;font-size:13px;font-weight:600;margin-bottom:4px;color:#333}
.lm-input,.lm-select{width:100%;font-size:16px;padding:12px;border:1px solid #cfd3d7;border-radius:10px;box-sizing:border-box;font-family:inherit}
.lm-input:focus,.lm-select:focus{border-color:var(--ep-green);box-shadow:0 0 0 3px rgba(21,155,151,.18);outline:none}

/* Padlock + secure microcopy under email */
.lm-secure{font-size:12px;color:#555;margin-top:6px;display:flex;align-items:center;gap:5px;line-height:1.3}
.lm-secure-icon{font-size:13px}

.popup-cta{font-size:22px;font-family:Arial,'Helvetica Neue',Helvetica,sans-serif;font-weight:700;text-align:center;padding:16px 26px;border-radius:18px;display:flex;align-items:center;justify-content:center;background:var(--ep-green);color:#fff;border:0;cursor:pointer;width:100%}
.popup-cta:hover{background:#118380}
.popup-cta[disabled]{opacity:.6;cursor:wait}

/* --- Mobile tweaks --- */
.lm-modal{max-width:520px}
@media (max-width:480px){.popup-title{font-size:26px}.popup-body{font-size:16px}.lm-badge{font-size:12px}}

/* --- RTL support (applied dynamically when language=ar is selected) --- */
.lm-body[dir="rtl"]{text-align:right}
.lm-body[dir="rtl"] .lm-secure{flex-direction:row-reverse;justify-content:flex-end}

/* --- Language-specific font stacks (applied dynamically) --- */
.lm-body[lang="ja"]{font-family:"Hiragino Kaku Gothic ProN","Hiragino Sans","Yu Gothic","Meiryo",Arial,sans-serif}
.lm-body[lang="ko"]{font-family:"Apple SD Gothic Neo","Malgun Gothic","Nanum Gothic",Arial,sans-serif}
.lm-body[lang="zh-hans"]{font-family:"PingFang SC","Hiragino Sans GB","Microsoft YaHei",Arial,sans-serif}
.lm-body[lang="zh-hant"]{font-family:"PingFang TC","Hiragino Sans CNS","Microsoft JhengHei",Arial,sans-serif}
.lm-body[lang="ar"]{font-family:"Geeza Pro","Simplified Arabic","Traditional Arabic",Arial,sans-serif}
</style>
<!-- CTA COLOUR SYSTEM — v48 (2026-08-04). Jeff's rule: colour means offer.
     ORANGE var(--ep-orange) = the free lesson. Every free-lesson button on the
     page, in either trial state, including both nav twins and the sticky twin
     that show while data-trial="off".
     GREEN var(--ep-green-cta) = anything that leads to paying. The trial pill,
     the mobile sticky trial, and "See plans & pricing".

     WHY NOT var(--ep-green), #159b97. White text on it measures 2.39:1, which
     fails WCAG AA at every size, and it was carrying the pricing button. The
     deeper #159b97 measures 5.43:1 and passes outright. Measured, not eyeballed.
     For reference the orange is 3.39:1, which clears AA for large text only and
     is under that threshold at the sizes these labels actually use. Jeff knows.

     This block is deliberately the LAST stylesheet in the file. .ep-btn-green
     sets its background with !important, so these rules need to come after it.
     The nav and sticky rules win on specificity instead, .ep-nav-cta.ep-trial-only
     beating .ep-nav-cta, so they hold wherever they sit. -->
<style id="cta-colour-v48">
/* paid path */
.ep-btn-green { background: var(--ep-green-cta) !important; }
.ep-nav-cta.ep-trial-only { background: transparent !important; border: 2px solid var(--ep-green2); color: var(--ep-green2) !important; box-shadow: none !important; }
.ep-mobile-stickycta a.ep-trial-only { background: var(--ep-green-cta); box-shadow: 0 6px 16px color-mix(in srgb, var(--ep-green-cta) 32%, transparent); }

/* free lesson: now uses the same violet CTA color, stated rather than inherited */
.ep-nav-cta.ep-notrial-only { background: var(--ep-orange); }
.ep-mobile-stickycta a.ep-notrial-only { background: var(--ep-orange); box-shadow: 0 6px 16px color-mix(in srgb, var(--ep-orange) 32%, transparent); }
</style>
<style id="ep-navfix-v1">
/* Mobile nav parity fix, 2026-08-12. Applied identically to Home, Apple
   Podcast, FAQ, Method, About and Listen so the six behave the same on a phone.

   1. The bar had a hard floor of 344px: logo 75 + Login 55 + language pill 76
      + burger 40, four 18px gaps and 40px of padding, and nothing shrinks. At
      or below 343px the burger was pushed off the right edge and the page
      scrolled sideways — on a 280px Galaxy Fold the menu was unreachable, and
      on any phone at 110% browser zoom it was half gone. Tightening the gap and
      padding on phones and dropping the Login link (it is already the fourth
      item inside the drawer) takes the floor to about 235px.
   2. The drawer was positioned 56px from the top, from a comment that assumed a
      56px bar. The bar actually measures 68-74px: the burger button is taller
      than the 32px mobile logo, and the language pill grows a few more pixels in
      Japanese, so the bar measures 68-74px depending on locale and width and no
      single offset is exactly right. Rather than guess, the drawer starts at
      60px — above the bottom of the shortest bar, so no strip of page can ever
      show through the seam — and its top padding puts the first item at 86px,
      clear of even the tallest bar. Do NOT set top:0 here: #ep-nav is sticky
      inside a lower stacking context, so a full-height drawer paints OVER the
      bar no matter what z-index the nav is given, and the burger becomes
      unclickable — the menu opens and cannot be closed. This overrides both
      spellings in the wild: Method sets inset via var(--ep-nav-h), the rest use
      a literal.
   3. The burger was a bare glyph beside a bright white language pill and did not
      read as a control. A hairline outline fixes that. Mobile only: at 961px and
      up the burger is display:none anyway and this rule never applies. */
@media (max-width: 960px) {
  .ep-nav-burger { border: 1px solid rgba(255,255,255,.28); }
  .ep-mobile-drawer { top: 60px; padding: 26px 20px 24px; }
}
@media (max-width: 420px) {
  .ep-nav-inner { gap: 10px; padding: 12px 12px; }
  .ep-nav-loginlink { display: none; }
}
/* v67: keep every nav item on one line. See the patch script for the
   measurements; the short version is that the links wrapping, not the CTA,
   is what broke German, Spanish, Indonesian and Thai. */
#ep-nav .ep-nav-cta,
#ep-nav .ep-nav-link,
#ep-nav .ep-nav-loginlink,
#ep-nav .ep-nav-list > li > a,
#ep-nav .ep-nav-list > li > span { white-space: nowrap; }
@media (min-width: 961px) and (max-width: 1100px) {
  #ep-nav .ep-nav-link { padding-left: 8px; padding-right: 8px; }
  #ep-nav .ep-nav-cta  { padding-left: 13px; padding-right: 13px; }
}
</style>

