Last time, we sorted out what those mysterious little markdown files are (you can catch up here: What Are Markdown Files?). Right at the end, I mentioned something called Skills and promised they deserved a post all of their own. Well, here we are.

If you’ve been hearing the word “Skills” floating around your AI tools lately and quietly nodding along while having no real idea what people mean, you’re in exactly the right place. I get this question a lot from clever business owners who are using AI every day but keep bumping into new bits of jargon. So let’s clear this one up properly. Plain English, no tech headache, and a genuinely good reason you’ll want to start paying attention to Skills.

So, what is a skill, really?

A Skill is a saved set of instructions that teaches your AI how to do a specific task properly, the same way, every single time.

That’s the heart of it. Instead of explaining the same task to your AI over and over (and getting a slightly different result each time because you worded it slightly differently), you write the instructions down once, save them as a Skill, and your AI follows them whenever that task comes up.

Think of it like a really good recipe card you hand to a brilliant but literal assistant. The recipe spells out the steps, the ingredients, the order, and the little details that make it turn out right. Give that same card to your assistant on a Monday or a Friday, when you’re fresh or frazzled, and you get the same lovely result either way. That’s a Skill.

Most Skills are written in Markdown (yep, the plain little file type from last time), saved in a file called SKILL.md. It holds a name, a short description of when to use it, and the step-by-step instructions for getting the job done.

A prompt, a template and a Skill walk into a bar

These three get muddled all the time, so let’s draw a clear line between them. Once you see the difference, the whole thing clicks.

A prompt is a one-off instruction. You type it, you get a result, and then it’s gone. Next time you want the same thing, you’re typing it all again from scratch (or digging through your chat history trying to find it). If you’re organised, you may save it somewhere for easy access and create a prompt template that you can copy and paste, maybe replacing [X] with the appropriate variable. But from my experience, most people don’t.

A Custom GPT or Gem is a whole assistant you set up once for a particular job, then go back to whenever you need it. (Custom GPTs live in ChatGPT, Gems live in Gemini, and they’re the same idea.) This is a big step-up because the instructions and knowledge are baked in, so you’re not starting from scratch. The catch is that it’s a separate destination. You have to stop what you’re doing, switch into that specific assistant, and work inside it, and it doesn’t retain any knowledge from outside that individual chat. Annoying!

A Skill flips that around. Instead of you going off to a separate assistant, the Skill comes to you. It lives inside the AI tool you’re already using, and it switches itself on automatically the moment the task it’s built for comes up. You don’t change tools, you don’t pick it from a menu, you just ask for the thing and the right Skill quietly does the work in the background.

Here’s the way I think about it. A prompt is the start of a conversation. A Custom GPT or Gem is a specialist you go and visit. A Skill is a specialist who already knows your business and shows up exactly when you need them, right where you’re already working.

Why Skills matter more than you’d think

This is where it gets genuinely useful for the way you run your business.

The real magic of a Skill is consistency. Most of us don’t have a marketing problem so much as a repeatability problem. You can write a cracking social post when you’re in the zone. The trouble is doing it again next week, and the week after, to the same standard, when you’re busy and tired and just want it off your plate. A Skill holds that standard for you, so the quality doesn’t wobble depending on your mood or your morning or who is using the skill (yes, you can share skills with your team members and take those tasks off your plate #winning).

It also saves you a heap of time. The instructions are already written, so you’re not re-explaining the task or fixing a half-right result. You ask, it delivers, you tweak. Done.
And here’s the part I care about most because it’s central to how I teach AI. A Skill is something you build and own. When you write your own Skill, you understand exactly what it’s doing and why, which means you can direct it, adjust it, and trust it. That’s a world away from using a Custom GPT you stumbled across on the internet and hoping for the best. Skills create freedom. Borrowed shortcuts create dependency.

That fits squarely into what I bang on about constantly: the goal isn’t collecting more AI tools, it’s getting genuinely good at directing the ones you’ve already got. A Skill is one of the clearest ways to turn “I figured out how to do this once” into “my AI does this brilliantly for me, on demand, forever.”

Where you’ll actually bump into Skills

Skills started life in Claude, which is my main tool, so that’s where you’ll meet them most directly. But here’s the exciting bit, and the reason Skills are worth your attention now rather than later, is because they’re quickly becoming a shared standard right across the AI world. That makes them genuinely portable, which is rare and rather wonderful.

Claude and Claude Cowork

This is Skills in their truest form, and where they began. In Claude, you can build a Skill as a SKILL.md file and Claude will use it automatically whenever the task it’s designed for comes up. In Claude Cowork (the version that can work with the files and folders on your computer), Skills become really powerful because they can carry out proper multistep jobs for you, not just write a paragraph. This is exactly the kind of thing I’ll be teaching in my next AI MegaClass.

And now, well beyond Claude

Anthropic (the company behind Claude) didn’t keep Skills to themselves. In late 2025 they published the Skill format as an open standard, meaning any AI platform can adopt it. And adopt it they have, fast. Skills now work in ChatGPT (only for business teams at the time of writing), in Google’s tools, in agent platforms like Manus, and across a growing list of others. Industry-watchers are already calling it the new best-practice standard for this kind of thing, much like what happened a year earlier when Anthropic opened up the technology that lets AI tools connect to your other apps.

What does that actually mean for you? Portability. Build a Skill once, and you’re increasingly able to use that same Skill across different tools, rather than being locked into one. The work you put into capturing how a task should be done isn’t trapped inside a single platform. As more tools adopt the standard (and it’s looking like most of them will), your Skills travel with you. That’s a smart place to be putting your effort.

How this sits alongside Custom GPTs and Gems

You might already have a Custom GPT in ChatGPT or a Gem in Gemini. These are close relatives. Saved, purpose-built assistants you set up once and return to. The key difference is that they tend to live inside one platform, whereas a Skill (built on the open standard) is designed to be far more portable. Tellingly, the platforms are converging on the Skill approach.

ChatGPT has rolled out its own Skills feature. The direction of travel is clear. And if you’ve created a Custom GPT of Gem before, it’s super easy to convert it into a skill. Just take your knowledge and instructions from the GPT or Gem over to Claude and ask Claude to convert it into a skill for you. Magic!

Your wider AI Dream Team

This is really what Skills, Custom GPTs and Gems all add up to. A personalised team of AI helpers, each one trained for a specific job in your business. One that writes your emails in your voice. One that turns a blog into social posts. One that drafts your ad copy. You build them once, and they show up ready to work every time. Build them as Skills, and that team can increasingly come with you wherever you go.

The thread running through all of these is the same. The more clearly you capture how a task should be done, the more reliably your AI does it. A Skill is simply the cleanest, most future-proof way to lock that in.

What sorts of tasks make a great Skill?

Not everything needs to be a Skill. The sweet spot is any task you do regularly, that follows a repeatable pattern, and that you want done to a consistent standard. A few that work beautifully for service-based business owners:

• Writing social posts in your brand voice using your preferred structure
• Turning one blog post into a week’s worth of social content
• Drafting email newsletters that actually sound like you
• Repurposing a podcast or video into written content
• Creating ad copy from a simple brief
• Converting meeting transcripts into accurate minutes with action items

The pattern to look for is simple. If you’ve ever thought “I wish I could just clone the version of me that does this really well,” that task is a strong candidate for a Skill.

What doesn’t make a good Skill? The one-off, the highly creative, or the genuinely strategic. Use Skills for the mundane, repeatable jobs. Keep the masterful, judgement-heavy work in your own hands. That’s the whole point: free yourself from the repetitive stuff so you’ve got more headspace for the work only you can do.

The honest drawbacks of Skills

I’m not going to pretend Skills are a magic wand because they aren’t, and you deserve the full picture.

• They take a bit of effort to set up. You’re investing time upfront to write good instructions. The payoff is real, but it isn’t instant. You’re building an asset, not flicking a switch.
• A Skill is only as good as the thinking behind it. Vague, woolly instructions give you vague, woolly results. The quality of what comes out depends entirely on the clarity of what you put in.
• They still need a human in the loop. A Skill gets you a strong, consistent draft. It doesn’t get you a finished product you should publish unread. You’re still the editor, the brand guardian, and the final say.
• They’re not “set and forget” forever. As your business shifts, or the tools update, your Skills will need the occasional tidy-up to stay sharp. But the good news here is that it’s not that hard to update. Just tell Claude where the Skill needs improvement, and it can make the adjustments for you. 

The simplest way to hold all this in your head is to remember that a Skill is a brilliant way to bottle your best work and repeat it. It isn’t a way to remove yourself from the work entirely. You stay in charge. The Skill just does the majority of the legwork, consistently and to your standards.

How to get started with Skills (without the tech headache)

Good news. You don’t need to code, and you can start far more simply than you might think.

• Begin with one task you do all the time. Don’t try to build ten Skills at once. Pick the single repetitive job that drains you most and start there.
• Write down how you do it well. Open a plain document and describe the steps, the rules, the tone, the format. If you can explain it to a new team member, you can write a Skill. That description is basically your first draft. And don’t forget to feed the skill with whatever core knowledge it requires, either by embedding that content as a Markdown file or by linking to the required knowledge in a shared folder in Claude Cowork (if that’s where you intend to use the Skill).
• Let your AI help you build it. This is the lovely bit. You can ask Claude to help you turn your thoughts or notes into a proper Skill. It speaks the format fluently, so you can stay hands-off on the technical side and focus on the thinking.
• Test it, then tweak it. Run the Skill on a real task, see where the output drifts from what you wanted, and sharpen the instructions. A couple of rounds of this, and you’ll have something genuinely reliable.

