Portent Interactive, Seattle, WA

Dear Yahoo, Please Stop Showing My Ads for EVERYthing

2009-10-30T16:29:33Z

Dear Yahoo,

While I admit that I was pretty annoyed with the Google session based keyword searches, I have to say that Yahoo has them beat. Ever since they advanced "advanced" matching that's been taking over in PPC accounts, keywords that used to see 3,000 impressions in a month are now seeing 3,000 within a week.

Why?

My Yahoo reps explained to me that it had to do with increasing traffic and trying to tell me that I was getting more for my money. What I actually find is that is quite the opposite. Based on the back end keyword query reports I have been requesting to see exactly what is triggering my ad, I'm building quite the negative keyword list. The key difference between these reports is that at least with the Google session based keywords I saw a couple of conversion and the frequency was far less.

Yahoo seems to be pumping out my ad at every chance they get. Include the word "free" anywhere in your keyword and sky is the limit, damn the relevancy. Change the match type to standard over advanced and cut off the traffic almost completely for anything over two words. Can't we have a medium or phrase match? Or at least a report I can run on my own to see what kinds of highway robbery might be occurring? The answer from Yahoo is to add to you negative keyword list, but just like with Google, should I really have to be gleaning through lines and lines of data constantly? Couldn't you just make your targeting better? (Hate to state the obvious here, ahem.)

As with the post on Google's session based matching, I bring you the highlights from a Yahoo report:

The keyword is "free bicycle"

The winners are:
bicycle trial
bicycle solitare free
animated bicycle
bicycle clip art free
bicycle blank
free old
freeflite
your free atlanta
bike funny

Add dynamic insertion ads to this and you've got yourself quite the mess. What this tells me two things, while the reports are chock full of potential long tail, VERY low traffic keywords, there are also full of some wacky queries. If you have to used advanced match in Yahoo, the lesson seems to be to add each word of your whole keyword as a negative.

For another account about a first grade math game, I had to add negative keywords of "game" "first grade" and "math"." If I choose the standard match for that keyword I end up with almost no impressions, which is just another example of if Yahoo doesn't get it together, they will be MicroHoo very soon and without the "hoo."

Since they launch of their new ad campaign for "It's You" I have yet to see how this will help me find what I am looking for if I type in "free bicycle in seattle" and come back with ads for a "free old" or a "animated bicycle."

Sorry Yahoo, it's not me. It's YOU.

The 10Things Site Review, by Portent

2009-10-07T23:20:47Z

We're expensive, I know.

A long-term engagement with a quality Internet marketing agency is pricey. So we've developed a 1-on-1 consulting package that lets you get started without breaking the bank, and without a long-term retainer contract.

As part of the program, you get:

A custom report listing 10 things you can do to improve your site.
1 hour of 1-on-1 consulting with me, via phone, to explain the recommendations.

If you're a blogger, a small business or just want to test out Portent's expertise, the 10Things package is perfect.

Click here to learn about it, and to sign up.

Shout-Out: Hanapin Marketing's $16,000 SEM Sweepstakes

2009-09-28T20:19:48Z

Our good friends over at Hanapin Marketing - you know, the folks who run the PPC Hero blog - are running a pretty amazing SEM Sweepstakes right now.

They have some great prizes available, including:

(3) Paid Subscriptions to SEOMoz
(1) 3-Month Subscription to ClickEquations (a neat bid management tool that we use at Portent)
(100) Various Internet Marketing Tomes
(1) Full PPC or SEO Audit by the Hanapin Marketing Team (worth $4,000)

What are you waiting for? Go sign up! They'll start randomly selecting winners on October 12th, 2009.

Dear Google, Please Give Me Back My Money

2009-09-23T21:03:37Z

Dear Google,
I was going through my search query reports the other day and noticed that more and more I am seeing these broad match (session based) results. Interesting. I know that there is no opt out button for this and like my fellow SEMs am in a bit of an annoyed snit, but since it's now rampant through one of my accounts, I'd like to just straight out ask for a refund.

My intended audience would be looking for a math game for first graders. I have included many negative keywords in my campaigns (And yes, I thought of adding more based on these session queries, but adding 30 or so negatives every other day trying to keep up just seems silly and waste of my time, you see.) and while a few of these session based queries were apt, their conversion and click through rates are not.