<!-- Google Tag Manager (noscript) -->
<div data-bigidea-style="grid" data-hero-bg="light" data-hero-cta="orange-primary" data-hero-layout="split" data-mobile-cta="orange" data-screen-label="Home" data-test-style="grid" data-trial="on" data-ue-layout="mockup-right" id="ep-page">
<!-- ╭─── 1. STICKY TOP NAV (Foundation §4.1) ─────────────────────╮ -->
<nav aria-label="Primary" data-i18n-aria="nav_primary_aria" id="ep-nav">
<div class="ep-nav-inner">
<a aria-label="ESLPod home" class="ep-logo" data-i18n-aria="nav_home_aria" href="/">
<img alt="ESLPod" src="https://eslpod.com/assets/ESLPodLogo.png"/>
</a>
<div class="ep-nav-spacer"></div>
<ul class="ep-nav-list">
<li>
<button aria-expanded="false" aria-haspopup="true" class="ep-nav-link" data-drop="lessons">
<span data-i18n="nav_lessons">Lessons</span><span class="ep-nav-caret">▼</span>
</button>
<div class="ep-nav-drop" data-dropfor="lessons">
<a href="https://www.eslpod.com/unlimited-english/"><span data-i18n="nav_ue_learn">Unlimited English</span></a>
<a href="https://www.eslpod.com/podcast/"><span data-i18n="nav_pod_learn">Apple Podcasts</span></a>
<a href="/listen/"><span data-i18n="nav_listen">Listen Free</span></a>
</div>
</li>
<li><a class="ep-nav-link" data-i18n="nav_pricing" href="/unlimited-english/#ep-pricing">Pricing</a></li>
<li><a class="ep-nav-link" data-i18n="nav_themethod" href="https://www.eslpod.com/method/">The Method</a></li>
<li><a class="ep-nav-link" data-i18n="nav_science" href="https://www.eslpod.com/research/">The Science</a></li>
</ul>
<a class="ep-nav-loginlink" data-i18n="nav_login" href="https://sso.teachable.com/secure/147717/users/sign_in">Login</a>
<div class="ep-lang-wrap">
<button aria-controls="ep-lang-menu" aria-expanded="false" aria-haspopup="listbox" aria-label="Change language" class="ep-lang-pill" data-i18n-aria="nav_change_lang_aria" id="ep-lang-pill" type="button">
<span aria-hidden="true" class="ep-globe">
<svg fill="none" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2.2" viewbox="0 0 24 24"><circle cx="12" cy="12" r="9"></circle><path d="M3 12h18"></path><path d="M12 3a14 14 0 0 1 0 18a14 14 0 0 1 0 -18"></path></svg>
</span>
<span id="ep-lang-pill-label">EN</span>
<span aria-hidden="true" class="ep-caret">▾</span>
</button>
<div aria-labelledby="ep-lang-pill" class="ep-lang-menu" id="ep-lang-menu" role="listbox"></div>
</div>
<!-- v40: the desktop nav CTA swaps with the trial state. The free-lesson
         button is .ep-notrial-only, the trial link is .ep-trial-only, and both
         carry .ep-nav-cta-desk so the 1055px media query hides whichever one is
         showing. Exactly one of the two is ever in the DOM flow. -->
<button class="ep-nav-cta ep-nav-cta-desk popup-trigger ep-notrial-only" data-i18n="nav_freecourse_cta" data-modal="#leadmagnet-modal" data-track="cta_click_freecourse_nav" style="font-family:inherit;cursor:pointer" type="button">Get a free lesson</button>
<a class="ep-nav-cta ep-nav-cta-desk ep-trial-only" data-i18n="nav_trial_cta" data-track="cta_click_trial" href="/unlimited-english/#ep-pricing">Start your 7-day trial</a>
<button aria-expanded="false" aria-label="Open menu" class="ep-nav-burger" data-i18n-aria="nav_open_menu_aria" id="ep-burger">☰</button>
</div>
<div class="ep-mobile-drawer" id="ep-mobile-drawer">
<div class="ep-mobile-section" data-i18n="nav_lessons">Lessons</div>
<a data-i18n="nav_ue_learn" href="https://www.eslpod.com/unlimited-english/">Unlimited English</a>
<a data-i18n="nav_pod_learn" href="https://www.eslpod.com/podcast/">Apple Podcasts</a>
<a data-i18n="nav_listen" href="/listen/">Listen Free</a>
<div class="ep-mobile-section" data-i18n="nav_more">More</div>
<a data-i18n="nav_pricing" href="/unlimited-english/#ep-pricing">Pricing</a>
<a data-i18n="nav_themethod" href="https://www.eslpod.com/method/">The Method</a>
<a data-i18n="nav_science" href="https://www.eslpod.com/research/">The Science</a>
<a data-i18n="nav_login" href="https://sso.teachable.com/secure/147717/users/sign_in">Login</a>
<button class="ep-nav-cta popup-trigger ep-notrial-only" data-i18n="nav_freecourse_cta" data-modal="#leadmagnet-modal" data-track="cta_click_freecourse_nav" style="font-family:inherit;cursor:pointer" type="button">Get a free lesson</button>
<a class="ep-nav-cta ep-trial-only" data-i18n="nav_trial_cta" data-track="cta_click_trial" href="/unlimited-english/#ep-pricing">Start your 7-day trial</a>
</div>
</nav>
<!-- ╭─── 2. HERO (#ep-hero) ──────────────────────────────────────╮ -->
<section data-screen-label="01 Hero" id="ep-hero">
<div class="ep-wrap-wide">
<div class="ep-hero-grid">
<p class="ep-eyebrow" data-i18n="hero_eyebrow">For intermediate English learners</p>
<div class="ep-hero-copy">
<h1 class="ep-h1 ep-h1-light" data-i18n="hero_h1"><span class="ep-h1-part">Finally understand real English</span> <span class="ep-h1-part">Then speak it with confidence</span></h1>
<p class="ep-lead ep-lead-light" data-i18n="hero_sub">Built on 40 years of research into how adults actually become fluent, delivered in 20-minute lessons you can do anywhere</p>
<div class="ep-hero-cta-row">
<div class="ep-hero-cta-col">
<button class="ep-btn ep-btn-large popup-trigger" data-i18n="cta_trigger" data-modal="#leadmagnet-modal">Get a free lesson</button>
<div class="ep-hero-microcopy" data-i18n="hero_freecourse_micro" style="margin-top:5px;font-size:12px;color:rgba(255,255,255,.72);line-height:1.5;text-align:center">No credit card needed</div>
<a class="ep-hero-pricing-link" data-i18n="hero_pricing_link" data-track="cta_click_pricing" href="/unlimited-english/#ep-pricing" style="display:inline-block;margin-top:11px;color:rgba(255,255,255,.78);font-size:13px;font-weight:600;text-decoration:none;border-bottom:1px solid rgba(255,255,255,.32);padding-bottom:1px">Already convinced? See plans →</a>
</div>
</div>
<div class="ep-hero-trust">
<span class="ep-stars">★ <span data-i18n="hero_rating">4.8</span></span>
<span data-i18n="hero_trust_apple">Apple Podcasts</span>
<span class="ep-dot">·</span>
<span><b data-i18n="hero_trust_count">1 billion</b> <span data-comment-anchor="b7d4ab7df9-span" data-i18n="hero_trust_learners">downloads</span></span>
</div>
</div>
<div class="ep-hero-portrait">
<img alt="Dr. Jeff McQuillan teaching Comprehensible Input i+1" data-i18n-alt="alt_portrait" src="/assets/headshot-052626.jpg"/>
<div class="ep-hero-portrait-fallback" style="display:none">
<span data-i18n="hero_portrait_name">Dr. Jeff McQuillan</span>
</div>
<div class="ep-hero-portrait-tag"><b data-i18n="hero_portrait_name">Dr. Jeff McQuillan</b> <span aria-hidden="true" class="ep-portrait-sep">· </span><span data-i18n="hero_portrait_cred">Ph.D. in Applied Linguistics + Education</span></div>
</div>
</div>
</div>
</section>
<!-- ╭─── 2b. PROOF BAR — STATS + PRESS (#ep-press) ───────────────╮ -->
<section data-screen-label="02 Proof bar" id="ep-press">
<div class="ep-wrap-wide">
<div class="ep-stats" style="margin-bottom:20px">
<div class="ep-stat"><b data-i18n="stat1_num">15M</b><span data-i18n="stat1">Learners</span></div>
<div class="ep-stat"><b data-i18n="stat2_num">189</b><span data-i18n="stat2">Countries</span></div>
<div class="ep-stat"><b data-i18n="stat3_num">1,800+</b><span data-i18n="stat3">Lessons</span></div>
<div class="ep-stat"><b data-i18n="stat4_num">21</b><span class="ep-stat-yrs"><span data-i18n="stat4_unit">years</span> <span data-i18n="stat4">of ESLPod®</span></span></div>
</div>
<hr class="ep-press-divider"/>
<p class="ep-press-eyebrow" data-i18n="press_eyebrow">As featured in</p>
<div class="ep-press-row">
<img alt="CNN" decoding="async" loading="lazy" src="/assets/cnn_white.png"/>
<img alt="El País" decoding="async" loading="lazy" src="/assets/el_pais_white.png"/>
<img alt="Macworld" decoding="async" loading="lazy" src="/assets/macworld_white.png"/>
<img alt="Deutschlandfunk" decoding="async" loading="lazy" src="/assets/deutschlandfunk_white.png"/>
<img alt="O Diário" decoding="async" loading="lazy" src="/assets/odiario_white.png"/>
<img alt="Techwd" decoding="async" loading="lazy" src="/assets/techwd_arabic_white.png"/>
</div>
</div>
</section>
<!-- ╭─── 4. AUDIO SAMPLE (#ep-audio) ─────────────────────────────╮ -->
<section class="ep-sec tl-audio" data-screen-label="03 Audio sample" id="ep-audio">
<div class="ep-wrap">
<div class="tl-audio-inner">
<div class="tl-audio-intro">
<p class="ep-eyebrow ep-eyebrow-muted" data-i18n="audio_eyebrow">Hear it for yourself</p>
<h2 class="ep-h2" data-i18n="audio_h2">120 seconds inside a Daily English lesson</h2>
<p class="ep-lead" data-i18n="audio_lead">Listen to the two-minute sample lesson below. We explain every difficult word in clear English you already understand.</p>
</div>
<div class="tl-audio-card">
<span class="tl-audio-tag" data-i18n="audio_tag">Sample Lesson</span>
<h3 data-i18n="audio_title">Competing in Business</h3>
<p class="tl-audio-meta" data-i18n="audio_meta">Complete lessons have 20-25 minutes of audio + 30-40 minutes of reading material</p>
<div class="tl-audio-player">
<button aria-label="Play sample" class="tl-audio-play" data-i18n-aria="audio_play_aria" id="tl-audio-play">
<svg aria-hidden="true" class="ic-play" fill="currentColor" viewbox="0 0 24 24"><path d="M8 5v14l11-7z"></path></svg>
<svg aria-hidden="true" class="ic-pause" fill="currentColor" viewbox="0 0 24 24"><path d="M6 4h4v16H6zM14 4h4v16h-4z"></path></svg>
</button>
<div class="tl-audio-track">
<div class="tl-audio-bar"><div class="tl-audio-bar-fill" id="tl-audio-bar-fill"></div></div>
<div class="tl-audio-time"><span id="tl-audio-cur">0:00</span><span id="tl-audio-total">2:00</span></div>
</div>
<audio id="tl-audio-el" preload="metadata" src="https://traffic.libsyn.com/secure/eslpod/de984.mp3"></audio>
</div>
<hr class="tl-audio-divider"/>
<div class="tl-audio-transcript" data-i18n="audio_transcript">
<p class="tl-audio-line">You hear the dialogue spoken at a clear speed in the beginning of the lesson, like this line:</p>
<div class="tl-audio-quote"><p><strong>Yoko:</strong> I'm thinking we could go in another direction. What do you think about offering <b>freebies</b> or <b>gifts with purchase</b>?</p></div>
<details class="tl-audio-toggle">
<summary><span class="tl-audio-show">Show the explanation ↓</span><span class="tl-audio-hide">Hide the explanation ↑</span></summary>
<p class="tl-audio-line">Then you hear an explanation of that line in the main lesson:</p>
<div class="tl-audio-quote"><p><strong>Dr. Jeff:</strong> Yoko says, "I'm thinking we could go in another direction," meaning we could do something different. She asks, "What do you think about offering <b>freebies</b> or <b>gifts with purchase</b>?" A <b>freebie</b> is something you get for free, that you do not pay for.</p></div>
<div class="tl-audio-quote"><p>If you go to the grocery store in the United States, sometimes the store will give you a small piece of food that they are selling for you to taste. That's a freebie. You don't have to pay anything for it. Of course, they want you to buy some.</p></div>
<div class="tl-audio-quote"><p>The phrase <b>gift with purchase</b> means you get something extra when you buy a certain product. So, a woman goes to the store to buy some makeup, things to make her face look beautiful, more beautiful, let's say. The store may offer her a gift with purchase. If she buys this 50-dollar package of makeup, they will give her a two-dollar lipstick or ten-dollar lipstick. I don't know how much lipsticks cost. I don't buy lipstick. Well, not anymore.</p></div>
<p class="tl-audio-line">Finally, you will hear the line again at a normal speed at the end of the lesson:</p>
<div class="tl-audio-quote"><p><strong>Yoko:</strong> I'm thinking we could go in another direction. What do you think about offering freebies or gifts with purchase?</p></div>
</details>
</div>
</div>
<p class="tl-audio-after"><span data-i18n="audio_after">Want to hear more?</span> <button class="popup-trigger" data-i18n="audio_after_link" data-modal="#leadmagnet-modal" style="background:none;border:0;color:var(--ep-dark);font-weight:700;text-decoration:underline;font-family:inherit;font-size:inherit;padding:0;cursor:pointer" type="button">Get a free lesson →</button></p>
</div>
</div>
</section>
<!-- ╭─── 3. BIG IDEA BAND (#ep-bigidea) ──────────────────────────╮ -->
<section class="ep-sec" data-screen-label="03 Big idea" id="ep-bigidea">
<div class="ep-wrap-wide">
<div class="ep-bigidea-intro" style="text-align:center">
<p class="ep-eyebrow" data-i18n="bigidea_eyebrow">Rich Comprehensible Input™</p>
<h2 class="ep-h2" data-i18n="bigidea_h2">Why this works when apps don't</h2>
<p class="ep-lead" data-i18n="bigidea_lead" style="max-width:720px;margin:0 auto 1em">Apps, AI chatbots, and live tutors focus on all the wrong things. Our Rich Comprehensible Input™ lessons give you what 40 years of research says actually works to make you a confident, fluent speaker.</p>
</div>
<div class="ep-pillars">
<div class="ep-pillar">
<div aria-hidden="true" class="ep-pillar-icon" data-tone="dark">
<svg fill="none" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2.2" viewbox="0 0 24 24"><path d="M12 5v14M8 9v6M16 9v6M4 11v2M20 11v2"></path></svg>
</div>
<h3 data-i18n="pillar1_title">Hear the words clearly</h3>
<p data-i18n="pillar1_body">Every Daily English lesson is recorded with our unique Slow-Explain-Fast Technique™. You understand every word, with real American English, not textbook scripts.</p>
<div class="ep-pillar-mech">
<b data-i18n="pillar1_mech_title">Slow-Explain-Fast Technique™</b>
<span data-i18n="pillar1_mech_body">Conversations recorded at slow speed, explained, then reviewed at natural speed</span>
</div>
</div>
<div class="ep-pillar">
<div aria-hidden="true" class="ep-pillar-icon" data-tone="dark">
<svg fill="none" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2.2" viewbox="0 0 24 24"><path d="M4 4h12a4 4 0 0 1 4 4v12H8a4 4 0 0 1-4-4z"></path><path d="M8 8h8M8 12h8M8 16h5"></path></svg>
</div>
<h3 data-i18n="pillar2_title">Get large amounts of input</h3>
<p data-i18n="pillar2_body">Each lesson goes beyond the audio with a 10-page Learning Guide. You get the cultural context for English idioms, phrases, and expressions, and learn what Americans know and use every day.</p>
<div class="ep-pillar-mech">
<b data-i18n="pillar2_mech_title">Daily &amp; Cultural English Library</b>
<span data-i18n="pillar2_mech_body">1,800+ lessons on daily life, business, travel, idioms, culture, and more</span>
</div>
</div>
<div class="ep-pillar">
<div aria-hidden="true" class="ep-pillar-icon" data-tone="dark">
<svg fill="none" stroke="currentColor" stroke-linecap="round" stroke-linejoin="round" stroke-width="2.2" viewbox="0 0 24 24"><circle cx="12" cy="12" r="9"></circle><circle cx="12" cy="12" r="5"></circle><circle cx="12" cy="12" fill="currentColor" r="1.6"></circle></svg>
</div>
<h3 data-i18n="pillar3_title">Speak with confidence</h3>
<p data-i18n="pillar3_body">No time is wasted on games, exercises, or English you cannot understand. Research-driven i + 1 Targeting™ makes sure every lesson is at your level. When you understand, you start speaking without stopping to translate.</p>
<div class="ep-pillar-mech">
<b data-i18n="pillar3_mech_title">i + 1 Targeting™</b>
<span data-i18n="pillar3_mech_body">Input one step above your current level, exactly what you need to improve</span>
</div>
</div>
</div>
<div class="ep-bigidea-cta">
<a class="ep-textlink" data-i18n="bigidea_cta" href="https://www.eslpod.com/research/">How Rich Comprehensible Input™ works →</a>
</div>
</div>
</section>
<!-- ╭─── 4b. CURRICULUM / WHAT'S INSIDE UE (#ep-curriculum) ──────╮ -->
<section class="ep-sec" data-screen-label="04 What is inside" id="ep-curriculum">
<div class="ep-wrap-wide">
<div class="ep-curriculum-intro" style="text-align:center">
<p class="ep-eyebrow" data-i18n="curr_eyebrow">What You Get</p>
<h2 class="ep-h2" data-i18n="curr_h2">What's Inside Unlimited English</h2>
<p class="ep-lead" data-i18n="curr_lead" style="max-width:760px;margin:0 auto 1em">1,800+ lessons across 16 categories • Every lesson comes with a 10-page Learning Guide</p>
</div>
<div class="ep-curriculum-grid">
<div class="ep-curriculum-col">
<h3 class="ep-curriculum-coltitle" data-i18n="curr_de_title">Daily English</h3>
<p class="ep-curriculum-colmeta" data-i18n="curr_de_meta">8 categories · 1,500+ topics</p>
<ul class="ep-curriculum-cats">
<li><b data-i18n="cat_business">Business</b> <span class="ep-cat-ex" data-i18n="cat_business_ex">Tough Negotiations, Making a Presentation, Hiring for a Job, Negotiating Salary, Giving Opinions, and more</span></li>
<li><b data-i18n="cat_travel">Travel + Transportation</b> <span class="ep-cat-ex" data-i18n="cat_travel_ex">Asking for Directions, Hotel Reservations, Going Through Customs, Reserving a Rental Car, Arranging Airport Transfer, and more</span></li>
<li><b data-i18n="cat_dailylife">Daily Life</b> <span class="ep-cat-ex" data-i18n="cat_dailylife_ex">Talking about Time, Being Late for an Event, Avoiding Giving an Opinion, Getting Dressed, Types of Neighborhoods, and more</span></li>
<li><b data-i18n="cat_money">Money</b> <span class="ep-cat-ex" data-i18n="cat_money_ex">Paying the Bills, Using an ATM, Talking about Prices, Investing Your Money, Being in Debt, Giving Correct Change, and more</span></li>
<li><b data-i18n="cat_shopping">Shopping</b> <span class="ep-cat-ex" data-i18n="cat_shopping_ex">Shopping at the Mall, At the Bakery, Garage Sales, At the Butchers, Finding a Bargain, Getting Bad Service, and more</span></li>
<li><b data-i18n="cat_health">Health + Medicine</b> <span class="ep-cat-ex" data-i18n="cat_health_ex">Scheduling a Medical Appointment, Childhood Illnesses and Diseases, Health Insurance Plans, Sports-Related Injuries, and more</span></li>
<li><b data-i18n="cat_entertainment">Entertainment + Sports</b> <span class="ep-cat-ex" data-i18n="cat_entertainment_ex">At the Movies, Seeing a Play, Bar Hopping, Pop and Jazz Music, Watching the Olympics, Betting on Sports, and more</span></li>
<li><b data-i18n="cat_relationships">Relationships + Family</b> <span class="ep-cat-ex" data-i18n="cat_relationships_ex">Seeing Old Friends, Going to a Wedding, Describing a Person's Looks, Dating a Younger Man/Woman, Saying Goodbye, and more</span></li>
</ul>
</div>
<div class="ep-curriculum-col">
<h3 class="ep-curriculum-coltitle" data-i18n="curr_ce_title">Cultural English</h3>
<p class="ep-curriculum-colmeta" data-i18n="curr_ce_meta">8 categories · 2,000+ topics</p>
<ul class="ep-curriculum-cats">
<li><b data-i18n="cat_popculture">Popular Culture</b> <span class="ep-cat-ex" data-i18n="cat_popculture_ex">Super Bowl Parties, Online Dating, Hip Hop Culture, Filming in L.A., and more</span></li>
<li><b data-i18n="cat_langqa">Common English Language Q&amp;A</b> <span class="ep-cat-ex" data-i18n="cat_langqa_ex">British vs. American English, Can vs. Can't, Effect vs. Affect, Pronouncing "a", Common Abbreviations, and more</span></li>
<li><b data-i18n="cat_writers">American Writers and Artists</b> <span class="ep-cat-ex" data-i18n="cat_writers_ex">Stephen King, Mark Twain, Bob Dylan, Yo-Yo Ma, Ansel Adams, Kate Smith, Georgia O'Keefe, Raymond Chandler, and more</span></li>
<li><b data-i18n="cat_places">Famous Places, Cities, and States</b> <span class="ep-cat-ex" data-i18n="cat_places_ex">Las Vegas, Niagara Falls, Yellowstone, Honolulu, Graceland, Disneyland, Boston, Carnegie Hall, California, and more</span></li>
<li><b data-i18n="cat_movies">Classic American Movies</b> <span class="ep-cat-ex" data-i18n="cat_movies_ex">Star Wars, Bonnie and Clyde, Apocalypse Now, Titanic, The Shawshank Redemption, Blockbuster Movies, and more</span></li>
<li><b data-i18n="cat_books">Well-Known Books</b> <span class="ep-cat-ex" data-i18n="cat_books_ex">Dr. Seuss Children's Books, Chicken Soup Series, Rich Dad Poor Dad, Catch-22, The Farmer's Almanac, The Road, and more</span></li>
<li><b data-i18n="cat_presidents">Presidents, Politics, and History</b> <span class="ep-cat-ex" data-i18n="cat_presidents_ex">John F. Kennedy, Pledge of Allegiance, George Washington, Barack Obama, Liberty Bell, Salem Witch Trials, and more</span></li>
<li><b data-i18n="cat_holidays">Holidays and Festivals</b> <span class="ep-cat-ex" data-i18n="cat_holidays_ex">Thanksgiving, Christmas and Kwanzaa, Mardi Gras in New Orleans, Fourth of July, Memorial Day, Martin Luther King Jr. Day, and more</span></li>
</ul>
</div>
</div>
<p class="ep-curriculum-foot" data-i18n="curr_foot">Every lesson: audio + 10-page Learning Guide • Listen anywhere • Cancel anytime</p>
</div>
</section>
<!-- ╭─── 5. UE FEATURE CARD (#ep-ue) ─────────────────────────────╮ -->
<section class="ep-sec" data-screen-label="04 Unlimited English" id="ep-ue">
<div class="ep-wrap-wide">
<div class="ep-ue-card">
<div aria-hidden="true" class="ep-ue-mockups">
<div class="ep-ue-desktop">
<img alt="Unlimited English on the web" data-i18n-alt="alt_ue_desktop" decoding="async" loading="lazy" src="https://eslpod.com/assets/webscreenshot.png" style="width:100%;height:100%;object-fit:cover;display:block"/>
</div>
<div class="ep-ue-phone">
<img alt="Unlimited English on mobile" data-i18n-alt="alt_ue_mobile" decoding="async" loading="lazy" src="https://eslpod.com/assets/appscreenshot.png" style="width:100%;height:100%;object-fit:cover;display:block"/>
</div>
</div>
<div class="ep-ue-body">
<p class="ep-eyebrow ep-eyebrow-muted" data-i18n="ue_eyebrow">The main course</p>
<h2 class="ep-h2" data-i18n="ue_h2">Unlimited English</h2>
<p class="ep-ue-lead" data-i18n="ue_lead">1,800+ lessons in clear, real English</p>
<p class="ep-ue-price-line"><b>$290<span class="ep-ue-price-per" data-i18n="ue_price_per">/year</span></b><span class="ep-ue-price-billed" data-i18n="ue_price_billed"> or $29/month</span></p>
<ul class="ep-ue-bullets">
<li class="ep-ue-check"><span data-i18n="ue_b1"><b>1,800+ lessons</b> on daily life, business, travel, idioms, culture, and more</span></li>
<li class="ep-ue-check"><span data-i18n="ue_b2"><b>Full English transcripts</b> for every lesson, every word spoken</span></li>