Start small, start with something annoying and repetitive, and let the tool do the fiddly part. But I will warn you, once you get started making Skills, it is highly addictive. You’ll have an AI Dream Team created in no time at all.

Frequently asked questions about Skills

What is an AI Skill in simple terms?

It’s a saved set of instructions that teaches your AI to do a specific task the same way every time. Rather than re-explaining a job whenever it comes up, you write the instructions once and your AI follows them on demand, giving you consistent results without the repetition.

Are Skills the same as prompts?

Not quite. A prompt is a one-off instruction you type in the moment. A Skill is the full, saved set of instructions for a task that your AI can trigger and follow whenever that task comes up. Think of a prompt as a single conversation and a Skill as a trained team member who already knows the job.

Do I need to know how to code to create a Skill?

Not at all. Skills are usually written in Markdown, which is plain text with a few simple symbols, and you can ask your AI to do the technical formatting for you. Your real job is the thinking: describing clearly how the task should be done. The tool handles the rest.

Which AI tools use Skills?

Skills started in Claude, including Claude Cowork, where they’re at their most powerful. In late 2025 the Skill format was made an open standard, and it’s been adopted quickly, so Skills now work in ChatGPT (including for business teams), Google’s tools, agent platforms like Manus, and a growing list of others. That’s the real beauty of them… Build a Skill once, and it’s increasingly portable across the tools you use, rather than locked into one.

What’s the difference between a Custom GPT or Gem?

They’re close cousins, and the main difference is how you reach them. A Custom GPT or Gem is a separate assistant you switch into and work inside whenever you need that job done. A Skill lives inside the AI tool you’re already using and triggers itself automatically when the relevant task comes up, so you don’t have to leave what you’re doing or pick it from a list. In short, you go to a Custom GPT or Gem, whereas a Skill comes to you.

Skills are how you bottle your best work

Here’s what I really want you to take away. Skills aren’t a developer’s plaything you need to feel intimidated by. They’re one of the most practical ways to get AI doing real work for your business, to your standard, without you having to be there every time.

The shift is subtle but powerful. You stop being the person who does the repetitive task, and you become the person who trains the thing that does it. You stay firmly in charge. AI just becomes a far more capable partner. Human-led, AI-powered, exactly as it should be.

If you want to see this in action, this is precisely what I’m teaching in my next AI Marketing MegaClass on Claude Cowork and Skills, happening on 19 June 2026. It’s a hands-on session where I’ll show you how to build Skills that actually work for your business, so your AI stops guessing and starts delivering.

Grab your spot here: impactiv8.com.au/ai

You’ve got this!

You asked Claude (or ChatGPT) for something useful. A summary, a brand brief, a set of instructions. It did the job beautifully, then handed you a file ending in .md and you thought, “Lovely. But what actually is this thing, and why isn’t it a normal Word doc?”

You’re not alone. I get this question constantly from clever, capable business owners who are doing brilliant things with AI but keep tripping over this odd little file type. So let’s clear it up properly. Plain English, zero tech jargon, and a few good reasons you’ll want to start using markdown on purpose.

So, what is a markdown file, really?

A markdown file is just a plain text file with a few simple symbols added in to handle the formatting. That’s genuinely it.

The file ends in .md (short for markdown), and instead of clicking a “bold” button like you would in Word, you pop a couple of symbols around your text and the formatting sorts itself out.

A quick taste of the symbols:

• A # at the start of a line makes a heading
• ## makes a slightly smaller heading
• **two asterisks** around a word makes it bold
• A dash at the start of a line makes a bullet point

That’s most of what you’ll ever need. No special software, no fancy editor, no code. Markdown was invented back in 2004 to let people write formatted text without wrestling with messy code, and two decades later it’s having a real moment. It turns out to be the format AI tools understand best.

Here’s a comforting thought. If you’ve ever jotted a quick list in your notes app and popped a dash in front of each item, you’ve already written markdown without knowing it. That dash is exactly how markdown makes a bullet point. Markdown just gives you a few more of these simple shortcuts, so both humans and machines can follow your structure.

Why a plain little text file matters more than you’d think

This is where it gets genuinely interesting for you, now that AI is part of how you run your business.

When you hand an AI tool a document to work from, it has to read and interpret that document before it can do anything useful with it. And the format you give it makes a real difference to how well it understands you.

AI tools read markdown more easily than almost any other format. The headings, bullets and bold text act like signposts, telling the AI exactly how your information is organised. It doesn’t have to guess where one idea ends and the next begins. It can simply see it.

But the even better bit, Markdown uses fewer tokens.

Tokens are the small chunks of text that AI tools process. They’re effectively the fuel AI runs on, and most tools charge or limit you based on how many you use. A Word doc or a PDF carries a lot of invisible extra baggage in the form of formatting code, layout instructions and hidden characters that make it look pretty on your screen. The AI has to wade through all of that before it even gets to your actual words. A markdown file strips it right back to the words and a handful of simple symbols.

That matters for two very practical reasons:

1. Cost. Fewer tokens means the AI does less work to read your file, which keeps your usage and your costs down. That adds up fast once you’re running things regularly or building tools you’ll use every day.
2. Quality. When the AI isn’t spending its energy untangling messy formatting, it has more room to focus on the actual task you’ve asked it to do. Cleaner input, sharper output. It really is that simple.

So a humble .md file isn’t just tidy. It’s cheaper to run, and it gets you better results. Not bad for a plain text file.

Where you’ll actually bump into Markdown in your AI tools

Once you know what to look for, you’ll start spotting markdown everywhere in your AI work. Here’s where it shows up most.

ChatGPT chats, Custom GPTs and Projects

When you build a Custom GPT or Project, the instructions and knowledge files you feed it work beautifully in Markdown. Clear headings and bullets help your GPT follow your instructions properly, rather than loosely interpreting a wall of text.

Gemini Gems

Same story over in Google’s world. If you’re building a Gem, structuring its instructions in Markdown helps it stay on track and behave the way you actually intended.

Claude and Projects

Claude is my main tool, and markdown is its native language. When you set up a Project and load it with knowledge files (your brand guide, your ideal customer profile, your offers), those files are at their most useful in Markdown. Claude reads them faster, understands them better, and produces work that actually sounds like you.

Files in your Claude Cowork folders

If you’ve started dropping files into folders that Claude Cowork can see, markdown is the format you want sitting in there. It’s the difference between Claude skimming a clean, well-signposted document and Claude squinting at a cluttered one.

Skills

Markdown is also the format behind Skills, the reusable instruction files you can build to get AI doing repeatable tasks for you the same way every single time. I won’t go down that rabbit hole here because Skills deserve a post all of their own (which you can read here).

The thread running through all of these? The cleaner and better structured your input, the better your AI behaves. Markdown is simply the easiest way to give it that.

Why .md beats Word, PDF and Google Docs for AI work

You might be wondering why you can’t just keep using the formats you already know and love. You can. AI can read a Word doc, a PDF or a Google Doc. But “can read” and “reads well” are two different things. Here’s why markdown wins for AI work specifically:

• It’s lighter and cheaper to process (those tokens again)
• It’s clearer for the AI to interpret, which lifts the quality of what comes back to you
• It works on any device and in any text editor, so it’s never locked inside one piece of software
• It’s future-proof. Plain text has been readable for decades, and it’ll still be readable decades from now
• It’s portable. You can move the same .md file between ChatGPT, Claude, Gemini and your own folders without anything breaking along the way

When the job is feeding AI good information, markdown is the right tool. (For a polished proposal you’re emailing a client, Word or a PDF still wins. More on that in a second.)

The honest drawbacks of markdown

I’m not going to pretend markdown is perfect, because it isn’t, and you deserve to know where it falls short.

• It looks plain. No fonts, no colours, no logos. If you need a beautifully branded document, this is not it.
• It can’t hold images inside the file the way a Word doc or PDF can. It’s words and structure, not visuals.
• The symbols feel strange at first. The first time you see ## and ** scattered through your text, it looks like a mistake. It isn’t, and it clicks quickly, but there’s a small learning curve.
• It’s not a client-facing format. You wouldn’t send a raw markdown file to a customer as your final deliverable.

The simplest way to hold all this in your head… Markdown is a brilliant working format for feeding and instructing AI. It is not a presentation format. Use it for the engine room, not the shop window.

How to read and edit a markdown file without the tech headache

Good news. You don’t need anything fancy, and you almost certainly already have what you need.

• Notion. If you use Notion, you can paste markdown straight in and it renders properly. Headings, bullets and all. It’s one of the friendliest ways to read and tidy up a .md file.
• The free text editor on your computer. On a Mac, TextEdit works (set it to plain text). On Windows, Notepad does the job. They’ll show you the raw markdown, symbols and all, which is perfectly fine for a quick read or edit.
• VS Code. Free, and it gives you a live preview of how your markdown will look once it’s rendered. Handy while you’re still getting comfortable with the symbols.
• Obsidian. If you fall for Markdown and want to build a proper knowledge library over time, this is a great free option to grow into. Not essential, but a nice one to have up your sleeve.
• Just ask your AI. Honestly, the simplest option of the lot. Ask Claude or ChatGPT to read it, edit it, or rewrite it for you. They speak Markdown fluently, so you can stay completely hands-off if you’d rather.