Here are some examples of session based keyword queries that I had to pay for on an account about first grade math:

math
math com
colorful math
math games grade
math partners for 5
math sheets
starfall math
negative number math
relearn math

and my personal favorite: 9th grade english

To be fair, out of the 39 keywords, 3 conversions did come out of them. At a CPA $2 higher than my target CPA. I will adjust for that, so please go ahead and deposit $19.97 in my account. I'm just going to spend it on PPC anyway.

Now, if you're going to insist on forcing us all to participate in your program, a good compromise that I would be willing to accept is: instead of a refund, a lower cost per click on these session based keyword offerings. Since I know you can adjust the cost per click on a per search basis, why don't you go ahead and give me a discount on the cost per click so that it's not on completely on my dime that you're screwing it up?

I won't hold my breath on this, but I'll sure squeal to anyone that will listen.

Sincerely,

A Google AdWords Advertiser

Update:After running another search query report for one of our clients in the "electronic signature" industry, my PPC Guru colleague Michael found some even more "relevant" queries. So far this September, session-based broad matching has generated over $400 in largely irrelevant clicks at an average cost per click of over $1.80 more than typical search network CPCs during that same stretch.
Some keywords included: "eleanor roosevelt signature", "webkinz stores online" and "electronic barcode". Why are we paying full price for Google to make us their guniea pigs?

Did Google Earn the Fail Badge or did the Advertiser?

2009-09-02T22:49:19Z

Periodically you'll see a post where someone has collected "epic fail" (and not just on the Fail Blog) screenshots of ads gone wrong. The serious plane crash news article next to the banner ad for discount tickets on the same airline that crashed. The diet pill flash banner across a forum for recovering anorexics. A recall notice with a coupon code site for $5 off your next purchase of the same item. You know the ones I'm talking about.

In fact this post on The Business Insider about Google's Worst Ads Ever, got me to thinking about how these ads come about and who is exactly at fault. In the case of an airline displaying a discounted tickets banner ad on the same site as a story about their plane crash- that's their fault. Negative keyword lists are there for a reason. Their keyword list should have been loaded with words like "crash" "faulty" "failure" "strike" and "breaks guitars." Then if their ads appear next to stories with those negative keywords, then it's Google's fault.

The most common seen "it's Google's fault" is typically when dynamic keyword insertion goes awry. So in the example below, the ad is clearly almost entirely dynamic and is inserting the word "terrorist."

If you run an HVAC school you'd have negative keywords, but "terrorist" probably isn't one of them. And why would it be? It's not on your keyword list either, right? So how did that word get in there? (Provided that they're not completely retarded and simply selecting "add all" when the keyword tool returns 350 results.) Clearly there's an issue here of correctly targeting relevant content and serving the ad with the right dynamic insertion. Hopefully the advertiser is checking their search query and placement performance reports to help themselves and filter out these kinds of things.

Sometimes Google just straight up earns that fail badge. In this example, the ad is clearly not using dynamic insertion, it's way too specific in it's ad text and reads well. Yet it's being displayed across a video that the advertiser would obviously not want to be associated with their ad.

Their negative keyword list might include words like "abuse," "brutality," "rodney king" and "wrongful arrest," but as an advertiser, how could they forsee that "Deputy Shown Kicking Teen Girl" would come up? I went and found the original video on YouTube to see what it was tagged with: "wa jail video raw deputy shown kicking teen girl" and with the description: Surveillance video released in an assault case against a King County, Wash. sheriff's deputy shows him kicking a young girl, slamming her to the jail cell floor and striking her repeatedly. The deputy has pleaded not guilty in case. (Feb. 27)

There is no longer a Google ad displaying with the video....can't say they didn't learn. That or the advertiser blocked it.

Some are just funny....Lesbian online dating and YAML coding language were meant to go together!

But when you look at these, keep in mind, that despite Google's targeting flaws, they are still way ahead of their competitors.

CNN is no longer showing banner ads across their crime section. I looked.

While these gems of advertising gone awry are funny, they often equal wasted advertising dollars. So avoid the "epic fail" and run those search query and performance placement reports. Google yourself. Go to the sites you're advertising on and look around. It's not a perfect system, but it is continually improving, we just have to help ourselves a little too!