<li class="ep-ue-check"><span data-i18n="ue_b3"><b>Vocabulary explained</b> in clear English you can easily understand</span></li>
<li class="ep-ue-check"><span data-i18n="ue_b4"><b>Listen anywhere</b> on your phone, browser, and iOS or Android app</span></li>
</ul>
<div class="ep-ue-cta-row">
<a class="ep-btn ep-btn-green" data-i18n="ue_cta_primary" data-track="cta_click_pricing" href="/unlimited-english/#ep-pricing">See plans &amp; pricing →</a>
<a class="ep-textlink" data-i18n="ue_cta_secondary" href="/unlimited-english/">See everything inside Unlimited English →</a>
</div>
</div>
</div>
</div>
</section>
<!-- ╭─── 5b. HOW IT WORKS STRIP (#ep-howitworks) ────────────────╮ -->
<section class="ep-sec" data-screen-label="05b How it works" id="ep-howitworks">
<div class="ep-wrap-wide">
<div class="ep-howitworks-intro" style="text-align:center">
<p class="ep-eyebrow" data-i18n="how_eyebrow">How it works</p>
</div>
<ol class="ep-howitworks-steps">
<li>
<span class="ep-howitworks-num">1</span>
<span class="ep-howitworks-text" data-i18n="how_step1">Pick your 20-30 minute lesson each day</span>
</li>
<li>
<span class="ep-howitworks-num">2</span>
<span class="ep-howitworks-text" data-i18n="how_step2">Listen to the audio while you are on the train, in your car, on your walk, or drinking your coffee</span>
</li>
<li>
<span class="ep-howitworks-num">3</span>
<span class="ep-howitworks-text" data-i18n="how_step3">i + 1 Targeting™ takes you to the next level so you can speak more fluently</span>
</li>
</ol>
</div>
</section>
<!-- ╭─── 6. FREE COURSE CAPTURE (#ep-magnet) ─────────────────────╮ -->
<section class="ep-sec" data-screen-label="05 Free lesson" id="ep-magnet">
<div class="ep-wrap">
<div class="ep-magnet-card">
<p class="ep-eyebrow ep-eyebrow-muted" data-i18n="magnet_eyebrow">Free lesson</p>
<h2 class="ep-h2" data-i18n="magnet_h2">5 Phrases That Make You Easier to Understand in English</h2>
<p class="ep-magnet-sub" data-i18n="magnet_sub">A 10-minute audio lesson with the five exact phrases you can use when someone doesn’t understand what you said in English</p>
<button class="popup-trigger" data-i18n="magnet_cta" data-modal="#leadmagnet-modal">Get a free lesson</button>
<p class="ep-magnet-microcopy" data-i18n="magnet_micro">No credit card needed</p>
</div>
</div>
</section>
<!-- ╭─── 8. FOUNDER CREDIBILITY (#ep-founder) ────────────────────╮ -->
<section class="ep-sec" data-screen-label="07 Founders" id="ep-founder">
<div class="ep-wrap-wide">
<div class="ep-founder-grid">
<div class="ep-founder-portrait">
<img alt="Bovard Administration Building at the University of Southern California" data-i18n-alt="alt_usc" decoding="async" loading="lazy" src="https://www.eslpod.com/assets/Bovard_USC.png"/>
<div class="ep-founder-portrait-fallback" style="display:none">
<span>University of Southern California</span>
</div>
</div>
<div class="ep-founder-body">
<p class="ep-eyebrow" data-i18n="founder_eyebrow">Your professors</p>
<h2 class="ep-h2" data-i18n="founder_h2">Dr. Jeff McQuillan &amp; Dr. Lucy Tse</h2>
<p data-i18n="founder_p1">ESLPod® launched in 2005 as the first English learning podcast in the world. Today, it is one of the most respected and popular English learning programs anywhere. More than 15 million people have learned English with ESLPod, with lessons downloaded over one billion times.</p>
<p data-i18n="founder_p2">The host is Dr. Jeff McQuillan. He holds a Ph.D. in applied linguistics and education from the University of Southern California, and has spent more than 35 years helping adult learners move from intermediate to advanced English. Dr. McQuillan has published more than 60 books and scientific articles on language learning and has appeared on CNN as a leading language expert.</p>
<p data-i18n="founder_p3">Dr. Lucy Tse writes and produces the ESLPod lessons. She holds a Ph.D. in second language acquisition from the University of Southern California. Dr. Tse wrote the best-selling book, <em>Why Don’t They Learn English?</em>, has published dozens of research articles, and has contributed to the <em>Harvard Educational Review</em>.</p>
<p data-i18n="founder_p4">The ESLPod method is built on 40 years of <a href="https://www.eslpod.com/research/" style="color:inherit;text-decoration:underline">research</a> by Dr. McQuillan, Dr. Tse, and their former professor and co-author, Dr. Stephen Krashen: When you give learners enough rich, interesting, and understandable input, the ability to speak naturally follows.</p>
<ul class="ep-founder-credlist">
<li><span data-i18n="cred3">15M learners taught · 21 years of ESLPod.com · 40+ years of comprehensible input research</span></li>
</ul>
</div>
</div>
</div>
</section>
<!-- ╭─── 9. TESTIMONIALS (#ep-celebrate) ─────────────────────────╮ -->
<section class="ep-sec" data-screen-label="08 Testimonials" id="ep-celebrate">
<div class="ep-wrap-wide">
<div class="ep-celebrate-head">
<p class="ep-eyebrow ep-eyebrow-muted" data-i18n="celeb_eyebrow">Learners speaking now</p>
</div>
<div class="ep-carousel" id="ep-carousel">
<div class="ep-carousel-track" id="ep-carousel-track">
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi1_quote">"I've been listening to ESLPod® week after week, more than 1,000 lessons. My English has improved dramatically. I can watch films in English without subtitles and read English books. The icing on the cake was when a native Englishman I met in Denmark mistook me for an American. I've obviously acquired a slight Californian accent."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi1_flag">🇩🇪</span><b data-i18n="testi1_name">Jake</b><span class="ep-tcard-loc" data-i18n="testi1_loc">Germany</span></div>
</article>
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi2_quote">"Thanks to your team I got the job of my dreams. I recently became a pilot for Aeroflot - Russian Airlines. All pilots who fly internationally have to pass the ICAO English language proficiency test at level 4. That test requires the ability to speak and think in English. I could never have passed it without your lessons. I even became the best student in my trainee group."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi2_flag">🇷🇺</span><b data-i18n="testi2_name">Dmitry</b><span class="ep-tcard-loc" data-i18n="testi2_loc">Russia (Aeroflot pilot)</span></div>
</article>
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi3_quote">"I have been listening to your lessons for several months. My English has improved ever since. Thank you for your work. My results for the TOEIC exam improved from 570 to 840. Thanks again for your work."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi3_flag">🇯🇵</span><b data-i18n="testi3_name">Yuzo</b><span class="ep-tcard-loc" data-i18n="testi3_loc">Japan (TOEIC 570 → 840)</span></div>
</article>
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi4_quote">"Words cannot express how appreciative I am. My close friend introduced your website to me, and I've been keeping up with it for nine months. It has been a great fortune to learn from linguists like you. Now I feel confident, because my English has improved to the point where I have no trouble communicating with native speakers. I'm no longer given away by my accent."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi4_flag">🇮🇷</span><b data-i18n="testi4_name">Farrie</b><span class="ep-tcard-loc" data-i18n="testi4_loc">Iran</span></div>
</article>
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi5_quote">"I'm a student from Brazil and I had a lot of problems learning English. One day I found your lessons while I was studying for the TOEFL, and you saved my life. I got 100% on the TOEFL listening test. Now I'm attending classes at UCSD. Here in San Diego I listen to five lessons a day, and because of you I can understand everybody around me, including the professor."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi5_flag">🇧🇷</span><b data-i18n="testi5_name">Raphael</b><span class="ep-tcard-loc" data-i18n="testi5_loc">Brazil</span></div>
</article>
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi6_quote">"I can't find the words to say how grateful I am. Your lessons are by far the best I've ever found, inspiring, comforting, and enjoyable. I've learned how to speak English not just adequately, but appropriately, fluently, and even with elegance. Whenever someone asks me for a source for learning English, I don't hesitate to say ESLPod® is the best you can find. A warm hug from beautiful Cali, Colombia to the people in beautiful Los Angeles."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi6_flag">🇨🇴</span><b data-i18n="testi6_name">Jhonatan</b><span class="ep-tcard-loc" data-i18n="testi6_loc">Cali, Colombia</span></div>
</article>
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi7_quote">"I would like to express my sincere gratitude to you all. I appreciate your pragmatic, approachable, and valuable online English lessons, prepared at a high professional level. I am so glad that through your online guidance, I can make progress in English. I am a Mongolian girl studying in Germany. A friend recently told me about this website, and I'm happy I know it now."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi7_flag">🇲🇳</span><b data-i18n="testi7_name">Tsolmon</b><span class="ep-tcard-loc" data-i18n="testi7_loc">Mongolia</span></div>
</article>
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi8_quote">"I found ESLPod® some months ago when I was searching the Internet for a way to finally improve my listening and comprehension. Now I have about 500 lessons on my phone, and I listen to your voice at least one hour a day during my commute to work. You have become a friend of mine. I have improved my English in these months as I never had before, and I'm one of those people who took lots of English courses and watched many hours of English-language TV shows in my life."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi8_flag">🇮🇹</span><b data-i18n="testi8_name">Marco</b><span class="ep-tcard-loc" data-i18n="testi8_loc">Italy</span></div>
</article>
<article class="ep-tcard">
<div class="ep-tcard-stars">★★★★★</div>
<p class="ep-tcard-quote" data-i18n="testi9_quote">"Since Jaime introduced me to it a couple of months ago, I listen to ESLPod® every day while going to work by bus. I work as a bioinformatician at a Cancer Research Center. I read scientific papers every day, but I have a real problem with speaking. The pronunciation of some words is tricky. So I find it very important to listen regularly to a native speaker."</p>
<div class="ep-tcard-meta"><span class="ep-tcard-flag" data-i18n="testi9_flag">🇪🇸</span><b data-i18n="testi9_name">Txema</b><span class="ep-tcard-loc" data-i18n="testi9_loc">Spain</span></div>
</article>
</div>
<div class="ep-carousel-controls">
<button aria-label="Previous" class="ep-carousel-btn" data-i18n-aria="carousel_prev_aria" id="ep-carousel-prev">‹</button>
<div class="ep-carousel-dots" id="ep-carousel-dots"></div>
<button aria-label="Next" class="ep-carousel-btn" data-i18n-aria="carousel_next_aria" id="ep-carousel-next">›</button>
</div>
</div>
</div>
</section>
<!-- ╭─── 10. FAQ SNIPPET (#ep-faq) ───────────────────────────────╮ -->
<section class="ep-sec" data-screen-label="09 FAQ" id="ep-faq">
<div class="ep-wrap" data-comment-anchor="c3def988de-div">
<div class="ep-celebrate-head" style="margin-bottom:24px">
<h2 class="ep-h2" data-i18n="faq_h2">Common Questions</h2>
</div>
<div class="ep-faq-list">
<details class="ep-faq-item">
<summary data-i18n="faq3_q">What level of English do I need to start?</summary>
<div class="ep-faq-item-body" data-i18n="faq3_a">Lessons are designed for adult intermediate learners: people who can read a lot of English, understand most of it, but still find it hard to speak easily or understand native speakers. If you've tried apps like Duolingo or Babbel and feel you're not making progress, you're in the right place.</div>
</details>
<details class="ep-faq-item">
<summary data-i18n="faq_english_q">Are all of your lessons 100% in English?</summary>
<div class="ep-faq-item-body" data-i18n="faq_english_a">Yes! All of our lessons are designed for intermediate and advanced English users. We do not use any translation. All lessons and instructions are completely in English to give you complete immersion in the language.</div>
</details>
<!-- v29: trial FAQ, hidden while data-trial="off". When the trial goes
           live, add the matching Q and A to the FAQPage JSON-LD in <head>. -->
<details class="ep-faq-item ep-trial-only">
<summary data-i18n="faq_trial_q">How does the 7-day free trial work?</summary>
<div class="ep-faq-item-body" data-i18n="faq_trial_a">You get 7 days to try all of our 1,800+ lessons for free. Cancel anytime before Day 7. On Day 8, your credit card is charged either $29 (for the monthly plan) or $290 (for the annual plan), depending on which one you signed up for. We will email you 2 days before we charge you to remind you.</div>
</details>
<details class="ep-faq-item">
<summary data-i18n="faq4_q">Does the membership auto-renew? Can I cancel anytime?</summary>
<div class="ep-faq-item-body" data-i18n="faq4_a">Yes. Monthly memberships renew every month, annual memberships renew every year, until you cancel. You can cancel from your account easily. Every course and membership has a 30-day money-back guarantee.</div>
</details>
<details class="ep-faq-item">
<summary data-i18n="faq5_q">How is this different from Duolingo, Babbel, or online tutors?</summary>
<div class="ep-faq-item-body" data-i18n="faq5_a">Most apps and tutors focus on either rote memorization or "output" practice, forcing you to speak before you're ready. They're based on now-disproven 19th century theories of how we learn languages. Unlimited English is for adults who enjoy practical English in lessons that are funny and interesting. We use current scientific research to create real, interesting English at a level you can actually understand. The result is the ability to listen to fast English and to speak more fluently.</div>
</details>
<details class="ep-faq-item">
<summary data-i18n="faq6_q">Is this only for American English?</summary>
<div class="ep-faq-item-body" data-i18n="faq6_a">The hosts are American, so the accent is American English. But the vocabulary, the grammar, and the method work for any English you'll need, whether it is British, Australian, or international business English. The way clear English works is the same everywhere.</div>
</details>
<details class="ep-faq-item">
<summary data-i18n="faq7_q">Can I download lessons for offline use?</summary>
<div class="ep-faq-item-body" data-i18n="faq7_a">Yes. Every lesson can be downloaded to our iOS or Android phone app so you can listen on a plane, on the subway, or anywhere your phone has no signal.</div>
</details>
</div>
<div class="ep-faq-cta">
<a class="ep-textlink" data-i18n="faq_cta" href="https://www.eslpod.com/faqs/">Read all FAQs →</a>
</div>
</div>
</section>
<!-- ╭─── 11. FOOTER (Foundation §5.1) ────────────────────────────╮ -->
<footer id="ep-footer">
<div class="ep-wrap-wide">
<div class="ep-footer-grid">
<div class="ep-footer-brand">
<a aria-label="ESLPod home" class="ep-logo" data-i18n-aria="nav_home_aria" href="/">
<span style="display:inline-block;font-family:-apple-system,BlinkMacSystemFont,Helvetica,Arial,sans-serif;font-size:17px;font-weight:900;color:#fff;letter-spacing:-0.5px;line-height:30px">ESLPod<span style="color:var(--ep-green2)">.com</span></span>
</a>
<p class="ep-footer-tagline" data-i18n="foot_tagline">Bringing science to language learning since 2005</p>
<small data-i18n="foot_byline">By Dr. Jeff McQuillan &amp; Dr. Lucy Tse · Los Angeles</small>
</div>
<div class="ep-footer-col">
<h4 data-i18n="foot_learn">Learn</h4>
<ul>
<li><a data-i18n="foot_ue" href="https://www.eslpod.com/unlimited-english/">Unlimited English</a></li>
<li><a data-i18n="foot_podcast" href="https://www.eslpod.com/podcast/">Apple Podcasts</a></li>
<li><a class="popup-trigger" data-i18n="cta_trigger" data-modal="#leadmagnet-modal" href="https://www.eslpod.com/free">Get a free lesson</a></li>
<li><a data-i18n="foot_listen" href="https://www.eslpod.com/listen/">Listen Free</a></li>
<li><a data-i18n="foot_collections" href="https://tv.eslpod.com/l/products?sortKey=recommended&amp;sortDirection=asc&amp;page=1&amp;courseCategory=Special+Courses">Special Courses</a></li>
</ul>
</div>
<div class="ep-footer-col">
<h4 data-i18n="foot_about">About</h4>
<ul>
<li><a data-i18n="foot_whoweare" href="https://www.eslpod.com/about/">Who we are</a></li>
<li><a data-i18n="foot_faqs" href="https://www.eslpod.com/faqs/">FAQs</a></li>
<li><a data-i18n="foot_themethod" href="https://www.eslpod.com/method/">The Method</a></li>
<li><a data-i18n="foot_method" href="https://www.eslpod.com/research/">The Science</a></li>
<li><a data-i18n="foot_library" href="https://www.eslpod.com/library">Lesson Library</a></li>
<li><a data-i18n="foot_blog" href="https://www.eslpod.com/eslpod_blog/">Blog</a></li>
</ul>
</div>
<div class="ep-footer-col">
<h4 data-i18n="foot_connect">Connect</h4>
<ul>
<li><a data-i18n="foot_youtube" href="https://www.youtube.com/user/eslpod">YouTube</a></li>
<li><a data-i18n="foot_x" href="https://www.x.com/eslpod">X (Twitter)</a></li>
<li><a data-i18n="foot_facebook" href="https://www.facebook.com/eslpod/">Facebook</a></li>
<li><a data-i18n="foot_instagram" href="https://www.instagram.com/eslpod/">Instagram</a></li>
<li><a data-i18n="foot_rss" href="https://www.eslpod.com/feed.xml">RSS</a></li>
<li><a data-i18n="foot_contact" href="https://tv.eslpod.com/p/contactus">Contact</a></li>
</ul>
</div>
</div>
<div class="ep-footer-trust">
<span>🔒 <span data-i18n="foot_trust1">Cancel anytime</span></span>
<span>🛡️ <span data-i18n="foot_trust2">30-day money-back guarantee on every course</span></span>
</div>
<div class="ep-footer-legal">
<span data-i18n="foot_copy">© 2026 ESLPod.com</span>
<div class="ep-footer-legal-links">
<a data-i18n="foot_privacy" href="https://tv.eslpod.com/p/privacy">Privacy</a>
<a data-i18n="foot_terms" href="https://tv.eslpod.com/p/terms/">Terms</a>
</div>
</div>
</div>
</footer>
<!-- ╭─── Sticky bottom mobile CTA ────────────────────────────────╮ -->
<div class="ep-mobile-stickycta" id="ep-mobile-stickycta">
<!-- v40: sticky bar swaps with the trial state, same pattern as the nav.
       The bar itself stays one CTA tall either way. -->
<a class="popup-trigger ep-notrial-only" data-i18n="mobile_sticky" data-modal="#leadmagnet-modal" href="#">Get a free lesson</a>
<a class="ep-trial-only" data-i18n="mobile_sticky_trial" data-track="cta_click_trial" href="/unlimited-english/#ep-pricing">Start My Free 7-Day Trial</a>
</div>
</div><!-- /#ep-page -->
<!-- ===== LEAD-MAGNET MODAL (verbatim from index-1.html) ===== -->
<div aria-hidden="true" aria-modal="true" class="lm-overlay" id="leadmagnet-modal" role="dialog">
<div class="lm-modal" role="document">
<button aria-label="Close" class="lm-x" data-i18n-aria="close_aria">×</button>
<div class="lm-grid">
<div class="lm-body" id="lm-body">
<div class="lm-badge" data-i18n="badge">Trusted by 15 million students in 189 countries</div>
<h2 class="popup-title" data-i18n="title">5 Phrases That Make You Easier to Understand in English</h2>
<p class="popup-body" data-i18n="body">A 10-minute audio lesson with the five exact phrases you can use when someone doesn’t understand what you said in English</p>
<form class="lm-form" id="lm-form">
<div class="full">
<label data-i18n="label_name" for="lm-name">Full Name *</label>
<input class="lm-input" data-i18n-placeholder="placeholder_name" id="lm-name" name="name" placeholder="First + Last Name" required="" type="text"/>
</div>
<div class="full">
<label data-i18n="label_email" for="lm-email">Email *</label>
<input class="lm-input" data-i18n-placeholder="placeholder_email" id="lm-email" name="email" placeholder="Email" required="" type="email"/>
<div class="lm-secure"><span class="lm-secure-icon">🔒</span><span data-i18n="secure_microcopy">This form is secure. We never share your email.</span></div>
</div>
<div class="full">
<label data-i18n="label_lang" for="language-select">Your Language *</label>
<select class="lm-select" id="language-select" name="language" required=""></select>
</div>
<!-- Hidden meta -->
<input id="browser_lang" name="browser_lang" type="hidden"/>
<input id="page_url" name="page_url" type="hidden" value=""/>
<input id="ga_client_id" name="ga_client_id" type="hidden" value=""/>
<input id="utm_source" name="utm_source" type="hidden" value=""/>
<input id="utm_medium" name="utm_medium" type="hidden" value=""/>
<input id="utm_campaign" name="utm_campaign" type="hidden" value=""/>
<input id="utm_content" name="utm_content" type="hidden" value=""/>
<input id="utm_term" name="utm_term" type="hidden" value=""/>
<input id="gclid" name="gclid" type="hidden" value=""/>
<input id="fbclid" name="fbclid" type="hidden" value=""/>
<input id="fbp" name="fbp" type="hidden" value=""/>
<input id="fbc" name="fbc" type="hidden" value=""/>
<input id="event_id" name="event_id" type="hidden" value=""/>
<input id="referrer" name="referrer" type="hidden" value=""/>
<div class="full">
<button class="popup-cta" data-i18n="cta" type="submit">Get Your Free Lesson</button>
<p class="popup-disclaimer" data-i18n="privacy">Your email stays private • No spam</p>
</div>
</form>
</div>
</div>
</div>
</div>
<!-- ===== ESLPod Popup Runtime (verbatim from index-1.html) ===== -->
<script id="eslpod-popup-js">
(function(){
  'use strict';
  // v35: the v6 Worker. This is the ONLY Worker that applies AC_FIXED_TAGS
  // ['lm-5-phrases','UEM-SL-NonApp','5Phrases','Popup'], which is what the
  // Active Campaign automation triggers on. v34 posted to the legacy v5
  // Worker, so no submission from this page could ever enter that automation.
  var POPUP_SECRET   = '8Qv3mKpR9xN2sJ5wL4hT6bY8cZ1fG0dM7v4nE9aJk2PuB3yX5tHrW6Mq8Vc5sZb1';
  var WORKER_BASE    = 'https://eslpod-analytics-v6.titleman.workers.dev';