Start with whatever you already use. For most of my clients, that’s Notion or simply asking Claude to handle it.

Frequently asked questions about markdown files

What does .md stand for?

It’s short for Markdown, the name of the simple formatting style the file uses. You’ll occasionally see .markdown too, but .md is the common one.

How do I open a markdown file?

You can open it in any text editor, including the free ones already on your computer (TextEdit on a Mac, Notepad on Windows). Notion and VS Code will display it nicely formatted, and you can always ask your AI tool to read it for you.

Do I need to learn to code to use Markdown?

Not at all. Markdown was designed specifically for people who don’t code. You only need a handful of simple symbols, and even those are optional. Plain paragraphs work perfectly well on their own.

Is markdown better than a Word document for AI?

For feeding information to AI, yes, in most cases. It’s lighter, cheaper to process and easier for the AI to understand, which tends to produce better results. For client-facing documents you want to look polished, Word or PDF is still the better pick.

Why does AI keep giving me markdown files?

Because it’s the format AI works with most naturally. When a tool like Claude or ChatGPT hands you a .md file, it’s giving you something clean, portable and easy to reuse across your other AI tools.

Your knowledge core starts with a few simple .md files

Here’s the thing I really want you to take away. Markdown isn’t a developer’s tool you need to feel intimidated by. It’s the quiet workhorse behind the smartest, most consistent AI setups, and it’s well within your reach.

The single best way to get AI producing work that actually sounds like you and your business is to give it a solid set of knowledge files to work from, written in clean, simple markdown. Your brand voice, your ideal customer, your offers, all of it.

That’s exactly what my free AI Knowledge Cloning Kit walks you through. It steps you through the most important markdown files to create for your knowledge core, so your AI stops guessing and starts working like it genuinely knows you and your business.

Grab it here: impactiv8.com.au/AIKnowledge

You’ve got this!

There’s a quiet panic I see in business owners right now, and it usually shows up as a confession.

“I feel like everyone else gets AI and I just don’t.”

They think there’s a technical knack they’re missing. Some prompt-engineering wizardry, a hidden menu of clever commands, a tool they haven’t found yet. They’ve convinced themselves that the people getting brilliant results from ChatGPT or Claude are simply more tech-savvy than they are.

I want to let you off the hook, because that’s not it. Not even close.

After years of teaching small business owners how to actually use this stuff, here’s what I’ve come to believe…

The people who get great results from AI aren’t the most technical. They’re the ones who bring a particular set of human skills to the table. Judgement. Discernment. The willingness to look at something and say “no, that’s not quite right, do it again.” And the good news, the genuinely brilliant news for service-based business owners, is that you’ve been building these skills your whole career.

The myth of the magic prompt

Let’s kill the biggest misconception first, because it’s holding so many people back.

There’s this idea that AI is a vending machine. You put in the right combination of words, the perfect prompt, and out pops the perfect result. So people hunt for the magic words. They collect prompt templates like Pokémon cards. And when the output is mediocre, they assume they got the magic words wrong.

But AI doesn’t really work like a vending machine. It works more like a very fast, very eager assistant who knows a little about everything and will confidently have a go at anything you ask, including the things it gets completely wrong.

The research backs this up in a way that should change how you think about the whole thing. Studies have found that people who use AI to get direct, finished answers actually get worse over time. The people who improve are the ones who use it as a thinking partner, who ask for options, push back, question the output and refine it. Same tool. Wildly different results. And the difference has nothing to do with technical skill and everything to do with how you engage with it.

So if it’s not about the magic prompt, what is it about?

The five human skills that actually matter

1. Effective briefing, or training your AI properly

You’ve heard the phrase “rubbish in, rubbish out.” Nowhere is it more true than with AI.

When you open a fresh AI chat and fire off a one-line request, you’re asking a brilliant assistant who knows nothing about your business to guess. Who’s your customer? What do you sound like? What’s worked for you before? It has no idea, so it fills the gaps with bland, could-be-anyone answers. Then people wonder why the output feels generic.

The fix isn’t a cleverer prompt. It’s better briefing. The most effective AI users invest time upfront teaching their AI about their business, their audience, their voice, their offers and their goals. They feed it context, examples and background, so it’s working from real knowledge instead of guesswork. Do that once, properly, and you dramatically cut the amount of correcting you have to do afterwards.

Think of it exactly like onboarding a new team member. You wouldn’t hand a new hire one vague instruction and expect brilliance on day one. You’d train them, show them examples of good work, explain your customers and your standards. AI deserves, and rewards, the same investment. (That’s the whole idea behind the AI Knowledge Cloning Kit, getting everything that’s in your head into a form your AI can actually use.)

The better you train it, the less you have to fix. That’s the skill.

2. Critical thinking, or knowing when something’s off

This one’s becoming genuinely rare, and it might be the most valuable of the lot.

AI is designed to sound confident. It will give you a wrong answer with exactly the same conviction as a right one. It’ll invent a statistic, misremember a fact, or write something that’s grammatically perfect and strategically useless, all while sounding like it knows precisely what it’s doing.

Your job is to be the one in the room who isn’t fooled.

When AI hands you a piece of copy, a strategy, a list of ideas, the most valuable thing you can do is read it with a slightly raised eyebrow. Does this actually make sense? Would my customer really respond to this? Is that claim even true? Is this the obvious answer anyone could have come up with, or is it genuinely good?

You already do this. Every time you’ve reviewed a quote from a supplier, vetted a contractor, reviewed a team member’s work, or read a contract and thought “hang on, that clause doesn’t sit right with me,” you’ve used exactly this skill. AI just gives you more to apply it to.

3. Taste, or knowing what good actually looks like

Here’s a word you don’t hear much in marketing circles, but it might be the most important one in the AI era… Taste.

Taste is the ability to look at three options and know which one is best. It’s knowing what to leave out, what to emphasise, and when something is technically fine but emotionally flat. AI can produce ten captions in four seconds. It may even suggest which one it thinks will land best with your audience. But is its recommendation actually the best? That’s where you come in.

And this is where being an expert in your field becomes your superpower. You know your customers. You know how they speak, what they worry about, what makes them roll their eyes and what makes them lean in. That deep knowledge is exactly the thing AI doesn’t have. When you bring it to the editing process, picking, cutting, reshaping, you turn a generic draft into something that sounds like you and speaks to them.

The people getting flat, generic, “AI-sounding” results? They’re usually the ones accepting the first draft without applying any taste at all. The output is only ever as good as the judgement you bring to it.

4. The willingness to correct and refine

This one’s a mindset more than a skill, and it trips people up constantly.

A lot of people treat AI’s first response as the final answer. If it’s not quite right, they sigh, give up, and decide AI “doesn’t work for them.” But the first draft was never meant to be the finished product. It’s a starting point. A rough first draft from an eager assistant, not the finished article.

The people who get brilliant results are relentless about refining. They say “good start, now make it warmer.” “Cut the waffle.” “That intro is boring, give me three other options that will hook my audience into reading on.” “You’ve missed the point, here’s what I actually meant.” They treat it as a back-and-forth conversation, not a one-shot transaction.

Think about how you’d work with a talented new team member. You wouldn’t write off their first draft and give up, you’d point them in a better direction and let them go again. AI deserves the same patience, and rewards it generously.

Good briefing (skill one) means there’s less to fix here. But there’s almost always something. The refining is where the magic happens.

5. Clear thinking, or knowing what you actually want

Even the best-trained AI can’t read your mind on a specific task.

Here’s where this skill and effective briefing part ways. Briefing is the standing context you give AI about your world, your business, your customer, your voice. Clear thinking is the clarity you bring to this particular request, right now. And it’s a mirror. If your own thinking is muddled, the output will be too.

When you can clearly picture what you’re trying to achieve, who it’s for, what you want it to feel like and what “done” looks like, AI becomes genuinely powerful. When you can’t, no amount of training will save you because you’re asking it to hit a target you haven’t defined.

This is just the skill of thinking something through before you ask for it. If you’ve ever had to get clear in your own head before briefing a designer or explaining a project to your team, you already have this muscle. AI simply rewards you for using it.

Why this is brilliant news for you

Step back and notice what’s on that list. Effective briefing. Critical thinking. Taste. The willingness to refine. Clear thinking about what you actually want.

Not one of them is technical. Not one of them requires you to understand how the model works under the bonnet. And every single one of them is a skill you’ve been sharpening for years as a business owner who’s an expert in what you do.

This is why I get a little frustrated when AI gets framed as a tech thing, a young person’s thing, a thing you need a computer science degree to be good at. It’s the opposite. The experience, judgement and discernment you’ve built over a whole career are exactly what makes AI useful instead of useless. You’re not behind. You’re holding the most valuable cards in the deck.

The flip side is worth saying plainly too. The people who outsource their thinking entirely to AI, who accept whatever it spits out without question, will produce the same forgettable, generic content as everyone else doing the same thing. AI doesn’t replace your judgement. It makes your judgement more valuable, because it’s the only thing that separates your work from the flood of average sitting in everyone else’s feed.