Write Short Simple Obvious Story Titles

2009-09-01T16:16:27Z

Keep your titles short, simple and obvious. Look at the image below. It's a partial screenshot of my RSS reader. This is your space to hook me. If you don't, I'm not opening your story. I have 241 unread items. I don't have time to read them all so I'll pick and choose based on my interest.

If you think I am an outlier, you're right. Most people don't use feed readers. Outside of search engines, people learn about stories on sites like Twitter. Twitter has a 140 character limit, including the link.

Consider who the people posting stories onto social media sites are.

They are your most engaged readers.
They are the people most likely to use feed readers.

Don't write Big Red Orb Rides Bull to China.

Write Apple Stock Up 30% after China Sales Agreement.

Remember the mantra: Short, Simple, Obvious.

230 Link Bait Ideas

2009-08-14T18:34:31Z

Generating great link bait ideas is hard work. When link bait works it's magical. Watching the page views, unique visitors and server load all go up at the same time is like watching the wheels spin on a slot machine right before it lands on 777. The difference is that a slot machine depends wholly on chance. With link bait, you have to earn your luck.

To help me create compelling ideas I created this poster - 230 Link Bait Ideas. I stuck it on my wall, printed in tabloid format, 11 inches by 17 inches, but it looks good in your PDF viewer too.

Download 230-link-bait-ideas.pdf

If you enjoyed this post please consider Tweeting it.

I've spent a lot of time thinking about and reading others' advice on building great link bait. Most of the ideas on this chart can be easily traced to others, so it's only fair that I mentions a few specific link bait thought leaders here: Todd Malicoat, Darren Rowse, Brian Clark & Rand Fishkin.

Download 230-link-bait-ideas.pdf

Feel free to print, share, email, or post this file. If you share it on the web, a link to this article is appreciated.

By the way: If you need help with link bait, you can always hire us.

Interactive Marketing - Ever Changing & Often Unclear

2009-08-07T19:46:44Z

Keeping up to date in interactive marketing is painstaking and frustrating. Yes, we are knowledgeable experts, but if you think we will have the answers at our fingertips every single time...you need to rethink your drink.

Let's take SEO as an example.

Things Change

This year Google announced that the nofollow attribute no longer reflects or redirects PageRank like a closed irrigation gate. Instead it evaporates link juice. By the way, it took Google almost a year to let everyone in on this change.

Here is another example of Google's changing tune. In 2005 Matt Cutts wrote,

If you're straight-out using CSS to hide text, don't be surprised if that is called spam. I'm not saying that mouseovers or DHTML text or have-a-logo-but-also-have-text is spam; I answered that last one at a conference when I said "imagine how it would look to a visitor, a competitor, or someone checking out a spam report. If you show your company's name and it's Expo Markers instead of an Expo Markers logo, you should be fine. If the text you decide to show is 'Expo Markers cheap online discount buy online Expo Markers sale ...' then I would be more cautious, because that can look bad.

It was clear to everyone that Google blessed CSS image replacement for company logos as long as the text faithfully represented the image content. When you read the full discussion this becomes even more apparent.

Then, on June 18, 2009 Google steps in front of video camera and changes everything.

In the video, Richard M. from Australia asks,

If you have a company logo on your site, what is the best way to include the text of the logo for SEO purposes? ALT tag, CSS hiding, or does it matter?

Matt Cutts answers emphatically,

Yes it does matter. It's much better to use an ALT tag than to use like, 'I'm hiding some CSS, you know, 9,000 pikels over to the left of the web page,' or something like that.

I would not hide it using CSS or anything like that when there's a perfectly valid, perfectly simple way to do it that does the job just fine.

Things Evolve

Google says they now read JavaScript links and index Adobe Flash content. It's true. They do. It is also true that any developer can create JavaScript and Flash content that Google cannot crawl. They don't have to resort to fancy tricks to roadblock Google either.

If you asked me right now what types of JavaScript links Google will crawl and which ones they will not, I'm not going to know. First, I'm not a developer or a connoisseur of Javascript links. Second, this is a fast evolving area where Google is constantly improving their abilities.