  // v56, 2026-08-06: SWAPPED TO /lgm/, on Jeff's instruction. The card-gated
  // thank-you page moves into /lgm/ at the same moment this file goes up, so
  // the destination and the page land together and there is no window in which
  // a signup reaches the retired $1 offer. The interim path is gone from this
  // script element, and the console guard that string-matched it is gone too.
  // The only surviving mentions of the old path are dated changelog entries in
  // <head>, which are history and should stay.
  var LGM_BASE       = 'https://www.eslpod.com/lgm/lgm.html';
/* ESLPod EXP-010 — post-signup destination. Assignment, stickiness, exposure.  v2.
 *
 * v2 fixes the three launch blockers found in the 2026-08-31 re-assessment. None of v1 ever
 * shipped, so nothing live carried them:
 *
 *   1. LOCALISATION. v1 sent every arm-B visitor to the ENGLISH sales page. The sales page
 *      localises by PATH (content/<locale>/unlimited-english/ — 17 per-locale dirs in the
 *      UE-Sales v9 tarball), not by query param, and ~40% of submits come from localised
 *      pages, so v1 would have biased the test against B. exp010SalesURL(lang) below maps
 *      the popup's effective language to the localised sales page. The dropdown's codes are
 *      lowercase (zh-hans) while the site's dirs are zh-Hans/zh-Hant — the map carries the
 *      exact dir casing because the host is case-sensitive.
 *   2. THE KILL SWITCH NOW UN-STICKS. v1 read the sticky cookie BEFORE the EXP010_LIVE
 *      gate, so flipping inert kept sending already-assigned visitors to arm B for up to
 *      180 days. The gate is now the first thing exp010Arm() checks.
 *   3. ONE ARM PER SUBMIT, EVEN WITH COOKIES BLOCKED. v1 relied on the cookie for
 *      stickiness, and document.cookie does not throw when blocked — so exp010Arm()'s two
 *      calls per submit (event, then URL) could flip different coins and a visitor could be
 *      LOGGED in one arm and SENT to the other. The arm is now memoised in a closure
 *      variable before any cookie I/O, so the two calls cannot disagree.
 *
 * Ships inside the Home page's popup script, immediately after the LGM_BASE constant, and is
 * wired into three existing lines by LandingPages_Home_EXP010_patch_v2.py. Tested whole by
 * Analytics_EXP010-Destination_v2_tests_v1.mjs against a stubbed browser.
 *
 * ─────────────────────────────────────────────────────────────────────────────────────
 * THE QUESTION. A visitor who submits the lead-magnet popup is sent straight to the lesson
 * at /lgm/lgm.html, and the sales page is something they must then click through to. Does
 * showing the sales page FIRST produce more trials, or does it spend the goodwill of a
 * promise just kept?
 *
 * ARM B DOES NOT BREAK THE PROMISE. The visitor asked for "5 Phrases That Make You Easier to
 * Understand in English", and arm B does not show it to them immediately. It is still
 * delivered: AC automation 704 mails the lesson link to every subscriber
 * (`https://www.eslpod.com/lgm/lgm.html?lang=…&ace=…`, verified in
 * Analytics_AC-06-5Phrases-Link-Edit-List_v2.md by a real test send on 27 Aug). Arm B changes
 * the order in which the two pages are seen, not whether the lesson is given. If 704 ever
 * stops carrying that link, THIS EXPERIMENT MUST BE STOPPED, because arm B would then be
 * taking an email address for a lesson it never delivers.
 *
 * ─────────────────────────────────────────────────────────────────────────────────────
 * WHY ASSIGNMENT HAPPENS AT SUBMIT, NOT ON PAGE LOAD
 *
 * The population is "people who submitted the popup" and nobody else. Randomising on page
 * load would put thousands of visitors who never submit into the cookie, and every one of
 * them would be a null row in the denominator. Assigning lazily at the moment of submit makes
 * the exposed set exactly the analysed set.
 *
 * WHY THE EXPOSURE RIDES `lead_magnet_submit` RATHER THAN ITS OWN EVENT
 *
 * Both arms navigate away immediately, so a fresh `experiment_view` pushed just before
 * `location.href` races the page teardown. EXP-009 solved that by firing on the landing page;
 * here the two arms land on DIFFERENT pages, so firing on arrival would mean instrumenting
 * two destinations and any difference between them would bias the result.
 *
 * Instead the arm is stamped onto `lead_magnet_submit`, an event that already fires at exactly
 * this moment and that demonstrably survives the navigation — GA4 recorded 258 of them in the
 * 18 days from 11 Aug. The denominator is therefore an event with a measured delivery rate,
 * not a new one we would be hoping about. `experiment_view` is pushed in the same call for
 * consistency with the registry's convention; if it is lost to the teardown, nothing depends
 * on it.
 *
 * The arm ALSO goes into the POST body as `exp`, so the Worker can persist it beside the lead
 * in D1 and the trial join can be made server-side. The Worker ignores unknown form fields
 * today, so shipping this before the Worker reads it is safe and costs nothing.
 */
  var SALES_BASE = 'https://www.eslpod.com/unlimited-english/';
  var EXP010 = 'EXP-010';
  var EXP010_COOKIE = 'esl_exp';
  var EXP010_COOKIE_DAYS = 180;