Where to start

You don’t need to overhaul anything. The next time you use AI for your marketing, just shift how you treat it.

Before you ask for anything, give it the context it needs about your business and your customer. Then stop hunting for the perfect prompt and start having a conversation. Read everything it gives you with that slightly raised eyebrow. When something’s not right, say so, and ask for better. Bring your taste, your knowledge of your customer, your standards. Refine until it’s actually great. Something you would be proud of. Not just good enough.

Do that, and you’ll quietly become one of those people who “just gets AI.” Not because you learned to code, but because you brought the one thing AI will never have on its own… YOU!

You’ve already got what it takes. Now go and use it.

There’s a quiet trap most people are falling into with AI.

It’s not a bad-prompts kind of trap. It’s a much sadder one. They’re using AI to do the same boring things they’ve always done, just faster. The result? Vanilla content that nobody notices. Vanilla emails that nobody opens. Vanilla marketing that blends into the same beige soup as everyone else’s.

I refuse to live there.

So when I was asked to speak at Social Media Marketing World 2026 with two weeks’ notice, I did the opposite of what most people would do.

Here’s what I made: 

That’s a 1 minute and 14 second Disney-style showtune, lip-synced across Disneyland, with every line carrying a teaching beat about how to actually use AI in your business. The closer is a flip of Disney’s most iconic lyric… don’t “wish upon a star”, BE the star.

I made every single piece of it in partnership with AI. Note that I said “in partnership”. AI didn’t create it for me. We made it together. And the result much better thanks to that partnership. 

A one-week deadline and one slightly goofy idea

One week. That’s all I had to put together a completely bespoke presentation for SMMW26 that held its own among all the other brilliant speakers on the program. Just writing the talk was a stretch goal, especially when I was also tyring to finish all my work before taking an international flight to Anaheim. Most people would’ve left it there.

I’m not most people.

My presentation was titled “AI + YOU: Becoming the go-to-expert in your niche”, and I wanted to make a point that no slide could carry on its own. The point being that AI isn’t the magic. You are. AI is your strategic partner, enabling you to do things you couldn’t possibly do on your own. And I wanted to demonstrate that, not just say it. But how?

I got my presentation in on time before I left home, but then upon arrival in Anaheim, the home of the happiest place on earth, I had an idea…

I thought it would be fun to make a Disney inspired music video about AI marketing, lip-synced at Disneyland, and play it as the intro to my presentation. It was past the presentation submission deadline, but I decided to make it anyway. I figured if it was good enough, they would make an exception and let me do a last-minute adjustment to my presentation. 

I was right. 

Here’s how I did it…

Step one: Writing the song with Claude

I started with Claude, my main man when it comes to AI.

The brief was specific. A 60-second-ish upbeat Disney-style show tune that paired iconic Disneyland attractions with teaching beats about using AI in business. Funny and self-aware. Easy locations to find inside the park. Built around my actual signature phrases, not generic AI fluff.

What followed was eight versions of back-and-forth.

Version one was decent but generic. It leant into “AI is the magic,” which is the very framing I disagree with. So we flipped it. Version two became “AI isn’t the magic. You are. AI’s your Dream Team.” That’s when the song started to feel like me.

From there, every revision sharpened a different layer. We swapped a line that accidentally implied my audience shouldn’t follow other teachers (including me) for one that positioned me as their guide through the AI maze.

We changed the productivity-versus-profit message to a pivot about becoming the go-to expert in your niche, where profitability actually lives. We rewrote the closing three times until “Don’t wish upon a star. BE the star. You’ve got this!” landed without awkward repetition.

By version eight, the song was right. Every line did two jobs at once. It paired with what was behind me at Disneyland, AND it dropped one of my signature phrases on the audience’s head. AI Dream Team. Tool mastery beats tool collection. AI Visibility. Joy-zapping busywork. In partnership with AI. Human-led. AI-powered. AI Mega Success. You’ve got this.

You don’t get to that level of polish in one prompt. You get there in dialogue. I treated Claude like a thought partner, not a magic content button. And the song is sharper for it.

Step two: Bringing the song to life with Suno

With the lyrics locked in, I uploaded them to Suno along with a detailed style brief that Claude had also helped me write. The brief was also super specific. Upbeat Disney-style showtune. Bright major key. 125 BPM. Female lead vocal with a confident belt delivery. Playful, contrarian wink. Broadway polish.

I generated several versions. Some were too musical theatre. Some were too pop. One had the right melody but the wrong energy. I kept iterating, tweaking the style description, regenerating, until I landed on the version that gave me exactly the feel I wanted. Bright, cheeky, confident, with that classic showtune build that makes you want to sing along.

This is the part most people skip. They generate one version, decide it’s “good enough,” and move on. Good enough doesn’t make people stop scrolling. 

We don’t remember “good enough”.

Step three: Filming across Disneyland

Then I took the song to Disneyland.

Claude had also given me a clustered filming plan, mapping each lyric to a specific attraction, grouped so I wasn’t crisscrossing the park all day. What I loved most about that is that I didn’t even have to ask for this. Claude just offered this to me, as you would expect from a true strategic partner that’s got your back.

I walked the park with my headphones on, playing the Suno track on loop, lip-syncing in real time so I could nail the sync. I filmed every single line at least twice, sometimes five or six times because at the end of a long content creation day you do not want to discover that the one shot you needed didn’t quite work.

I also filmed everything in both landscape and portrait. I only needed landscape for the conference presentation, but I wanted the portrait versions banked so I could repurpose into Reels later. That’s an extra few seconds per shot that future Loren will thank me for.

I kept what I was doing quiet. The people I was visiting the park with picked up that I was making something, but they didn’t know what. I wanted the reveal at SMMW26 to be a genuine surprise, not a thing people had already half-seen or already heard about. 

Step four: Editing on the go

Here’s where most people would’ve gone wrong.

I edited the entire landscape version on my phone in CapCut, throughout the day, while standing in queues. I did not wait until I got back to the hotel that night.

Two reasons. First, I had a full conference schedule, combined with networking events that I wanted to be fully present for. If I’d left the editing until “later,” there would have been no later. Second, and more importantly, editing on-site meant I could see what was working and what wasn’t WHILE I was still in the park. If a shot didn’t quite land, I could re-film it. If I needed extra b-roll for the moments I wasn’t singing, I could go grab it. If I discovered a better location for a line, I could swap it.

This wouldn’t have been possible if I’d waited.

I also discovered something annoying about CapCut on mobile that turned into a useful constraint. It’s hard to drop in placeholder content and replace it later without messing with the lip-sync timing. So I had to film in order. Once I committed to that discipline, the editing got significantly faster.

I also realised the brief I’d worked on with Claude, as thorough as it was, hadn’t accounted for enough b-roll for the moments between sung lines. So I added that into my filming plan in real time, capturing extra establishing shots, location reveals, and atmospheric snippets to layer in.

The night I went back for two extra seconds

There was one scene I wanted to film that needed nighttime to truly land the line. So the next evening, I went back into the park, just to film what ended up being a two-second clip.

While I was there, I also re-shot a couple of scenes that I knew could be a fraction better than what I’d captured the day before.

Most people wouldn’t bother with any of that.

I am not most people.

What happened in the room

The video opened my session at SMMW26. It lifted the energy in the room before I even stepped foot on the stage. And then when I did make my entrance, wearing my Mickey Mouse ears (of course), it was to thunderous applause. And I loved it. 

I stepped on stage knowing I’d already won the room. The video had landed the message before I’d opened my mouth. The rest was easy. It felt like I’d earned my place up there before I’d said a word, with 500+ people in the audience already on my side, cheering me on.

Here’s some of what came back:

“When I listened to Loren Bartley, I said to myself: THIS is what I came to SMMW26 for!”

Linda Sherman Gordon

“Your session was mindblowingly brilliant. I just wanted to say thank you. It was the best session I’ve seen in the whole conference.”

Rufus Meakin

“Awesome speech. Well done. The work that goes into going to Disneyland and literally video taping every single section just to make the song work… Amazing. Loved it.”

Alan Thomas

The video set the scene. It made the philosophy land in a way that no bullet point ever could.

Why this matters (and what I want you to take from it)

Here’s the lesson I want you to walk away with.

Most people are using AI to do the same things they’ve always done, just faster. They’re using it to write the same emails, generate the same posts, churn out the same blog content, build the same generic landing pages. AI productivity, sure. But productivity doesn’t equal profitability. Vanilla doesn’t get you noticed. Vanilla doesn’t make you the go-to.

The big difference happens when you use AI to do things you couldn’t do on your own.

I couldn’t have written that song on my own. I’m not a Broadway lyricist. I couldn’t have composed and produced a 1 minute and 14 second Disney-style showtune on my own. I’m not a music producer. Plus, I am tone-deaf. I couldn’t have mapped all those attractions to the lyric beats, built a clustered filming plan, and edited the whole thing in a little over 24hrs from concept to completion. 

But I could do all of those things in partnership with AI, using the tools I already have access to with no additional budget (if you don’t include the ticket to Disneyland, as I was going anyway). And the result was something genuinely creative, super fun, and incredibly memorable. Something that landed a contrarian message about AI in a way that nobody in that room is going to forget.

That’s where AI gets exciting.

Not when you use it to be average faster. When you use it to be the version of yourself that you couldn’t be alone.