There is a First Time for Everything

I learned HTML in 1995. Since then, I've marketed and optimized countless websites, big and small. Even now, 14 years later, I'll come across situations I've never dealt with. It happened just last month (sorry, I cannot be more specific).

When I come across something new I research it and interview other experts. When the unexpected rears its head, I become the expert.

Sometimes I Don't Know

Google measures over 200 ranking factors or signals. I know this because they said so. I don't know what all 200 signals are. Google closely guards that information. Even though, I have lots of practical knowledge from experience and research. For example, I can tell you that title element content is highly influential (Now there's a revelation!).

Chances are good that I've not worked with your niche before. Otherwise, I'd already represent your competitor. ;)

I'm not going to know who the major players are, what the best keywords are, or where to find good topical links. Well, not today. But tomorrow? Watch out!

Why? Because I know where to look and how to do the research. Here at Portent we do it better than most.

So yes, interactive marketing tools like SEO, social media, PPC, email marketing, banner ads and affiliate management change. Interactive marketing changes constantly.

Deal with it.

We do.

Write Compelling Messages - Think Greeting Cards

2009-08-04T23:41:54Z

As Mad Men's Don Draper so eloquently puts it,

"You are the product. You feeling something. That's what sells. Not them. Not sex. They can't do what we do and they hate us for it."

To sell, you must evoke emotion. Make your customers feel. Stoke their desire to buy.

But, how can you get a customer to feel anything when you cannot see them? You have no idea if you're trying to reach a 75-year-old woman in Alabama or a salsa-dancing single in New York City. You're waving your hands like a five-year-old child trying to capture everyone's attention. First, stop that. Never yell at your audience or write desperate copy. Pretend exclamation points and all-caps are extinct. Second, stop thinking you're going to reach the entire world wide web with your message. Not only is it impossible, this notion prevents you from writing a targeted message.

Narrow your audience

Narrow your audience down to a small select group (about three or four), create marketing personas, and then get inspired by greeting cards. What message will your persona listen to? What words will make them feel? Greeting cards sell feelings. Walk down that aisle of colorful cards. Spend sometime in the Happy Birthday section. You'll laugh. Move onto the wedding cards and you'll weep, giggle, or get hungry for cake. Every message was created to stir your emotions. Greeting cards are copywriting and the product is the emotion itself. You wouldn't send a chimpanzee eating a birthday cake card to your Grandma (unless your Grandma is my Grandma). Just like you shouldn't write humorous copy when you're trying to reach a 55-year-old man who's about to retire. You don't know if Bob, the future retiree shares your love of fart jokes. You do know he's a nostalgic empty nester who's soon to set out a new adventure. That's why all those Father's Day cards have sailboats on them.

Can you tell the difference between the advertisement copy and greeting card?

"Don't you just love the 12-seconds when all the laundry is done?"

"For those nights you want everything to be just right..."

The first line is from a greeting card and the second is from an advertisement for stereos in the 70's. Both have a similar target: the middle-aged-bored-out-of-her-mind mom. Both make you long for a break, a little romance, or relaxation. Send the wrong message and your customers will feel nothing and move on. A good example of what happens when you send the wrong message to the wrong person, comes from my Uncle Jim, who once gave his wife a Far Side birthday card, complete with an egregious joke about cows and utters. She expected a flowery card full of tear-inducing sentiment. He gave her the humorous card he would want. Ultimate *FAIL*.

Speaking at SES San Jose

2009-08-02T21:22:01Z

I'll be running the Curmudgeon's Web Copywriting Site Clinic on Tuesday at SES San Jose. 10-11 AM.

Bring your web site and I'll put it up on the big screen, walk through tweaks and changes to copy, and hopefully teach a little at the same time.

Here's all the info:

Tuesday, August 11th, 10:00am - 11:00am
The Curmudgeon's Web Copywriting Clinic

Is This How Vince's Change Works?

2009-07-31T18:41:38Z

Early this year Google made a change to their organic rankings algorithm. Named Vince's Change, after the engineer who worked on it, the search marketing industry describes it as favoring brands. Here is what Google's Matt Cutts had to say about Vince's Change in March.

A few things that Matt says about Vince's change caught my ears.