  // False = inert: everyone gets arm A, which is today's behaviour exactly, and the arm is
  // still reported so the baseline is measured on the same code path the live test will use.
  // Flip in LandingPages_Home_EXP010_patch_v2.py to launch.
  var EXP010_LIVE = false;   // set by the patch script: INERT

  /* Arm B's destination, localised the way the sales page actually localises: by PATH.
     The keys are the popup's effective-language codes (lowercase, from the dropdown /
     browser detection); the values are the site's directory names, whose casing matters —
     the two Chinese dirs are zh-Hans and zh-Hant on a case-sensitive host. Anything not in
     the map ('en', '', an unexpected code) gets the English sales page, which is also what
     the lesson page does with an unknown ?lang=. */
  var EXP010_SALES_DIRS = {
    'ar': 'ar', 'de': 'de', 'es': 'es', 'fr': 'fr', 'id': 'id', 'it': 'it',
    'ja': 'ja', 'ko': 'ko', 'pl': 'pl', 'pt': 'pt', 'ru': 'ru', 'th': 'th',
    'tr': 'tr', 'vi': 'vi', 'zh-hans': 'zh-Hans', 'zh-hant': 'zh-Hant'
  };
  function exp010SalesURL(lang) {
    var dir = EXP010_SALES_DIRS[String(lang || '').toLowerCase()];
    return dir ? 'https://www.eslpod.com/' + dir + '/unlimited-english/' : SALES_BASE;
  }

  /* The cookie is shared with the head framework and holds every experiment the visitor is
     in — "EXP-009:A|EXP-010:B". These helpers must round-trip that format EXACTLY, or
     assigning EXP-010 would evict a live EXP-009 assignment and silently re-randomise people
     mid-test. The test suite asserts the interop in both directions rather than trusting it. */
  function exp010ReadArm(exp) {
    try {
      var m = document.cookie.match(new RegExp('(?:^|;\\s*)' + EXP010_COOKIE + '=([^;]*)'));
      if (!m) return '';
      var parts = decodeURIComponent(m[1]).split('|');
      for (var i = 0; i < parts.length; i++) {
        var kv = parts[i].split(':');
        if (kv[0] === exp) return (kv[1] === 'A' || kv[1] === 'B') ? kv[1] : '';
      }
    } catch (e) {}
    return '';
  }

  function exp010WriteArm(exp, arm) {
    try {
      var m = document.cookie.match(new RegExp('(?:^|;\\s*)' + EXP010_COOKIE + '=([^;]*)'));
      var parts = m ? decodeURIComponent(m[1]).split('|') : [];
      var out = [];
      for (var i = 0; i < parts.length; i++) {
        // Keep every OTHER experiment untouched. This is the line that protects EXP-009.
        if (parts[i] && parts[i].split(':')[0] !== exp) out.push(parts[i]);
      }
      out.push(exp + ':' + arm);
      var d = new Date();
      d.setTime(d.getTime() + EXP010_COOKIE_DAYS * 864e5);
      document.cookie = EXP010_COOKIE + '=' + encodeURIComponent(out.join('|')) +
        ';expires=' + d.toUTCString() + ';path=/;SameSite=Lax;Secure';
      return true;
    } catch (e) { return false; }
  }

  /* One arm per page life, then sticky across visits, then a coin flip. Returns 'A' or 'B'
     and never anything else, because the destination switch treats anything that is not 'B'
     as the control.

     ORDER IS LOAD-BEARING, twice over:
       - The memo comes first. exp010Arm() runs twice per submit (event, then URL), and
         document.cookie does not throw when blocked — without the memo those two calls
         could flip different coins and log one arm while navigating to the other.
       - The EXP010_LIVE gate comes BEFORE the sticky read. It is the kill switch: when the
         patch ships inert, an assignment stored by an earlier live build must NOT keep
         routing that visitor to arm B. v1 had these reversed and the switch could not
         un-stick anyone for up to 180 days. */
  var exp010Assigned = '';
  function exp010Arm() {
    if (exp010Assigned) return exp010Assigned;
    if (!EXP010_LIVE) { exp010Assigned = 'A'; return 'A'; }
    var existing = exp010ReadArm(EXP010);
    if (existing) { exp010Assigned = existing; return existing; }
    var arm = Math.random() < 0.5 ? 'A' : 'B';
    exp010WriteArm(EXP010, arm);
    exp010Assigned = arm;
    return arm;
  }

  // v42: destination corrected to the ROOT signup.html. v41 used /lgm/signup.html
  // on Jeff's answer, and his own .htaccess later showed the rewrite
  // "^signup/?$ -> /signup.html" at the site root, with no signup folder and no
  // copy under /lgm/. Jeff confirmed: eslpod.com/signup.html is the page.
  // That page localises itself, so the ?lang= appended below still matters.
  // Only used if a blocker suppresses the
  // modal. Published on window, because the blocker-fallback script lower down
  // runs in its own IIFE and cannot see this scope. Before v40 that script
  // carried its own hard-coded copy of this URL and this constant was never
  // read by anything, so changing it here did nothing.
  var FALLBACK_URL   = 'https://www.eslpod.com/signup.html';
  window.ESLPOD_FALLBACK_URL = FALLBACK_URL;

  var LANGS = [
    {code:'ar',      name:'Arabic | العربية'},
    {code:'zh-hans', name:'Chinese (Simplified) | 中文（简体）'},
    {code:'zh-hant', name:'Chinese (Traditional) | 中文（繁體）'},
    {code:'fr',      name:'French | Français'},
    {code:'de',      name:'German | Deutsch'},
    {code:'id',      name:'Indonesian | Bahasa Indonesia'},
    {code:'it',      name:'Italian | Italiano'},
    {code:'ja',      name:'Japanese | 日本語'},
    {code:'ko',      name:'Korean | 한국어'},
    {code:'pl',      name:'Polish | Polski'},
    {code:'pt',      name:'Portuguese | Português'},
    {code:'ru',      name:'Russian | Русский'},
    {code:'es',      name:'Spanish | Español'},
    {code:'th',      name:'Thai | ไทย'},
    {code:'tr',      name:'Turkish | Türkçe'},
    {code:'vi',      name:'Vietnamese | Tiếng Việt'},
    {code:'other',   name:'Other'}
  ];
  var RTL_LANGS = {ar:1};