That’s how you stand out. That’s how you become the go-to expert in your niche. That’s how you make people stop scrolling and pay attention. And as an unexpected bonus, that’s how you have a ridiculous amount of fun while you do it.

So the next time you sit down to use AI, ask yourself a better question. Not “how can AI help me do this faster?” Try “how can AI help me do something I couldn’t do on my own?”

That’s the question that changes everything.

Don’t wish upon a star.

BE the star.

You’ve got this!

A few weeks ago, I ran an audit on my own business that made me want to crawl under a rock.

I teach service-based business owners how to build profitable marketing systems in partnership with AI.

That’s the work. That’s what I want to be known for. That’s what pays the bills.

But when I asked the AI tools what they actually knew about me, the story they told wasn’t that one. It was the old one. The Facebook Ads story. The “social media strategist” story. The story I’d been quietly walking away from for the past couple of years.

I scored 4.3 out of 10.

My first instinct, if I’m being honest, was to go back through a decade’s worth of content and start deleting things. All those Facebook Ads tips. The email marketing tutorials. The “Clubhouse Tips and Etiquette for Newbies” blog post that was genuinely useful in 2021 and slightly embarrassing now. Wipe them all out. Start fresh.

I’m glad I didn’t.

Why deleting your old content is the wrong move

Your old content isn’t a liability. It’s credibility scaffolding.

Those pieces prove you’ve been in the room for a long time. They prove you didn’t just discover the topic you’ve pivoted to six weeks ago and rebrand yourself as an expert overnight. They prove you’ve earned the right to an opinion by doing the work, year after year, across whatever the trend of the moment happened to be.

If you wipe all of that out, you’re not repositioning. You’re starting from zero. And nobody takes the brand-new expert as seriously as the expert who’s clearly been building for years.

So don’t kill off your old work to start afresh. Bury them alive instead.

The three-step method that actually works

Here’s the sequence that shifted my AI visibility score from 4.3 to 5.7 in four weeks. That might not sound like a lot, but hear me out, as I’m only partway through my journey of improving my AI visibility and I think the progress I’ve made in a relatively short time has been quite impressive, given the minimal effort it took.

I’ve ordered these the way I actually tackled them, which also happens to be the order that makes the most sense for impact and effort. Start with the quick wins you fully control. Build the long-term habit. Then round it out with the optional polish.

Step one: Update the platforms you control.

So where do you start? Before you create new content that reflects your positioning, you need to discover all the places you already show up and update them first.

Your LinkedIn profile. Your website’s About page. Your homepage hero copy. Your Facebook page intro. Your speaker bio. Every directory listing you’ve ever claimed. Every guest author bio on every publication you’ve ever written for. That last one might require a friendly email request to whoever is hosting that content.

If any of those still describe the old version of you, they’re actively working against the new version of you. An AI tool that crawls your LinkedIn and your homepage and gets two different stories doesn’t split the difference. It averages you into something generic and moves on.

The first thing I did here was use my Paige Platform Positioner Skill (that I gave away as a Custom GPT, Gem and Skill inside my most recent AI Marketing MegaClass on Build Your Authority Engine workshop). I used her to create consistent positioning statements matched to the exact character limits and formats of every platform I show up on. One coherent story, tailored to each platform.

Then I went and updated the places. LinkedIn first, because LLMs re-scrape it often and I wanted that signal moving as fast as possible. Then the static pages on my website. I’d known for months that this work needed doing, and I’d been quietly putting it off. The audit gave me the motivation I needed to finally get it done. It took less than a day of actual work, spread across a week of chipping away at it in small pockets of time.

This is the quickest play with the biggest upside. The platforms you already control are the lowest-friction fixes you’ll ever make.

Step two: Outproduce the old content with new content.

This is where the long-term work is. You create so much new content, so consistently, in your new positioning, that the new content becomes the loudest signal both to AI tools and to the humans looking at your blog, your LinkedIn feed, and your socials.

I started publishing a blog post every other day. Not because I’m a content-production machine, but because I’m using my own AI tools (Boris Blog Builder in particular) to assist me. It used to take me roughly half a day of writing time, stretched across a week, to produce a single blog post. Now I can go from idea to a published 1,500 to 2,000 word post on my site in under an hour.

That’s the part most people don’t realise. Burying your old content with new content sounds like a monumental task, but with AI in the mix, it really isn’t. The AI doesn’t replace the thinking or the voice, that’s still me. But it handles the research, the structure, the first draft, and the SEO scaffolding so I can spend my energy on the parts that actually need a human brain.

Even this post is part of the burying process. I say “process”, because positioning yourself as the expert in your niche is an ongoing job. You can do a great job flooding the zone initially, but you also have to keep going. AI favours recent content. The more you publish over time, the more fresh material the LLMs have to draw from when somebody asks them about your topic. So this isn’t a project with a finish line. It’s a practice.

Step three (optional): Reframe the old content.

This is where most people want to start, but it’s the third priority for a reason.

You can, if you want, go back through your old content and update it to connect more obviously to your new positioning. And of course, AI should assist you with this too, as it’ll speed the process up enormously.

But I’d hold off on this until you’ve made real progress with steps one and two. The impact from reframing old content isn’t nearly as significant as updating your owned platforms or producing new content in your current positioning. You’ll get more lift from a week of new blog posts than a week of rewriting old ones.

When you do get to it, prioritise the pieces still pulling traffic. If that five-year-old blog post you wrote is still bringing in loads of readers every month, that’s the one to reframe first. It’s already doing the hard work of attracting an audience, so redirecting those readers towards your current offers and positioning is a high-leverage move. The pieces nobody’s reading anymore can wait, or stay exactly as they are. They’re still playing their credibility scaffolding role in the background.

I haven’t even started this step yet, and I made the right call not to prioritise it. My dial moved further, faster, from steps one and two than it would have from the reframe work.

What actually moved the needle

When I re-ran the audit four weeks later, the score had jumped to 5.7 out of 10. Clarity up one point. Consistency up one point. Corroboration, the hardest signal of the three because it depends on other people, up two points.

And here’s what I didn’t expect…

Without me chasing them down, three new podcast invitations and two speaking gigs landed in my inbox after I started making these changes.

None of this happened because the LLMs surfaced me. It happened because real humans in my existing network were paying attention to the content I was putting out, and THAT started opening doors. But these new opportunities as a result of this strategy are providing me with corroboration signals that are already starting to make an impact. The full AI visibility benefits will probably take longer to land. But, I didn’t have to wait for the LLMs to catch up for this strategy to start working.

This is the benefit nobody tells you about. The work you do to show up better in AI tools is the same work that helps you show up better to the people who matter most in your business. Clients. Peers. Event organisers. Podcast hosts. The algorithm is one audience. Your network is the other. You’re building for both at once.

Now that I’ve seen how quickly the corroboration signal can lift, I’m going to be a lot more proactive about creating those opportunities rather than waiting for them to find me.

The bonus nobody tells you about

The re-audit also showed me where I still have work to do. I thought I’d nailed the website repositioning. Turns out the homepage hero still read like the old me, and the AI tools picked that up immediately. Of course, I fixed that immediately.

That’s the single most important lesson in all of this… You don’t know until you check. The initial audit matters. The re-audit matters just as much. And the one after that. Because repositioning isn’t a project you complete. It’s a practice you maintain.

The other thing worth noting is that it doesn’t take a huge amount of effort to move the dial. Four weeks. One positioning exercise. Less than a day on the updates. A handful of blog posts. In partnership with AI, of course (my knowledge, my stories, and my voice).  And I’m already seeing significant results. The work is doable. The missing piece, for most people, is knowing where to put the effort in the first place.

That’s what the audit gives you.

Want to see where you actually stand?

If you’ve been quietly repositioning your business, and you’re not sure whether the work is landing, get yourself an AI Visibility Audit. It’s the same audit I ran on myself, the one that scored me 4.3 and gave me the map I needed to shift the dial.

You’ll learn exactly what the AI tools are saying about you right now, what’s pulling your score down, and what to fix first. No guessing. No wasted effort. Just a clear picture of where you actually stand, and what to do next.

Get your AI Visibility Audit here. 

Because you can’t bury the bodies if you don’t know where they are.

How to clone your knowledge, so your marketing sounds like you

If you’ve been using AI in your marketing and still spending more time editing than creating, that is not just a prompting problem.

It is usually a knowledge problem.

Most service-based business owners are trying to make AI sound like them before they have given it anything meaningful to think with. So they open ChatGPT, ask for a post, an email, or a sales page, and get back something that is technically fine but personally off. It sounds polished enough, but not like them. Not like the way they explain things. Not like the way they think. And definitely not like the person their clients actually hired.

That is where a lot of people get stuck. They assume the answer is a better prompt. Better instructions. A few more examples. Maybe even a voice clone or an avatar.

Lovingly warned, that is backwards.

If you want AI to help you create marketing that still sounds like you, the first thing you need to clone is not your face or your voice.

It is your knowledge.

 Why better prompts are not fixing the real problem

One of the biggest misconceptions I see is this idea that AI already knows everything and all you need to do is prompt it properly.

That sounds convenient, but it is exactly why so many people keep ending up with generic output.