It affects a small number of queries

People made 9.1 billion searches on Google during June of 2009. That's 9,100,000,000 searches. Let's say Vince's Change impacts 5% of searches. Five percent is 455,000,000 or 455 million searches. That does not seem like such a small number to me.

It does not affect lots of long-tail queries

The inverse of this statement is that Vince's Change is more likely to affect high-tail searches. Is Vince's change more likely to skew the results of competitive queries that companies like Portent Interactive's clients compete for?

Google does not think in terms of brands

Matt says Google cares about, "trust, authority, reputation, PageRank, high quality..." "Sometimes that's a brand search, sometimes that's an informational search, sometimes it's navigational, sometimes it's transactional..."

This sounds good, however, the search marketing community's concern about Vince's change is not the type of search--whether it's a brand search or not. Our concern is with the results--are brands more likely to appear in the organic search results because they are brands?

Trust, Reputation, Authority, PageRank

Matt Cutts says, "Try to make a great site. Try to make it a site that is so fantastic that you sort of become known as an authority in your niche. And it doesn't have to be a big niche. It doesn't have to be a huge well known keyword. It can be a smaller niche. And if you're still the expert that's the sort of of thing that they'll link to, that they'll talk about, the sort of thinks that people really enjoy and those are the type of sites, the experts that we want to bring back."

At Portent we like to advise our clients to become a resource, not just for their specific product or service, but for the market or niche that uses it. This expands clients' ability to create great content and to become a recognized expert and not solely bent on self-promotion. When you do this other websites are more likely to mention you and link to you.

The Power of the Mention

I think that Vince's change places added emphasis on mentions. Notice in the quote above Matt says, "...that's the sort of thing...that they'll talk about." Google likes to do things algorithmically. When creating a list of brands, they do not want to use someone else's list. They want to create their own list from scratch, using math and statistics. Nouns that are mentioned frequently throughout the World Wide Web are more likely to be important, like brands. This is an important change. Previously search engine optimizers placed almost all of their off-site focus on getting links from other websites, not just mentions.

Matt also emphasizes reputation. This indicates that sentiment is important too. If lots of people write good things about you on the web, then your business will have a good reputation. Inversely, if lots of people write negative things, your business will have a poor reputation. If nobody or few people mention you, that's not a bad reputation. It's no reputation or a lack of reputation.

Another word for reputation in this context is sentiment. Terrific sentiment, good sentiment, non-existent sentiment, negative sentiment and wretched sentiment. Measuring sentiment is fairly straight-forward. Some words and phrases are typically used to say good things while others are used to say bad things.

So now we have the frequency of mentions and sentiment.

Is This How Vince's Change Works?

Is this graph a good representation of Vince's Change?

It's a frequency of scatter graph on two variables, placement value and sentiment.

Placement value means the credibility and importance of the web page on which the brand mention takes place. All websites are not created equal. Some have more trust and authority than others. Also, within a website, some pages have more authority than others. A website's homepage, for example, is typically more important to search engines than a page that is five clicks away from the home page.

Each brand has a footprint. In the image below this website receives a lot of positive sentiment mentions.

The brand also receives many mentions on high-value pages or high authority pages. This brand is quite likely to benefit from Vince's Change and get higher search engine rankings.

Next is a brand that Vince's Change might hurt.

Once again the brand gets a lot of mentions on the Internet, except this time most of those mentions contain negative sentiment. This means lots of people are saying bad things about this brand, which can contribute to a poor reputation.

Before Vince's Change affects your brand it needs to exceed a threshold.

The brand I graphed above has neutral sentiment and most mentions are on pages with average authority. This brand is unlikely to receive any direct affect from Vince's Change. However, this brand should work to get positive mentions on high value or authority pages because a different brand, one which receives a boost from Vince's Change, could push this neutral brand down in the rankings.

It also makes sense that a few mention on highly trusted pages will have more impact than the same number of mentions on average or low trust pages. However, a large number of mentions on average trust pages will likely trigger Vince's Change.

I do wonder if there is a cutoff for low authority or low trust pages. It makes sense that pages with negative trust or very low authority cannot trigger Vince's change, no matter how many of these mentions occur. Of course, if many many mentions occurred naturally on low authority pages, one would expect to see brand mentions on average trust pages and even some high trust pages too.