  var STRINGS = {
    badge:{"en":"Trusted by 15 million students in 189 countries","es":"15 millones de estudiantes en 189 países confían en nosotros","pt":"Com a confiança de 15 milhões de alunos em 189 países","fr":"15 millions d'étudiants dans 189 pays nous font confiance","de":"15 Millionen Englischlerner in 189 Ländern vertrauen uns","it":"15 milioni di studenti in 189 Paesi si fidano di noi","ja":"189カ国、1,500万人の学習者に選ばれています","ko":"189개국 1,500만 명의 학습자가 선택했습니다","ru":"Нам доверяют 15 миллионов учеников в 189 странах","tr":"189 ülkede 15 milyon öğrenci güveniyor","vi":"15 triệu học viên ở 189 quốc gia tin dùng","zh-hans":"189个国家的1500万名学员都在学","zh-hant":"189個國家、1,500萬名學員的選擇","ar":"يثق بنا 15 مليون طالب في 189 دولة","id":"15 juta siswa di 189 negara memercayai kami","pl":"Zaufało nam 15 milionów uczących się w 189 krajach","th":"ผู้เรียน 15 ล้านคนใน 189 ประเทศไว้วางใจ"},
    title:{"en":"5 Phrases That Make You Easier to Understand in English","es":"5 frases para que te entiendan mejor en inglés","pt":"5 frases que deixam o seu inglês mais fácil de entender","fr":"5 phrases qui vous rendent plus facile à comprendre en anglais","de":"5 Sätze, mit denen Sie auf Englisch leichter verstanden werden","it":"5 frasi che La rendono più facile da capire in inglese","ja":"英語がぐっと伝わりやすくなる5つのフレーズ","ko":"영어로 말할 때 상대가 더 쉽게 알아듣게 해 주는 5가지 표현","ru":"5 фраз, благодаря которым Вас будут лучше понимать по-английски","tr":"İngilizcede Sizi Daha Anlaşılır Kılan 5 Kalıp","vi":"5 câu nói giúp người khác hiểu bạn dễ hơn khi bạn nói tiếng Anh","zh-hans":"让别人更容易听懂你英语的5句话","zh-hant":"5句話，讓別人更容易聽懂你的英文","ar":"5 عبارات تجعل الآخرين يفهمونكم بالإنجليزية بسهولة أكبر","id":"5 Frasa agar Orang Lebih Mudah Memahami Bahasa Inggris Anda","pl":"5 zwrotów, dzięki którym łatwiej Państwa zrozumieć po angielsku","th":"5 ประโยคที่ทำให้คนเข้าใจภาษาอังกฤษของคุณง่ายขึ้น"},
    body:{"en":"A 10-minute audio lesson with the five exact phrases you can use when someone doesn’t understand what you said in English","es":"Una lección en audio de 10 minutos con las cinco frases exactas que puedes usar cuando alguien no entiende lo que dijiste en inglés","pt":"Uma lição em áudio de 10 minutos com as cinco frases exatas que você pode usar quando alguém não entende o que você disse em inglês","fr":"Une leçon audio de 10 minutes avec les cinq phrases exactes à utiliser quand quelqu'un ne comprend pas ce que vous venez de dire en anglais","de":"Eine 10-minütige Audio-Lektion mit den fünf Sätzen, die Sie sagen können, wenn jemand Ihr Englisch nicht versteht","it":"Una lezione audio di 10 minuti con le cinque frasi esatte da usare quando qualcuno non capisce ciò che Lei ha detto in inglese","ja":"英語が相手に通じなかったとき、そのまま使える5つのフレーズ。10分の音声レッスンでお届けします","ko":"상대가 내 영어를 알아듣지 못했을 때 바로 쓸 수 있는 다섯 가지 표현을 담은 10분짜리 오디오 레슨","ru":"10-минутный аудиоурок с пятью точными фразами, которые помогут Вам, когда собеседник не понял, что Вы сказали по-английски","tr":"İngilizce söylediğinizi karşınızdaki anlamadığında kullanabileceğiniz beş kalıbı veren 10 dakikalık ses dersi","vi":"Một bài học nghe 10 phút, với đúng năm câu bạn dùng được khi ai đó không hiểu điều bạn vừa nói bằng tiếng Anh","zh-hans":"一节10分钟的音频课，给你五句现成的话，专门用在别人听不懂你说的英语的时候","zh-hant":"一堂10分鐘的音訊課程，教你五句話，專門用在對方聽不懂你的英文的時候","ar":"درس صوتي من 10 دقائق يضم العبارات الخمس التي تستعملونها حين لا يفهم أحدهم ما قلتموه بالإنجليزية","id":"Pelajaran audio 10 menit berisi lima frasa yang bisa Anda pakai ketika lawan bicara tidak menangkap ucapan Anda dalam bahasa Inggris","pl":"10-minutowa lekcja audio z pięcioma konkretnymi zwrotami na sytuację, gdy ktoś nie rozumie tego, co Państwo powiedzieli po angielsku","th":"บทเรียนเสียง 10 นาที พร้อมห้าประโยคที่ใช้ได้ทันทีเมื่อมีคนไม่เข้าใจสิ่งที่คุณพูดเป็นภาษาอังกฤษ"},
    label_name:{"en":"Full Name *","es":"Nombre completo *","pt":"Nome completo *","fr":"Nom complet *","de":"Vollständiger Name *","it":"Nome e cognome *","ja":"お名前（フルネーム） *","ko":"성명 *","ru":"Имя и фамилия *","tr":"Ad Soyad *","vi":"Họ và tên *","zh-hans":"姓名 *","zh-hant":"姓名 *","ar":"الاسم الكامل *","id":"Nama Lengkap *","pl":"Imię i nazwisko *","th":"ชื่อ-นามสกุล *"},
    placeholder_name:{"en":"First + Last Name","es":"Nombre y apellido","pt":"Nome + Sobrenome","fr":"Prénom + Nom","de":"Vor- + Nachname","it":"Nome + Cognome","ja":"姓 + 名","ko":"성과 이름","ru":"Имя и фамилия","tr":"Ad + Soyad","vi":"Họ + Tên","zh-hans":"姓名","zh-hant":"姓名","ar":"الاسم الأول + اسم العائلة","id":"Nama Depan + Nama Belakang","pl":"Imię i nazwisko","th":"ชื่อ + นามสกุล"},
    label_email:{"en":"Email *","es":"Correo electrónico *","pt":"E-mail *","fr":"E-mail *","de":"E-Mail *","it":"Email *","ja":"メールアドレス *","ko":"이메일 *","ru":"Электронная почта *","tr":"E-posta *","vi":"Email *","zh-hans":"邮箱 *","zh-hant":"電子郵件 *","ar":"البريد الإلكتروني *","id":"Email *","pl":"E-mail *","th":"อีเมล *"},
    placeholder_email:{"en":"Email","es":"Correo electrónico","pt":"E-mail","fr":"E-mail","de":"E-Mail","it":"Email","ja":"メールアドレス","ko":"이메일","ru":"Электронная почта","tr":"E-posta","vi":"Email","zh-hans":"邮箱","zh-hant":"電子郵件","ar":"البريد الإلكتروني","id":"Email","pl":"E-mail","th":"อีเมล"},
    secure_microcopy:{"en":"This form is secure. We never share your email.","es":"Este formulario es seguro. Nunca compartimos tu correo con nadie.","pt":"Este formulário é seguro. Nós nunca compartilhamos o seu e-mail.","fr":"Ce formulaire est sécurisé. Nous ne communiquons jamais votre e-mail.","de":"Dieses Formular ist sicher. Wir geben Ihre E-Mail nie weiter.","it":"Questo modulo è sicuro. Non comunichiamo mai la Sua email a nessuno.","ja":"このフォームは安全です。メールアドレスを第三者に渡すことは決してありません。","ko":"이 양식은 안전합니다. 이메일 주소를 다른 곳에 넘기는 일은 절대 없습니다.","ru":"Форма защищена. Мы никогда не передаём Ваш адрес третьим лицам.","tr":"Bu form güvenlidir. E-postanızı asla paylaşmayız.","vi":"Biểu mẫu này an toàn. Chúng tôi không bao giờ chia sẻ email của bạn.","zh-hans":"这个表单是安全的。我们绝不会把你的邮箱给别人。","zh-hant":"這份表單是安全的。我們絕不會把你的電子郵件給別人。","ar":"هذا النموذج آمن. لا نشارك بريدكم الإلكتروني أبدًا.","id":"Formulir ini aman. Kami tidak pernah membagikan email Anda.","pl":"Formularz jest bezpieczny. Nie udostępniamy adresów e-mail.","th":"แบบฟอร์มนี้ปลอดภัย เราไม่เปิดเผยอีเมลของคุณให้ใคร"},
    label_lang:{"en":"Your Language *","es":"Tu idioma *","pt":"Seu idioma *","fr":"Votre langue *","de":"Ihre Sprache *","it":"La Sua lingua *","ja":"お使いの言語 *","ko":"사용 언어 *","ru":"Ваш язык *","tr":"Diliniz *","vi":"Ngôn ngữ của bạn *","zh-hans":"你的语言 *","zh-hant":"你的語言 *","ar":"لغتكم *","id":"Bahasa Anda *","pl":"Państwa język *","th":"ภาษาของคุณ *"},
    lang_placeholder:{"en":"— Select your language —","es":"— Selecciona tu idioma —","pt":"— Selecione seu idioma —","fr":"— Sélectionnez votre langue —","de":"— Sprache wählen —","it":"— Selezioni la sua lingua —","ja":"— 言語を選択 —","ko":"— 언어 선택 —","ru":"— Выберите язык —","tr":"— Dilinizi seçin —","vi":"— Chọn ngôn ngữ —","zh-hans":"— 选择您的语言 —","zh-hant":"— 選擇您的語言 —","ar":"— اختر لغتك —","id":"— Pilih bahasa Anda —","pl":"— Wybierz swój język —","th":"— เลือกภาษาของคุณ —"},
    cta:{"en":"Get Your Free Lesson","es":"Recibe tu lección gratis","pt":"Receba sua lição grátis","fr":"Recevoir ma leçon gratuite","de":"Holen Sie sich Ihre kostenlose Lektion","it":"Riceva la Sua lezione gratuita","ja":"無料レッスンを受け取る","ko":"무료 레슨 받기","ru":"Получить бесплатный урок","tr":"Ücretsiz Dersinizi Alın","vi":"Nhận bài học miễn phí của bạn","zh-hans":"领取你的免费课程","zh-hant":"領取你的免費課程","ar":"احصلوا على درسكم المجاني","id":"Dapatkan Pelajaran Gratis Anda","pl":"Odbierz darmową lekcję","th":"รับบทเรียนฟรีของคุณ"},
    privacy:{"en":"Your email stays private • No spam","es":"Tu correo se queda en privado • Nada de spam","pt":"Seu e-mail fica em sigilo • Sem spam","fr":"Votre e-mail reste privé • Aucun spam","de":"Ihre E-Mail bleibt privat • Kein Spam","it":"La Sua email resta privata • Niente spam","ja":"メールアドレスは他社に渡しません • 迷惑メールは一切なし","ko":"이메일 주소는 공개하지 않습니다 • 스팸 없음","ru":"Ваш адрес остаётся конфиденциальным • Без спама","tr":"E-postanız gizli kalır • Spam yok","vi":"Email của bạn được giữ kín • Không spam","zh-hans":"你的邮箱不会外泄 • 不发垃圾邮件","zh-hant":"你的電子郵件不會外流 • 不寄垃圾信","ar":"بريدكم يبقى خاصًا • بلا رسائل مزعجة","id":"Email Anda tetap pribadi • Tanpa spam","pl":"Adres e-mail zostaje u nas • Bez spamu","th":"อีเมลของคุณเป็นความลับ • ไม่มีสแปม"},
    close_aria:{"en":"Close","es":"Cerrar","pt":"Fechar","fr":"Fermer","de":"Schließen","it":"Chiudi","ja":"閉じる","ko":"닫기","ru":"Закрыть","tr":"Kapat","vi":"Đóng","zh-hans":"关闭","zh-hant":"關閉","ar":"إغلاق","id":"Tutup","pl":"Zamknij","th":"ปิด"},
    cta_trigger:{"en":"Get a free lesson","es":"Recibe una lección gratis","pt":"Receba uma lição grátis","fr":"Recevoir une leçon gratuite","de":"Holen Sie sich eine kostenlose Lektion","it":"Riceva una lezione gratuita","ja":"無料レッスンを受け取る","ko":"무료 레슨 받기","ru":"Получить бесплатный урок","tr":"Ücretsiz ders alın","vi":"Nhận bài học miễn phí","zh-hans":"领取免费课程","zh-hant":"領取免費課程","ar":"احصلوا على درس مجاني","id":"Dapatkan pelajaran gratis","pl":"Odbierz darmową lekcję","th":"รับบทเรียนฟรี"},
    fallback_link:{"en":"Having trouble? Click here instead.","es":"¿Tienes problemas? Haz clic aquí.","pt":"Problemas? Clique aqui.","fr":"Des problèmes ? Cliquez ici.","de":"Probleme? Klicken Sie hier.","it":"Hai problemi? Clicca qui.","ja":"問題がありますか？こちらをクリック。","ko":"문제가 있나요? 여기를 클릭하세요.","ru":"Проблемы? Нажмите здесь.","tr":"Sorun mu yaşıyorsunuz? Buraya tıklayın.","vi":"Gặp sự cố? Nhấp vào đây.","zh-hans":"遇到问题？点击这里。","zh-hant":"遇到問題？點擊這裡。","ar":"هل تواجه مشكلة؟ انقر هنا بدلاً من ذلك.","id":"Ada masalah? Klik di sini.","pl":"Masz problem? Kliknij tutaj.","th":"มีปัญหาใช่ไหม คลิกที่นี่"}
  };

  function $(s,r){return (r||document).querySelector(s);}
  function $$(s,r){return Array.prototype.slice.call((r||document).querySelectorAll(s));}
  function getCookie(n){var m=document.cookie.match('(^|;)\\s*'+n+'\\s*=\\s*([^;]+)');return m?m.pop():'';}
  function getGAClientId(){var raw=getCookie('_ga');if(!raw)return '';var p=raw.split('.');return p.length>=4?p.slice(2).join('.'):raw;}
  function getQueryParam(n){var u=new URL(window.location.href);return u.searchParams.get(n)||'';}

  // v81: campaign parameters survive navigation INSIDE one visit.
  //
  // getQueryParam above reads the current URL and nothing else. Measured on the
  // live site: none of the internal research links carry utm_* forward, so a
  // reader who arrived from a DM on the hub and then clicked into an article
  // signed up with no campaign at all — the Worker wrote src:research and
  // neither kw: nor utm_src:, which is indistinguishable from an organic visit.
  //
  // ATOMIC: a URL carrying any campaign parameter replaces the WHOLE stored set,
  // so a later link with only utm_source= cannot inherit an earlier link's
  // utm_content and invent a campaign nobody ran. A URL carrying none leaves the
  // stored set alone; that is the case this exists for.
  //
  // SESSION-SCOPED: sessionStorage, not a cookie. These last for this tab and
  // this visit and are gone when it closes, so a signup days later is never
  // credited to a DM.
  //
  // Every access is wrapped. sessionStorage THROWS in Safari private mode, and a
  // signup must not die because the reader is browsing privately.
  var CAMPAIGN_KEYS = ['utm_source','utm_medium','utm_campaign','utm_content','utm_term','fbclid','gclid'];
  var CAMPAIGN_STORE = 'eslpod_campaign';
  var _campaignMemo = null;

  function campaign(){
    if (_campaignMemo) return _campaignMemo;
    var fresh = {}, any = false;
    CAMPAIGN_KEYS.forEach(function(k){
      var v = getQueryParam(k);
      if (v) { fresh[k] = v; any = true; }
    });
    if (any) {
      try { sessionStorage.setItem(CAMPAIGN_STORE, JSON.stringify(fresh)); } catch(e){}
      _campaignMemo = fresh;
      return _campaignMemo;
    }
    var stored = {};
    try {
      var raw = sessionStorage.getItem(CAMPAIGN_STORE);
      if (raw) {
        var parsed = JSON.parse(raw);
        // Only keys we recognise, and only strings. Anything else in there was
        // not put there by this code.
        if (parsed && typeof parsed === 'object') {
          CAMPAIGN_KEYS.forEach(function(k){
            if (typeof parsed[k] === 'string' && parsed[k]) stored[k] = parsed[k];
          });
        }
      }
    } catch(e){}
    _campaignMemo = stored;
    return _campaignMemo;
  }

  function campaignParam(n){ return campaign()[n] || ''; }
  function detectBrowserLang(){return (navigator.language||navigator.userLanguage||'').toLowerCase();}

  function normalizeBrowserLang(raw){
    raw=(raw||'').toLowerCase();if(!raw)return '';
    if(raw.indexOf('ar')===0)return 'ar';
    if(raw.indexOf('fr')===0)return 'fr';
    if(raw.indexOf('de')===0)return 'de';
    if(raw.indexOf('it')===0)return 'it';
    if(raw.indexOf('ja')===0)return 'ja';
    if(raw.indexOf('ko')===0)return 'ko';
    if(raw.indexOf('pt')===0)return 'pt';
    if(raw.indexOf('es')===0)return 'es';
    if(raw.indexOf('tr')===0)return 'tr';
    if(raw.indexOf('vi')===0)return 'vi';
    if(raw.indexOf('ru')===0)return 'ru';
    if(raw.indexOf('zh')===0)return /tw|hk|mo|hant/.test(raw)?'zh-hant':'zh-hans';
    // v57: th, id and pl were missing here since the picker was built, so those
    // three fell through to '' and the page was forced back to English — on the
    // Thai static page that meant the baked Thai CTA got overwritten in English.
    if(raw.indexOf('th')===0)return 'th';
    if(raw.indexOf('id')===0||raw.indexOf('in')===0)return 'id';  // 'in' is the legacy ISO code for Indonesian
    if(raw.indexOf('pl')===0)return 'pl';
    if(raw.indexOf('en')===0)return 'en';
    return '';
  }

  function resolveEffectiveLang(code){return (!code||code==='other')?'en':code;}

  function buildLangDropdown(preselect, fallback){
    var sel=$('#language-select');if(!sel)return;
    sel.innerHTML='';
    var ph=document.createElement('option');
    ph.value='';ph.textContent=STRINGS.lang_placeholder.en;
    ph.setAttribute('data-i18n-option','lang_placeholder');
    ph.disabled=true;
    sel.appendChild(ph);
    LANGS.forEach(function(L){var o=document.createElement('option');o.value=L.code;o.textContent=L.name;sel.appendChild(o);});
    // v64: only pre-select a code the dropdown actually offers. LANGS holds the
    // 16 localisations plus 'other' — there is no 'en' option, so an English
    // browser (or any locale normalizeBrowserLang cannot place) used to assign a
    // value with no matching <option>, which leaves selectedIndex at -1 and
    // renders the control BLANK on open. Now anything unoffered falls back to
    // the placeholder row, which is what the user should always see first.
    // 'other' is excluded on purpose: it is a choice, never a default.
    // v80: two candidates, tried in order — the page's own language, then the
    // browser's. The `offered` test below is still the ONLY thing that decides
    // what may be preselected, so neither candidate can set a value with no
    // matching <option>. 'other' is excluded from both: it is a choice, never a
    // default. When neither is offered (the English root with an English browser)
    // this lands on the placeholder, which is v64's behaviour untouched.
    var candidates = [preselect || '', fallback || ''];
    var wanted = '';
    var offered = false;
    for (var c = 0; c < candidates.length && !offered; c++) {
      var cand = candidates[c];
      if (!cand || cand === 'other') continue;
      for (var i = 0; i < LANGS.length; i++) {
        if (LANGS[i].code === cand) { wanted = cand; offered = true; break; }
      }
    }
    if (offered) { sel.value = wanted; }
    else { ph.selected = true; sel.selectedIndex = 0; }
  }