Your prompts only work as well as the knowledge behind them. Without your own thinking, context, stories, beliefs, and expertise in the mix, AI fills the gap with average internet language and recycled patterns. That is when you end up with content that feels flat, vague, or painfully generic. The kind of output that makes you think, “I may as well have written this myself.”

And that creates a frustrating cycle.

You ask AI for help. It gives you something half-right. You rewrite it. You soften bits. You add your own language back in. You fix the advice. You pull out the fluff. Forty minutes later, you have saved yourself ten.

That is not real leverage.

That is AI acting like a slightly faster version of the old bottleneck.

The issue is not that AI cannot write.

The issue is that it does not know enough about you to think in a way that produces useful work on the first pass.

The real asset is not your avatar. It is your thinking

A lot of people hear the phrase “clone yourself” and immediately think of voice clones, avatars, or digital likeness.

That is the flashy version of the conversation. It gets attention. It looks impressive. But for most service-based business owners, it is not where the real value sits.

The more valuable thing to clone is your knowledge.

Your judgement. Your explanations. Your decision-making. Your values. Your lived experience. The way you connect dots. The way you make calls. The way you talk clients through a problem. The way you know what matters and what does not.

That is the part people actually pay you for.

And that is the part most people skip.

When you clone your knowledge well, the upside is not just “more content”.

You get better quality output across content, strategy, products, and decisions. You get faster execution.

You reduce the amount of your intellectual property trapped in your head. You lower key-person risk. You create more capacity because your thinking can start working without you being present for every single task.

That is a much better bang for your buck than creating a digital talking head that still does not actually think like you.

There are two types of knowledge, and most people skip the one that matters most

If you want AI to sound like you, there are two types of knowledge you need to understand.

The first is core knowledge.

This is who you are. Your identity, values, beliefs, stories, opinions, priorities, and the way you think. It is the deeper layer. The stuff underneath the tactics.

The second is task knowledge.

This is what you do. Copywriting. Customer support. Ads. Planning. Admin. Sales follow-up. Content creation. The role-specific tasks that AI can support once it understands the job at hand.

Most people go straight to task knowledge.

They build a GPT for writing captions. Or emails. Or proposals. Or customer support. But they never do the harder work of capturing the core knowledge underneath it all.

So the tool can perform the task, but it still cannot sound like them.

That is why the content feels close, but not quite right. It may follow instructions. It may even sound polished. But it does not carry the weight of real perspective.

Because the thing that makes your marketing sound like you is not just your wording.

It is the thinking behind the wording.

You are not just training AI what you know.

You are training it how you think.

The unsexy step that determines whether AI sounds like you or not

Here is the part most people want to skip.

Processing.

Not collecting. Not building. Processing.

This is the unsexy but most important step, and it is usually the one that takes the most time.

Processing is not just tidying up documents.

It is deciding how your knowledge should be grouped, interpreted, prioritised, and used. It is removing duplication. Resolving contradictions. Cutting outdated thinking. Deciding what is core. Stripping out fluff. Making sense of messy raw material so that AI has something clear and useful to work with.

This is where people often go wrong.

They dump transcripts, notes, folders, websites, random documents, and years of half-finished thinking into a tool and hope it will sort it all out. But when you overload AI with irrelevant information, it struggles to prioritise what matters. Important information gets drowned out by noise. You lose precision. And the output gets worse, not better.

AI can help clean and summarise.

But humans still need to decide what matters.

That judgement call is everything.

A simple way to build your knowledge clone

You do not need to overcomplicate this.

A practical build path looks like this:

1. Collect

Capture the raw materials.

This is not about perfection. It is about gathering your unique source material in one place. Your stories, opinions, values, notes, podcast interviews, blogs, emails, training content, voice notes, transcripts, and lived experience. Focus on what is yours, not generic information anyone could Google.

2. Process

Curate, clean, and align it.

This is where you organise what belongs together, remove what no longer fits, and identify the pieces that actually represent your core thinking. This stage cannot be rushed, because the quality of everything downstream depends on it.

3. Build

Integrate the knowledge into your AI tools.

This is where your processed knowledge gets put to work. The mistake here is building one giant brain and hoping it can do everything. A better approach is to think in mini-brains: one core knowledge base for who you are, then separate task-specific knowledge bases for different jobs.

4. Test

Never assume it learned correctly.

Testing is not just asking, “Is this right?” It is also asking, “Would I say this?” “Would I make this call?” “Does this align with my values and priorities?” If the answers are off, correct it, add better examples, and keep training through feedback.

5. Use

Put it into real work.

A knowledge clone that never gets used is just a filing system. The value only shows up when your knowledge is embedded into content, emails, decisions, planning, and day-to-day execution.

6. Refine

Keep evolving it.

Your knowledge clone should change as you change. As your positioning shifts, your thinking improves, or your beliefs evolve, your knowledge base should evolve too. That is how the value compounds over time.

You don’t win by creating the most content. You win by being the clearest, most consistent, and most corroborated voice on a specific topic.

What this changes in the real world

This matters because knowledge trapped in one person’s head creates bottlenecks everywhere.

A good example from the interview was Beauty and the Bee’s customer support. The owner knew the product range deeply and could quickly guide people towards the right shampoo or conditioner bars based on their needs. But everyone else had to piece answers together from FAQs, emails, notes, ingredient lists, price lists, blogs, and other scattered information. So the owner kept becoming the default person for answers.

That is not just a customer service problem.

It is a scale problem.

When key knowledge only lives in your head, your business slows down around you.

Marketing slows down. Decision-making slows down. Delegation slows down. Quality control gets harder. And every “quick question” comes back to you.

Cloning your knowledge does not solve everything overnight.

But it does start shifting your role. Instead of being the only source of truth, you become the builder of a system that can carry more of your thinking forward consistently.

That changes the game.

The mindset shift a lot of business owners still need to make

Some people resist this whole idea because they are protective of their intellectual property.

I understand that.

But the answer is not to ignore what is happening and hope the shift goes away. The smarter move is to be intentional. Put protections in place. Turn off training on your data where possible. Ring-fence knowledge inside the tools and environments you choose to use.

At the same time, it is worth being honest.

Your knowledge is already more visible, searchable, and inferable than most people think.

The real risk is often not that AI will know too much about your business.

It is that your business will stay overly dependent on you because you never took the time to capture what only you know.

This is not sexy work.

But it is some of the highest-impact work you can do if you want AI to be genuinely useful in your business

Clone your knowledge before you chase polish

If your AI content still does not sound like you, I would not start by rewriting your prompt library.

I would start by asking a better question:

Have I actually given AI enough of my thinking to work with?

Because that is the real shift.

AI does not become useful just because it writes faster.

It becomes useful when it understands the knowledge behind your marketing. Your perspective. Your decisions. Your stories. Your standards. Your way of seeing the problem.

That is when the editing drops.

That is when the output improves.

That is when your marketing starts sounding like the person your clients hired.

Watch or listen to the full interview

If this has hit a nerve, I discussed this topic in much more detail on the AI Explored Podcast with Mike Stelzner. The following full interview goes deeper into what cloning your knowledge actually looks like, the mistakes people make, and how to start building it properly.

Watch or listen to the full interview now on Apple Podcasts, Spotify, YouTube Music, YouTube or Amazon Music.

Because the sooner you stop trying to make AI perform without the right foundation, the sooner it can start doing work that actually sounds like you.

Most business owners I talk to treat their AI subscription the same way they treat their Netflix account. It’s a monthly cost to keep as low as possible. They’ll use the free tier until it runs out. They’ll hit their usage limit and switch to something else while they wait for it to reset. They’ll spend twenty minutes Googling a workaround rather than upgrading to the next plan.

And they’ll tell themselves they’re being smart with their money.

They’re not. They’re losing far more than they’re saving, and most of them don’t even realise it.

I know this because I used to do exactly the same thing.

The Saving That’s Actually Costing You

Here’s a scenario that might sound familiar. You’re deep in a workflow. Writing content, building a strategy, refining a proposal. And then your AI tool hits its usage limit. You get that message: “You’ve reached your limit. Try again in a few hours.”

So you stop. You switch to email, or jump onto social media, or start another task entirely. When the limit resets, you come back to where you left off and try to pick up the thread.

Except you can’t just pick up the thread. Not really.

Research from the University of California, Irvine, found that it takes an average of 23 minutes and 15 seconds to fully regain focus after an interruption. That’s not 23 minutes of feeling a bit distracted. That’s 23 minutes before you’re back to the level of deep thinking you were at before you stopped.

If that happens twice in a day, you’ve lost nearly an hour of your most productive thinking. And that’s not counting the actual downtime waiting for the usage window to refresh. Over a week, that’s hours of deep, creative, strategic work gone. Not because you weren’t capable of doing it, but because you chose only to save a few dollars on your AI subscription costs.

I learned this the hard way. I was using Claude on the lower tier and working until I hit the limit, then switching to other tasks until the window refreshed. It felt efficient at the time. I was still “working,” just on different things. But the reality was that every time I switched, I lost momentum. I’d come back to a half-finished blog post or a strategy I was in the middle of developing, and it would take me ages to get back into the right headspace.

The real wake-up call came when I was running a Click Engage Convert Academy workshop. I was demonstrating how to use AI for content creation, walking the group through the process live, step-by-step. And right near the end of the process, I hit my usage limit. Luckily, I already had a finished example of what we were building together that I could share with the group. But it completely broke the flow. I couldn’t take them from start to finish in a single session, which made what I was teaching far less effective. That moment made it crystal clear: the “saving” wasn’t saving anything. It was costing me credibility, momentum, and the quality of the experience I was delivering.

When I upgraded my Claude subscription, that credit anxiety vanished immediately. I stopped rationing my usage, stopped interrupting my own workflow, and my productivity went through the roof.

What You’d Actually Pay a Human to Do This

Let’s talk about what this actually costs when you compare it to the alternative.

If you were to hire a freelance copywriter in Australia to write a single SEO-optimised blog post, you’d be looking at somewhere between $500 and $1,500. That’s one post. And a good copywriter charges $125 to $240 an hour, so even smaller tasks add up fast.

Now compare that to an AI subscription at $20 a month. Even the premium tiers sit around $100 to $200 a month, which is still less than a single outsourced blog post.

Let that sink in: one outsourced blog post can cost you more than an entire year of an AI Pro subscription.

And that’s just copywriting. These tools also help with strategy, research, brainstorming, email sequences, social content, and loads of other tasks that would normally require different specialists. Not to mention the agentic capabilities. When you start adding up what you’d pay humans for all of that, the comparison is almost laughable.

I used to rely on team members for tasks like content creation. The process would go something like this: brief the team member, wait for the first draft, review it, send back feedback, wait for revisions, review again, maybe send back more notes. A single blog post could take up to a week to go from brief to final version. I didn’t track the actual staff cost at the time, but when I think about the hours involved on both sides, it was significant.

Now I can do the equivalent in about 30 minutes using Claude Cowork. From idea to researched, written, optimised, and published. That’s not an exaggeration. That’s what I do every week.

The Access Most Small Businesses Never Had

What gets missed in most conversations about AI costs is that for a lot of small business owners, the comparison isn’t AI versus a copywriter. It’s AI versus nothing.

Many small business owners have never been able to afford a professional copywriter. They’ve never had a strategist on call. They’ve never had someone to bounce ideas off, refine their messaging, or help them create content that actually sounds polished and professional. Those services, even at the lower end of the market, are thousands of dollars a month. And that’s for businesses that can afford to invest in content at all.

For the business owner who’s been writing their own social posts at 10pm, agonising over their email newsletter, or putting off their blog because they don’t have the time or confidence to write it, AI at $20 a month isn’t replacing a team. It’s giving them access to a level of support and capability they’ve simply never had. And it means they might actually get those evenings back, instead of sitting up late trying to do it all themselves.

That’s not a cost. That’s a completely new category of opportunity.

And the businesses that are leaning into this are seeing it in their results. They’re producing more, producing better, and doing it faster than they ever could before. Not because they’ve suddenly become better marketers, but because they’ve got a capable partner working alongside them for the first time.

What You’re Not Doing While You’re Penny-Pinching

There’s another cost that almost nobody calculates, and it might be the biggest one of all: opportunity cost.

Every hour you spend doing a task that AI could handle in minutes is an hour you’re not spending on revenue-generating work. Every blog post that takes you half a day to write is half a day you could have spent on client work, sales conversations, product development, or strategic planning.

Think about your own week. How many hours do you spend on tasks that AI could help you complete in a fraction of the time? Even if it’s just four or five hours a week, that adds up to nearly five full working weeks over a year. Five weeks you could spend on client work, business development, or the projects that actually move the needle. Or, honestly, just use that time to go on a holiday, spend it with your family, or do all those non-work things you keep saying you’ll get around to but never do.

Now ask yourself: what’s an hour of your time worth when you’re doing the work only you can do? Whatever that number is, multiply it by the hours you’d get back, and then compare it to a $20 monthly subscription. The gap is enormous.

And this isn’t just about earning more. It’s about doing less of the work that drains you and more of the work that energises you. Most business owners didn’t start their business because they love writing social media captions or formatting newsletters. They started because they’re brilliant at what they do, and AI gives them the space to actually do it.

The Quality Multiplier Nobody Talks About

This is the part that surprises most people because the assumption is that AI-assisted work is somehow lesser. Cheaper, faster, sure. But not as good.

In my experience, it’s the opposite.

When I’m working with AI as a coworking partner, not just delegating to it, but genuinely collaborating, the quality of what I produce goes up. I can bounce ideas off it. I can get a first draft down in minutes and then spend my time refining, shaping, and adding my experience and perspective. I can ask it to challenge my thinking, find gaps in my argument, or suggest angles I haven’t considered.

The result is content that’s more thoroughly researched, better structured, and more strategically considered than what I’d produce on my own. Not because the AI is smarter than me, but because having a thinking partner makes everyone’s output better. It’s the same reason great editors make great writers even better.

But there’s an even bigger knock-on effect that occurs because when the quality of your output improves:

You can charge more. Clients and customers can see and feel the difference between rushed, thrown-together content and polished, strategic work. Higher quality positions you as more professional, more capable, and more valuable, which means you can command higher prices for your services.

Your clients are more satisfied. When you deliver better work, faster, clients notice. They get better results, they feel better supported, and they stick around longer. Customer satisfaction isn’t just a nice feeling. It’s the foundation of a sustainable business.

Referrals increase. Satisfied clients talk. They recommend you. They share your content. One piece of genuinely excellent content can lead to a chain of new business that far outweighs the cost of the tool that helped you create it.

The compounding effect kicks in. Better content leads to more visibility. More visibility leads to more opportunities. More opportunities lead to more revenue. More revenue means more capacity to invest in your business. It’s a virtuous cycle, and it starts with producing better work, more consistently.

How to Calculate Your Own AI ROI

If you’re still on the fence, here’s a simple framework to work out what your AI subscription is actually worth to you.

Step 1: Calculate your replacement cost. What would you pay someone else to do the tasks you’re using AI for? Get quotes for a copywriter, a virtual assistant, or a strategist. Even rough numbers will do. If you’d pay a copywriter $500 for a blog post and you’re producing four posts a month with AI, that’s $2,000 a month in replacement value from a $20 subscription.

Step 2: Calculate your time savings. Track how long tasks take you with AI versus without. Multiply the hours saved by your hourly rate (or what you could earn in that time). This is your opportunity cost recovery.

Step 3: Factor in the quality premium. If AI helps you produce better work, estimate what that means for your pricing, retention, and referrals. Even a conservative estimate adds up quickly.

Step 4: Add up the real cost of interruptions. If you’re hitting usage limits and losing 23 minutes of focus each time, calculate how many times that happens per week. Multiply by your hourly rate. This is the hidden tax you’re paying by staying on a cheaper plan.

When you add it all up, most business owners are genuinely shocked by the gap between what they’re paying and what they’re getting back. And that’s before you factor in the intangible benefits: reduced stress, better work-life balance, and the confidence that comes from consistently producing work you’re proud of.

Stop Asking What It Costs. Start Asking What It’s Worth.

I see three levels of resistance to investing in AI tools, and all three are costing the people who hold them.

The first is the business owner who won’t pay for AI at all. They’re using free tiers only, which means they’re stuck with older, less capable models, severely limited usage, and none of the features that actually make these tools useful for real work. They’re making decisions about AI based on the worst possible experience of it. That’s like test-driving a car with the handbrake on and deciding cars aren’t worth buying.

The second is the business owner who’s paying for an AI subscription and refusing to upgrade. They’re using AI constantly, getting real value from it, and then hitting their usage limit right when they need it most. They’re effectively choosing to stop working mid-flow rather than invest another $80 or $180 a month. When you compare that cost to the value of the output they’re losing, it makes no sense at all.

The third, and this one is increasingly common, is the business owner who has one AI subscription and feels they can’t justify adding a second tool. But the reality is, different AI tools have different strengths. The features that a second or third tool brings to your workflow might be exactly what takes your business to the next level. You wouldn’t hire one person and expect them to do every single role in your business, yet that’s what people do with AI tools. Sometimes the thing standing between you and a real breakthrough is an extra $20 a month on a tool you haven’t tried yet.

Your AI subscription isn’t a line item to minimise. It’s an investment that, dollar for dollar, is likely giving you a better return than almost anything else in your business.

So the next time you find yourself hesitating over an upgrade, or talking yourself out of trying a new tool, ask yourself this: what’s it costing you not to?

The answer, when you actually do the maths, will almost certainly surprise you.

The businesses that will look back on this year as the year they fell behind are already identifiable. They’re the ones still “exploring.”

There is a specific kind of business owner I’ve been watching over the past 18 months. They’re curious about AI. They’re genuinely interested in getting better results. They have tried a few tools. They have a ChatGPT account. They have bookmarked approximately 47 articles about AI for marketing and read approximately four of them.

They’re experimenting. And they feel like they’re keeping up.

They are not.

The businesses pulling ahead are not the ones who know about AI. They’re the ones who have built AI into the operating rhythm of their business. The ones who have moved from “trying things out” to “this is just how we work now.” That is a fundamentally different place to be, and the gap between the two is widening every single week.

I am not writing this to scare you. I am writing this because the window for catching up is still open, but it will not stay open indefinitely. And the good news is that building an AI-powered business is far more accessible than most people think. It does not require technical expertise. It does not require a big budget. It requires a shift in approach.

Here’s what that shift looks like, and why it matters more urgently now than it did even six months ago.