  function applyLang(lang){
    if(!lang||!STRINGS.title[lang])lang='en';
    var body=$('#lm-body');if(!body)return;
    body.setAttribute('lang',lang);
    body.setAttribute('dir',RTL_LANGS[lang]?'rtl':'ltr');
    $$('[data-i18n]',body).forEach(function(el){var k=el.getAttribute('data-i18n');if(STRINGS[k]&&STRINGS[k][lang])el.textContent=STRINGS[k][lang];});
    $$('[data-i18n-placeholder]',body).forEach(function(el){var k=el.getAttribute('data-i18n-placeholder');if(STRINGS[k]&&STRINGS[k][lang])el.setAttribute('placeholder',STRINGS[k][lang]);});
    var closeBtn=$('.lm-x',$('#leadmagnet-modal'));
    if(closeBtn){var k=closeBtn.getAttribute('data-i18n-aria');if(k&&STRINGS[k]&&STRINGS[k][lang])closeBtn.setAttribute('aria-label',STRINGS[k][lang]);}
    var phOpt=$('#language-select option[data-i18n-option="lang_placeholder"]');
    if(phOpt&&STRINGS.lang_placeholder[lang])phOpt.textContent=STRINGS.lang_placeholder[lang];
  }

  function openModal(){
    var m=$('#leadmagnet-modal');if(!m)return;m.style.display='block';m.setAttribute('aria-hidden','false');
    // GA4: lead_magnet_open
    try { (window.dataLayer = window.dataLayer || []).push({event:'lead_magnet_open'}); } catch(e){}
  }
  function closeModal(){var m=$('#leadmagnet-modal');if(!m)return;m.style.display='none';m.setAttribute('aria-hidden','true');}

  function applyPageLang(lang){
    if(!lang||!STRINGS.cta_trigger[lang])lang='en';
    var rtl=!!RTL_LANGS[lang];
    var ctaText=STRINGS.cta_trigger[lang];
    $$('.popup-trigger').forEach(function(btn){
      // Only translate triggers that opt-in via data-i18n="cta_trigger"
      if(btn.getAttribute('data-i18n')==='cta_trigger'){
        btn.textContent=ctaText;
      }
      btn.setAttribute('lang',lang);
      if(rtl)btn.setAttribute('dir','rtl');else btn.removeAttribute('dir');
    });
    var fbText=STRINGS.fallback_link[lang];
    $$('.fallback-link').forEach(function(a){
      a.textContent=fbText;
      a.setAttribute('lang',lang);
      if(rtl)a.setAttribute('dir','rtl');else a.removeAttribute('dir');
    });
    $$('a[href^="https://www.eslpod.com/signup.html"]').forEach(function(a){
      try{var u=new URL(a.href);u.searchParams.set('lang',lang);a.href=u.toString();}catch(e){}
    });
  }

  function buildRedirectURL(lang, email){
    // v35: lang is now the FIRST query param, because the thank-you page reads
    // ?lang= (lowercase, all 17 codes) to pick its language. v34 built a path
    // at /trial/<lang>/trial-<lang>-a.html, which are retired pages, and never
    // sent lang as a param at all.
    // v67: ace=<email> identifies the arrival to ActiveCampaign. Without it every
    // visitor who did not come through the AC email link is anonymous, and
    // trackcmp.net accepts then silently discards their events.
    // encodeURIComponent is load-bearing: it turns jeff+esl@gmail.com into
    // jeff%2Besl%40gmail.com, which survives the URLSearchParams read on the
    // far side. A raw + would arrive decoded as a space.
    var qs=['lang='+encodeURIComponent(lang||'en')];
    if(email)qs.push('ace='+encodeURIComponent(email));
    // v81: campaignParam, not getQueryParam — the /lgm/ arrival carries the
    // campaign even when the signup happened a click away from the landing page.
    ['utm_source','utm_medium','utm_campaign','utm_content','utm_term','fbclid','gclid'].forEach(function(k){var v=campaignParam(k);if(v)qs.push(k+'='+encodeURIComponent(v));});
    var cid=getGAClientId();if(cid)qs.push('ga_client_id='+encodeURIComponent(cid));
    // EXP-010: the query string is identical for both arms; only the base changes, so
    // a sales-page arrival carries exactly the attribution a lesson arrival carries.
    // v2: arm B goes to the LOCALISED sales page. That page localises by path, not by
    // ?lang=, so sending every arm-B visitor to the English one (v1) would have biased
    // the test against B for the ~40% of submits that come from localised pages.
    // `lang` is the popup's `effective` language — this function's only caller passes
    // resolveEffectiveLang(selected) — so it is a dropdown code, 'en', or ''.
    var exp010 = exp010Arm();
    qs.push('exp=' + encodeURIComponent('EXP-010:' + exp010));
    return (exp010 === 'B' ? exp010SalesURL(lang) : LGM_BASE)+'?'+qs.join('&');
  }

  function readCookie(name){
    try{
      var m=document.cookie.match(new RegExp('(?:^|; )'+name.replace(/[.*+?^${}()|[\]\\]/g,'\\$&')+'=([^;]*)'));
      return m?decodeURIComponent(m[1]):'';
    }catch(e){return '';}
  }

  function newEventId(){
    try{if(window.crypto&&crypto.randomUUID)return crypto.randomUUID();}catch(e){}
    return 'xxxxxxxx-xxxx-4xxx-yxxx-xxxxxxxxxxxx'.replace(/[xy]/g,function(c){
      var r=Math.random()*16|0,v=c==='x'?r:(r&0x3|0x8);return v.toString(16);
    });
  }

  function bindForm(){
    var form=$('#lm-form');if(!form)return;
    $('#browser_lang').value=detectBrowserLang();
    $('#page_url').value=window.location.href;
    $('#ga_client_id').value=getGAClientId();
    $('#referrer').value=document.referrer||'';
    // v81: campaignParam, not getQueryParam. These seven fields are what the
    // Worker turns into the src:, kw: and utm_src: tags, so this line is the
    // one that decides whether a multi-page visit is attributable at all.
    ['utm_source','utm_medium','utm_campaign','utm_content','utm_term','fbclid','gclid'].forEach(function(k){var el=document.getElementById(k);if(el)el.value=campaignParam(k);});
    $('#fbp').value=readCookie('_fbp');
    $('#fbc').value=readCookie('_fbc');

    form.addEventListener('submit',function(ev){
      ev.preventDefault();
      // v36: validate BEFORE the button is disabled, so a rejected attempt
      // leaves the CTA live and the visitor can correct and resubmit. Trim
      // first, because a field of spaces satisfies `required` untrimmed.
      var nameEl=$('#lm-name',form), emailEl=$('#lm-email',form);
      if(nameEl)nameEl.value=nameEl.value.trim();
      if(emailEl)emailEl.value=emailEl.value.trim();
      if(form.checkValidity&&!form.checkValidity()){if(form.reportValidity)form.reportValidity();return;}
      var sel=$('#language-select',form);
      var selected=(sel&&sel.value)||'';
      if(!selected){if(form.reportValidity)form.reportValidity();return;}
      var btn=$('.popup-cta',form);if(btn)btn.disabled=true;
      $('#ga_client_id').value=getGAClientId();
      $('#page_url').value=window.location.href;
      $('#fbp').value=readCookie('_fbp');
      $('#fbc').value=readCookie('_fbc');
      var eventId=newEventId();
      $('#event_id').value=eventId;

      try{
        if(typeof fbq==='function'){
          fbq('track','Lead',{
            value:0,currency:'USD',
            content_name:'5 Phrases That Make You Easier to Understand in English',
            content_category:'Lead Magnet'
          },{eventID:eventId});
        }
      }catch(e){}

      // GA4: lead_magnet_submit
      // EXP-010: the arm rides the submit event because that event already fires at
      // this exact moment and is known to survive the navigation below (258 delivered
      // in GA4 over 11-28 Aug). A fresh event pushed here would race the teardown.
      // exp010Arm() memoises its answer, so this call and the one inside
      // buildRedirectURL below cannot produce two different arms for one visitor —
      // not even when cookies are blocked, which is the v1 bug this replaces.
      var exp010Variant = exp010Arm();
      try { (window.dataLayer = window.dataLayer || []).push({event:'lead_magnet_submit', event_id:eventId, experiment_id:'EXP-010', variant:exp010Variant}); } catch(e){}
      try { (window.dataLayer = window.dataLayer || []).push({event:'experiment_view', experiment_id:'EXP-010', variant:exp010Variant}); } catch(e){}

      var effective=resolveEffectiveLang(selected);
      var body=new URLSearchParams(new FormData(form));
      body.set('language_selected',selected);
      body.set('language',effective);
      body.set('exp','EXP-010:'+exp010Variant);   // persisted by the Worker; unknown fields are ignored until it reads this
      try {
        fetch(WORKER_BASE+'/popup/'+POPUP_SECRET,{
          method:'POST',
          headers:{'Content-Type':'application/x-www-form-urlencoded;charset=UTF-8'},
          body:body.toString(),
          keepalive:true
        }).catch(function(){});  // v35: keepalive request survives the navigation
                                 // below, so a network hiccup must not surface as an
                                 // unhandled rejection in the user's console.
      } catch (e) {}
      window.location.href=buildRedirectURL(effective, (emailEl && emailEl.value) || '');
    });
  }

  function init(){
    // v57: the PAGE's own language wins. v56 read navigator.language only, so a
    // visitor with an English browser landing on the German page got an English
    // modal AND applyPageLang overwrote the baked German CTA text from the stale
    // inline STRINGS dictionary. On the static per-language pages the split
    // produces, <html lang> is the truth; the browser is only a hint on the
    // English root, where it drives the language-picker default.
    var pageLang=normalizeBrowserLang((document.documentElement.getAttribute('lang')||'').toLowerCase());
    var browserLang=normalizeBrowserLang(detectBrowserLang());
    var uiLang=(pageLang&&STRINGS.title[pageLang])?pageLang
             :((browserLang&&STRINGS.title[browserLang])?browserLang:'en');
    applyPageLang(uiLang);
    // v80: the dropdown default follows the PAGE first and the browser second.
    // v57 made <html lang> win for the interface text but left this call reading
    // navigator.language, so an English browser on a localised page preselected
    // nothing: 'en' is not one of the options and v64 correctly falls back to the
    // placeholder. buildLangDropdown now tries these in order and still refuses
    // anything LANGS does not offer, so the English root is unchanged.
    buildLangDropdown(pageLang, browserLang);
    applyLang(uiLang);
    bindForm();
    var sel=$('#language-select');
    if(sel)sel.addEventListener('change',function(){applyLang(resolveEffectiveLang(sel.value));});
    $$('.popup-trigger').forEach(function(t){t.addEventListener('click',function(e){e.preventDefault();openModal();});});
    var modal=$('#leadmagnet-modal');
    if(modal){modal.addEventListener('click',function(e){if(e.target===modal||e.target.classList.contains('lm-x'))closeModal();});}
    document.addEventListener('keydown',function(e){if(e.key==='Escape')closeModal();});
  }

  if(document.readyState==='loading'){document.addEventListener('DOMContentLoaded',init);}else{init();}
})();
</script>
<!-- ===== Page-level interactivity (nav drops, lang pill, audio, carousel, tracking, tweaks) ===== -->
<script>
(function(){
  'use strict';
  var $ = function(s, r){ return (r||document).querySelector(s); };
  var $$ = function(s, r){ return Array.prototype.slice.call((r||document).querySelectorAll(s)); };

  // ─── Nav dropdowns (click to open; close on outside click) ───
  document.addEventListener('click', function(e){
    var trigger = e.target.closest('[data-drop]');
    var openDrops = $$('.ep-nav-drop.open');
    if (trigger){
      var key = trigger.getAttribute('data-drop');
      var drop = $('[data-dropfor="'+key+'"]');
      var wasOpen = drop && drop.classList.contains('open');
      openDrops.forEach(function(d){ d.classList.remove('open'); });
      if (drop && !wasOpen) {
        drop.classList.add('open');
        trigger.setAttribute('aria-expanded','true');
      } else if (trigger) {
        trigger.setAttribute('aria-expanded','false');
      }
    } else if (!e.target.closest('.ep-nav-drop')) {
      openDrops.forEach(function(d){ d.classList.remove('open'); });
      $$('[data-drop]').forEach(function(t){ t.setAttribute('aria-expanded','false'); });
    }
  });

  // ─── Mobile drawer ───────────────────────────────────────────
  var burger = $('#ep-burger');
  var drawer = $('#ep-mobile-drawer');
  if (burger && drawer) {
    burger.addEventListener('click', function(){
      var open = drawer.classList.toggle('open');
      burger.setAttribute('aria-expanded', open ? 'true' : 'false');
      document.documentElement.classList.toggle('nav-locked', open);
    });
    drawer.querySelectorAll('a').forEach(function(a){
      a.addEventListener('click', function(){
        drawer.classList.remove('open');
        burger.setAttribute('aria-expanded','false');
        document.documentElement.classList.remove('nav-locked');
      });
    });
  }

  // ─── Top-of-page language pill ──────────────────────────────
  // ─── Language picker ───────────────────────────────────────
  // v61: this used to set its own label and fire a dataLayer event, then stop.
  // It never navigated, so all 17 localised Home pages were reachable only by
  // direct URL. It also listed 14 languages; id, pl and th were missing.
  // Now it matches the Sales page: 17 languages, one real anchor each so a
  // crawler can follow them and a visitor can open one in a new tab, and the
  // choice is remembered before the navigation happens.
  var LANGS_TOP = [
    {code:'en',      name:'English | English',                   code2:'EN'},
    {code:'ar',      name:'Arabic | العربية',                    code2:'AR'},
    {code:'zh-Hans', name:'Chinese (Simplified) | 中文（简体）',  code2:'ZH'},
    {code:'zh-Hant', name:'Chinese (Traditional) | 中文（繁體）', code2:'ZH-T'},
    {code:'fr',      name:'French | Français',                   code2:'FR'},
    {code:'de',      name:'German | Deutsch',                    code2:'DE'},
    {code:'id',      name:'Indonesian | Bahasa Indonesia',       code2:'ID'},
    {code:'it',      name:'Italian | Italiano',                  code2:'IT'},
    {code:'ja',      name:'Japanese | 日本語',                    code2:'JA'},
    {code:'ko',      name:'Korean | 한국어',                      code2:'KO'},
    {code:'pl',      name:'Polish | Polski',                     code2:'PL'},
    {code:'pt',      name:'Portuguese | Português',              code2:'PT'},
    {code:'ru',      name:'Russian | Русский',                   code2:'RU'},
    {code:'es',      name:'Spanish | Español',                   code2:'ES'},
    {code:'th',      name:'Thai | ไทย',                          code2:'TH'},
    {code:'tr',      name:'Turkish | Türkçe',                    code2:'TR'},
    {code:'vi',      name:'Vietnamese | Tiếng Việt',             code2:'VI'}
  ];