Experimentation Has a Ceiling

I see this pattern constantly, and I recognise it because I lived a version of it myself in the early days.

You try an AI tool. It produces something that is pretty good. You’re impressed. You use it a few more times, get a few more decent outputs, and then life gets busy, and you stop using it consistently. A month later, you try a different tool because you saw it recommended somewhere. The cycle repeats.

That is experimentation. And it produces exactly what you would expect from experimentation: occasional wins, occasional frustrations, and no lasting change to how your business operates.

The frustrating thing about this cycle is that it feels productive. You’re learning. You’re staying current. You’re trying things. But when you step back and ask, “has this changed how my business runs?” the honest answer is usually “no”.

Here’s the ceiling I am talking about. Experimentation gives you familiarity. It gives you a rough sense of what AI can and cannot do. It might even give you the occasional piece of content that saves you an hour or two. But it does not give you leverage. It does not free up your time in a way that compounds. It does not change your capacity in a meaningful or sustainable way.

Systematisation does that. And that’s where most small business owners have not yet arrived.

The businesses that have arrived there are not necessarily smarter or more tech-savvy. They just made a different decision: instead of continuing to explore, they picked something specific, built a system around it. Then they did it again for something else. That compounding process is what separates the businesses pulling ahead from the ones standing still.

The other part of this problem is the cost of not acting. Most business owners think that “not yet” is a neutral decision. It’s not. Every week you’re running a process manually that could be handled by a well-designed AI workflow, you are spending time. And time, as I keep reminding the business owners I work with, is the most expensive thing in your business. Not money. Time. You can make more money. You cannot make more time.

But what if you got 10 hours of that time back every week? What would change?

The Gap Is Measurable and Growing

The data on this is compelling, and the trajectory is clear.

In Australia, the shift is visible and accelerating. Research from the Australian Department of Industry found that 40% of small and medium businesses had adopted AI tools by the end of 2024. A 2025 BizCover study put the figure even higher: 80% of Australian small businesses are now either actively using AI or planning to implement it. That’s the mainstream. That’s not a niche of early adopters anymore.

But here is the number that matters most to me: a Xero study of 500 Australian small businesses found that 57% of businesses reporting revenue growth use AI tools weekly, while those with flat or declining revenue are adopting AI at significantly lower rates. That’s not a coincidence. That’s a pattern. The businesses doing better are, at a disproportionate rate, the ones that have moved from exploration to integration.

Research from COSBOA published in 2025 adds an important nuance to that picture. While 30% of Australian small business owners report using AI daily, only 14% have integrated AI into their core business operations. That gap between daily use and genuine integration is exactly what I’m talking about. Familiarity with a tool is not the same as building a system around it. And it is the system, not the tool, that creates the compounding advantage.

There’s also a compounding dynamic at work that’s easy to underestimate. A business that built its first AI workflow in January 2025 did not just save time in January. Every month since then, that workflow has been running. Every month, the person who built it got slightly better at briefing AI, slightly faster at using the outputs, and slightly more confident about building the next workflow. That’s compounding in the truest sense.

The business that is still experimenting in April 2026 has not been standing still for a year. They’ve been falling further behind a moving target.

I’m not saying this to be alarming. I’m saying it because the solution is genuinely straightforward. You don’t have to build 12 AI workflows at once. You need to build one properly, see what it changes, and then build the next one. That process, done consistently, is what creates the compounding advantage. And it is available to you right now, regardless of where you are starting from.

From Experimentation to Systematisation

The shift from experimentation to systematisation is not a technology shift. It’s a thinking shift.

It starts with a different question. Instead of “what can I use AI for?”, start asking “what process in my business is still running at manual speed that doesn’t need to be?”

That is the question I walk every client through when we start building their AI Dream Team. It’s a question about your business, not about technology. And most business owners, when they sit with it honestly, can identify three to five processes immediately. Writing first drafts of content. Pulling together client research. Drafting emails. Creating social media posts. Repurposing long-form content into shorter formats.

These are all processes that are already well within what AI can handle reliably, right now, with the tools that are already available to you.

The reason most business owners have not systematised these processes is not that they lack the tools. It is that they have not taken the step of building a proper brief and workflow around the handoff. That’s the difference between using AI occasionally and using AI systematically.

A proper handoff includes four things.

1. A clear brief – for AI that explains the context: who you are, who your audience is, what the goal of the content is, and what your voice sounds like.

2. A defined process – where does the task start, what does AI handle, where does it hand back to you, and what does “done” look like?

3. A feedback loop – how do you improve the brief when the output is not quite right? 

4. A decision – to stop touching these tasks manually once the system works.

That fourth point is the one that most people skip. They build the system, use it once, and then go back to doing it manually when they’re in a hurry. Systematisation only stacks up if you actually use the system. Consistently. Not just when you have time to.

I started building my own AI Dream Team with a single, specific task: writing the first draft of my weekly email newsletter. Once I had a brief that AI could work from reliably, and once I had committed to using it every single week rather than only when I remembered, I got that time back permanently. Then I moved to content repurposing. Then to client research. Each one added capacity. Each one compounded.

That’s the path. Not a dramatic transformation overnight, but a steady, deliberate process of handing off the right things to AI and getting your own time back in return.

How to Start Systematising This Week

You don’t need a plan for every AI tool in existence. You need a plan for one task. Here’s how to build it.

1. Audit your weekly tasks for manual speed.

Write down everything you do in a typical week that feels repetitive, time-consuming, or like it does not strictly require your expertise. Be honest. You are looking for the tasks you do out of habit, not necessity.

2. Rank them by time cost.

Which of those tasks takes the most time per week? That is your starting point. Not the most interesting one, and not the one you are most curious about. The most expensive one.

3. Build a brief before you open the AI tool.

Before you touch ChatGPT or any other tool, write down: who you are, who your audience is, what the task is, what the output should look like, and what your voice sounds like (include a sample of your own writing). This brief is the foundation of your system.

4. Test the brief with a single task.

Run the task through AI using your brief. Evaluate the output. Note what was close and what was off. Refine the brief. Run it again. Repeat until the output consistently meets your standard.

5. Document the process.

Write down the steps: what you type, what AI produces, how you review it, and where it goes next. This documentation is your workflow. It means you can hand this process to a team member later, or return to it after a holiday without rebuilding it from scratch.

6. Commit to using it.

Decide now that you will use this workflow every single time this task comes up. Not sometimes. Every time. Systematisation only compounds if you actually use the system.

7. Build the next one.

Once the first workflow is running smoothly, identify the next most expensive task and repeat the process. That is the whole strategy. One at a time. Consistently.

Frequently Asked Questions

How do I know which AI tool to start with?

Start with the tool that best fits the task you are systematising first, rather than the tool that has the most features. For most written content tasks, ChatGPT or Claude are solid starting points. The tool matters less than the brief you give it. A great brief with a basic tool will almost always outperform a vague brief with the most sophisticated model available.

What if I try to build a system and the AI outputs are still not good enough to use?

In almost every case, the problem is the brief rather than the tool. Before deciding a process is not suitable for AI, revisit your brief and ask whether you have given AI enough context: your voice, your audience, the specific outcome you need, and a clear example of what “good” looks like. Briefing AI properly is a skill that improves with practice.

How long does it take to see a meaningful time saving?

Most business owners who commit to building one well-structured AI workflow report a noticeable time saving within the first two to three weeks of consistent use. The brief takes time to build and refine, but once it is working, the time saving is immediate and repeatable every time that task comes up.

Is this relevant for small service-based businesses, or mainly for larger companies?

It is particularly relevant for small service-based businesses. The time savings that matter most are in the tasks you handle personally: writing, researching, communicating, planning. Those are exactly the tasks AI handles well. You do not need a team or a tech department to get started. You need a clear brief and a commitment to using it consistently.

Will my content still sound like me if I use AI to write it?

Yes, but only if you build the right foundation. AI produces content in the voice and style you give it. If you brief it with your actual writing samples, your specific audience, and your particular way of explaining things, the output will sound close to you. If you give it a generic brief, it will produce generic content. The voice is in the briefing, not in the tool.

The Window Is Still Open. For Now.

There’s a moment in every technology cycle when the early advantage shifts from “interesting” to “structural.” We’ve already passed that moment with AI.

The businesses that moved first are not just ahead because they started earlier. They’re ahead because they’ve been compounding that advantage, quietly, every single week. Each workflow they built made the next one easier. Each hour they got back went into building something else. The gap is not a few months. It is a compounding system running at pace.

But here’s what’s important to know… you have not missed this opportunity. The window for catching up is still open, and it’s more accessible than the noise around AI would have you believe.

You don’t need to understand how large language models work. You don’t need a technical background. You don’t need to clear your diary and spend a week doing an AI overhaul of your entire business.

You just need to pick one task. Build one brief. Commit to using it consistently. And then do it again.

That’s the whole path. And every week you walk it, you’re closing the gap.

The question is not whether AI is right for your business. The question is which task you’re going to hand off this week.

Start there.

You’ve got this!

About the Author Loren Bartley is an AI marketing educator and founder of Impactiv8, a Melbourne-based digital marketing education business. She teaches service-based small business owners how to build profitable marketing systems powered by AI through her AI Marketing MegaClass, Click Engage Convert Academy, and Lead Activator platform.