  // One page per language, so switching language is a navigation, not a swap.
  // Case is PRESERVED: Chinese is /zh-Hans/ and /zh-Hant/, matching FAQ,
  // Method and the Sales page.
  var PAGE_LANG   = 'en';
  var LANG_ORIGIN = 'https://www.eslpod.com';

  function urlForLang(code){
    return code === 'en' ? LANG_ORIGIN + '/' : LANG_ORIGIN + '/' + code + '/';
  }

  var langPill = $('#ep-lang-pill');
  var langMenu = $('#ep-lang-menu');
  var langLabel = $('#ep-lang-pill-label');
  if (langPill && langMenu){
    var selfEntry = LANGS_TOP.filter(function(L){ return L.code === PAGE_LANG; })[0];
    if (langLabel && selfEntry) langLabel.textContent = selfEntry.code2;
    LANGS_TOP.forEach(function(L){
      var b = document.createElement('a');
      b.setAttribute('role','option');
      b.setAttribute('lang', L.code);
      b.setAttribute('href', urlForLang(L.code));
      b.innerHTML = '<span>'+L.name+'</span><span class="ep-lang-opt-code">'+L.code2+'</span>';
      b.dataset.lang = L.code;
      b.dataset.code = L.code2;
      if (L.code===PAGE_LANG) { b.classList.add('active'); b.setAttribute('aria-selected','true'); }
      b.addEventListener('click', function(e){
        // v62: remember the choice either way, then let a modifier click fall
        // through to the browser so it opens a new tab or window. v61 called
        // preventDefault() first, which suppressed exactly the behaviour the
        // comment above promises.
        try { (window.dataLayer = window.dataLayer || []).push({event:'lang_change', lang:L.code}); } catch(err){}
        // Remember the choice so the other pages can honour it later.
        try { localStorage.setItem('eslpod_ue_lang', L.code); } catch(err){}
        try { document.cookie = 'eslpod_ue_lang=' + L.code + ';path=/;max-age=31536000;SameSite=Lax'; } catch(err){}
        if (e.metaKey || e.ctrlKey || e.shiftKey || e.altKey || e.button === 1) return;
        e.preventDefault();
        langMenu.classList.remove('open');
        langPill.setAttribute('aria-expanded','false');
        if (L.code === PAGE_LANG) return;
        window.location.href = urlForLang(L.code);
      });
      // v62: the options were <button> before v61 and answered Enter AND Space.
      // They are anchors now, which answer Enter only, so Space is restored.
      b.addEventListener('keydown', function(e){
        if (e.key === ' ' || e.key === 'Spacebar'){ e.preventDefault(); b.click(); }
      });
      langMenu.appendChild(b);
    });
    langPill.addEventListener('click', function(e){
      e.stopPropagation();
      var open = langMenu.classList.toggle('open');
      langPill.setAttribute('aria-expanded', open?'true':'false');
    });
    document.addEventListener('click', function(e){
      if (!e.target.closest('.ep-lang-wrap')) {
        langMenu.classList.remove('open');
        langPill.setAttribute('aria-expanded','false');
      }
    });
  }

  // ─── AudioSample (v14: real duration + click/drag seek + play/pause icon sync) ─
  var audioPlay = $('#tl-audio-play');
  var audioEl = $('#tl-audio-el');
  var audioBar = $('#tl-audio-bar-fill');
  var audioCur = $('#tl-audio-cur');
  var audioTotal = document.querySelector('#tl-audio-total');
  var audioBarTrack = document.querySelector('.tl-audio-bar');
  if (audioPlay && audioEl) {
    var DURATION = 120; // fallback until metadata loads
    function fmt(t){ if (!isFinite(t) || t < 0) t = 0; var m = Math.floor(t/60); var s = Math.floor(t%60); return m+':'+(s<10?'0':'')+s; }
    function setBar(t){
      audioBar.style.width = (DURATION ? (t/DURATION*100) : 0) + '%';
      audioCur.textContent = fmt(t);
    }
    audioEl.addEventListener('loadedmetadata', function(){
      if (isFinite(audioEl.duration) && audioEl.duration > 0) {
        DURATION = audioEl.duration;
        if (audioTotal) audioTotal.textContent = fmt(DURATION);
        setBar(audioEl.currentTime || 0);
      }
    });
    audioEl.addEventListener('durationchange', function(){
      if (isFinite(audioEl.duration) && audioEl.duration > 0) {
        DURATION = audioEl.duration;
        if (audioTotal) audioTotal.textContent = fmt(DURATION);
      }
    });
    audioEl.addEventListener('timeupdate', function(){ setBar(audioEl.currentTime); });
    // v14: keep play-button icon in sync with the actual audio state, no matter the source
    audioEl.addEventListener('play', function(){
      audioPlay.classList.add('playing');
      audioPlay.setAttribute('aria-label','Pause');
      try { (window.dataLayer = window.dataLayer || []).push({event:'audio_play'}); } catch(e){}
    });
    audioEl.addEventListener('pause', function(){
      audioPlay.classList.remove('playing');
      audioPlay.setAttribute('aria-label','Play sample');
    });
    audioEl.addEventListener('ended', function(){
      audioPlay.classList.remove('playing');
      audioPlay.setAttribute('aria-label','Play sample');
      try { (window.dataLayer = window.dataLayer || []).push({event:'audio_complete'}); } catch(e){}
    });
    audioPlay.addEventListener('click', function(){
      if (audioEl.paused) {
        var p = audioEl.play();
        if (p && p.catch) p.catch(function(){});
      } else {
        audioEl.pause();
      }
    });

    // v14: click + drag to seek (mouse and touch). Resumes playback if it was paused mid-seek.
    if (audioBarTrack) {
      audioBarTrack.style.cursor = 'pointer';
      audioBarTrack.style.touchAction = 'none';
      var isDragging = false;
      var wasPlayingBeforeDrag = false;
      function getClientX(ev){
        if (typeof ev.clientX === 'number') return ev.clientX;
        if (ev.touches && ev.touches[0]) return ev.touches[0].clientX;
        if (ev.changedTouches && ev.changedTouches[0]) return ev.changedTouches[0].clientX;
        return null;
      }
      function seekFromEvent(ev){
        if (!DURATION) return;
        var rect = audioBarTrack.getBoundingClientRect();
        var x = getClientX(ev);
        if (x === null) return;
        var pct = Math.max(0, Math.min(1, (x - rect.left) / rect.width));
        var t = pct * DURATION;
        audioEl.currentTime = t;
        setBar(t);
      }
      function startDrag(ev){
        isDragging = true;
        wasPlayingBeforeDrag = !audioEl.paused;
        // While dragging, pause so the scrubber doesn't fight live playback
        if (wasPlayingBeforeDrag) audioEl.pause();
        seekFromEvent(ev);
      }
      function endDrag(){
        if (!isDragging) return;
        isDragging = false;
        // Resume playback after a seek — auto-restart whether or not it was playing before
        var p = audioEl.play();
        if (p && p.catch) p.catch(function(){});
      }
      // Mouse
      audioBarTrack.addEventListener('mousedown', function(ev){ ev.preventDefault(); startDrag(ev); });
      document.addEventListener('mousemove', function(ev){ if (isDragging) seekFromEvent(ev); });
      document.addEventListener('mouseup', endDrag);
      // Touch
      audioBarTrack.addEventListener('touchstart', function(ev){ startDrag(ev); }, {passive: true});
      document.addEventListener('touchmove', function(ev){ if (isDragging) seekFromEvent(ev); }, {passive: true});
      document.addEventListener('touchend', endDrag);
      document.addEventListener('touchcancel', endDrag);
    }
  }

  // ─── Testimonial carousel (manual scroll-snap + dots) ────────
  // v13: default is grid (testStyle="grid"); carousel logic only runs if user flips style back to carousel
  var track = $('#ep-carousel-track');
  var prevBtn = $('#ep-carousel-prev');
  var nextBtn = $('#ep-carousel-next');
  var dotsHost = $('#ep-carousel-dots');
  var _page = document.getElementById('ep-page');
  var _isGrid = _page && _page.getAttribute('data-test-style') === 'grid';
  if (track && dotsHost && !_isGrid) {
    var cards = $$('.ep-tcard', track);
    function dotCount(){
      var w = track.clientWidth;
      var per = Math.max(1, Math.floor(w / 296));
      return Math.max(1, Math.ceil(cards.length / per));
    }
    function renderDots(){
      dotsHost.innerHTML = '';
      var n = dotCount();
      for (var i=0; i<n; i++){
        var s = document.createElement('span');
        if (i===0) s.classList.add('active');
        dotsHost.appendChild(s);
      }
    }
    function updateActiveDot(){
      var dots = $$('span', dotsHost);
      if (!dots.length) return;
      var pct = track.scrollLeft / Math.max(1, track.scrollWidth - track.clientWidth);
      var idx = Math.min(dots.length-1, Math.round(pct * (dots.length-1)));
      dots.forEach(function(d,i){ d.classList.toggle('active', i===idx); });
    }
    renderDots();
    track.addEventListener('scroll', updateActiveDot, {passive:true});
    window.addEventListener('resize', function(){ renderDots(); updateActiveDot(); });
    if (prevBtn) prevBtn.addEventListener('click', function(){ track.scrollBy({left: -track.clientWidth*0.85, behavior:'smooth'}); });
    if (nextBtn) nextBtn.addEventListener('click', function(){ track.scrollBy({left:  track.clientWidth*0.85, behavior:'smooth'}); });
    // v22 item 10: dead carousel auto-rotate JS removed (testimonials render as
    // a static grid via data-test-style="grid"; the timer never engaged and
    // should not ship). Manual prev/next + dots retained for the carousel
    // fallback if the grid style is ever turned off.
  }

  // ─── Tracking: cta_click_trial, scroll_75 ────────────────────
  // v32: was listening for cta_click_trial only, so the two other data-track
  // attributes on the page (cta_click_freecourse_nav, cta_click_pricing) were
  // decorative and fired nothing. Now any data-track value pushes its own event.
  document.addEventListener('click', function(e){
    var el = e.target.closest('[data-track]');
    if (!el) return;
    var name = el.getAttribute('data-track');
    if (!name) return;
    try { (window.dataLayer = window.dataLayer || []).push({event:name, cta_label: el.textContent.trim()}); } catch(err){}
  });
  var scrollFired = false;
  window.addEventListener('scroll', function(){
    if (scrollFired) return;
    var h = document.documentElement;
    var pct = (h.scrollTop + window.innerHeight) / h.scrollHeight;
    if (pct >= 0.75) {
      scrollFired = true;
      try { (window.dataLayer = window.dataLayer || []).push({event:'scroll_75'}); } catch(e){}
    }
  }, {passive:true});

  // v22 item 10: dev tweaks panel removed before production. The seven layout
  // toggles it used to apply at runtime (heroBg, heroLayout, heroCta,
  // bigideaStyle, ueLayout, testStyle, mobileCta) are now hardcoded as static
  // data-attributes on #ep-page, so the rendered page is byte-for-byte the
  // same without the panel, the postMessage host integration, or the editmode
  // hooks shipping to production.

})();
</script>
<!-- ===== v10: popup-blocker fallback for popup-trigger CTAs ===== -->
<script>
(function(){
  'use strict';
  // After any popup-trigger click, check 500ms later whether the modal opened.
  // Modal open-state signal in this page: #leadmagnet-modal has aria-hidden="false"
  // (set in openModal() in the popup runtime above) AND its inline display is 'block'.
  // If neither is true, the browser/blocker likely suppressed the modal -> redirect.
  document.addEventListener('click', function(e){
    var trigger = e.target.closest('.popup-trigger');
    if (!trigger) return;
    setTimeout(function(){
      // v40: reads the single hoisted constant in the popup runtime above.
      // The literal that used to sit here was a second copy, so the VERIFY
      // destination had to be changed in two places or it silently disagreed.
      var FALLBACK = window.ESLPOD_FALLBACK_URL || 'https://www.eslpod.com/signup.html';
      var modal = document.querySelector('#leadmagnet-modal');
      if (!modal) { window.location.href = FALLBACK; return; }
      var ariaOpen = modal.getAttribute('aria-hidden') === 'false';
      var displayOpen = window.getComputedStyle(modal).display !== 'none';
      var isOpen = ariaOpen || displayOpen;
      if (!isOpen) {
        window.location.href = FALLBACK;
      }
    }, 500);
  }, true);
})();
</script>
<!-- ===== v18: mobile sticky CTA reveal via IntersectionObserver ===== -->
<script>
(function(){
  if (!('IntersectionObserver' in window)) return;
  var sticky = document.querySelector('#ep-mobile-stickycta');
  var heroCta = document.querySelector('#ep-hero .ep-btn-large.popup-trigger');
  if (!sticky || !heroCta) return;
  var obs = new IntersectionObserver(function(entries){
    entries.forEach(function(e){
      if (e.isIntersecting) {
        sticky.classList.remove('ep-sticky-revealed');
      } else {
        sticky.classList.add('ep-sticky-revealed');
      }
    });
  }, { rootMargin: '-20% 0px 0px 0px' });
  obs.observe(heroCta);
})();
</script>
<script>
/* v33: trial launch guard. Warns in the console when the trial is switched on
   but a companion launch edit was missed. v35 adds the third check, the popup
   redirect base. Development aid only, it renders nothing and sends nothing. */
(function(){
  try {
    var page = document.getElementById('ep-page');
    if (!page || page.getAttribute('data-trial') !== 'on') return;
    var ld = document.querySelector('script[type="application/ld+json"]');
    if (!ld || ld.textContent.indexOf('How does the 7-day free trial work?') === -1) {
      console.warn('[ESLPod] data-trial is "on" but the trial question is missing from the FAQPage in the @graph, so the visible FAQ and the structured data disagree. See the TRIAL FAQ SCHEMA note in <head>.');
    }
  } catch (e) {}
})();
</script>
</body></html>