Let's put it all together. Here on one graph is how I think Vince's Change may operate.

Here's a caveat. By collecting the scatter graphs for a large sampling of brands then charting their patterns or footprints, you will likely generate a normal distribution or bell-curve Participating in a scheme to manipulate Vince's Change by creating an unnatural pattern--one which falls outside of the bell-curve--could identify a website as being a member of a bad neighborhood and cause Google to penalize that website. Vince's Change emphasizes the importance of participating in aggressive brand building while avoiding brand manipulation.

Simplify PPC Reports

2009-07-27T19:00:00Z

After attending a Stratigent workshop with Jennifer Vessenmeyer called "Pimp your Reports" it got me thinking about PPC reporting and how I could be "pimping" my own reports. I don't mean throwing so much data together that no one will ever read it like a house bill nor exporting something out of AdWords and handing it over, I mean putting together the suggested meaningful and insightful report.

Currently the PPC report I put together each month now for clients is actually right in line for what the workshop covered- but definitely in need of a tune up and there were a few pearls of wisdom that we should all stop and think about before will fill out the next report on auto pilot once again.

Is your report meaningful?

That's a loaded question. Meaningful in the way like a wedding ceremony? Your mother's day gift?

No, meaningful in the way that it conveys information that is unique and needed to the client. That it almost tells a story about what is going on in the campaigns and account structure and that it is answering whether or not PPC is meeting the intended goals meant specifically to PPC.

Did you pick the right statistics?

Ah, KPIs (aka Key Performance Indicators). How do you pick those anyway?

Do you just include the ones the client said to, just because? When was the last time you reviewed what those were? Is the value still the same or has it changed? Check those out. For PPC it's going to vary greatly on that end goal. If it's to drive traffic, do you need a list of all top referrers or is it more that you should have a total and breakdown by engine of clicks/visits from paid search?

Does it matter what the average cost per click is across all your search engines? What does that metric do and what can you do with it? If the average cost per click is 0.40 and all 3 of the platforms you are using a really different as far as the reach, the type of person who uses it and pricing, what could you do with that metric anyway?

Which statistics are you showing together?

Meaning don't cram all the stats together on one graph, particularly if the data isn't related enough to be on the same graph. It skews perception and overshadows successes and areas of trouble. For example, if my leads based client is tracking their cost per lead month to month and their conversion rate, should those two metrics be on the same graph? No, a conversion rate is measured in percentage and the CPL is in dollars. You wouldn't believe it, but some folks really do mix and match their units of measurement just to save a little space.

Or how about putting your overall conversion rate against individual ad click throughs? Technically they're both percentages, but how do you connect (by looking at the graph only- remember you can't be standing there waiting to explain every time) a conversion rate with many sources with a single ad's click through?

Same goes for too many metrics on one graph: CPC, CPL, revenue, cost, CPC per engine, CPL per engine, cost per engine, revenue per engine and previous month's CPL should not be all on one graph.

What kind of graphics are you using?

Line graph? Bar graph? Scatter chart?

So many choices, so little time. Clients like data visualizations. It makes things pretty. But what are your graphics really telling you? And can clients actually read and interpret the data in them? Go with a chart that is easy to interpret and doesn't require a two day workshop on how to read it from you. For PPC, I prefer the line graph. Clean, simple and a clear indication in the up direction. (Revenue of course!)

What colors should I be using?

There's a rainbow of palettes and I want to use them all!

Please don't. Colors are a great way of drawing attention to points you want to draw attention to (but also a good way to hide some stuff!) Use colors that come from the same palette. Don't use red, yellow, aqua, purple and then magenta cause you feel like it. Gradiate the colors. Have lower numbers be lighter and bigger numbers be darker for example. Remember that if they print the report, it'll probably be in black and white. Can you tell the difference between sunshine yellow and mint green when it's in black and white?

Using text and wording appropriately

There's a key with your graph right? Or each line/bar is labeled? And when I say labeled I don't mean with everything you can possibly think of. Name the report something specific like "2009 PPC Performance for X" or "January 2009 PPC Ad Performance -Google." Then you don't have to put "Jan-2009" "Feb-2009" "Mar-2009" you can just put "Jan" "Feb" and "Mar." Label what needs to be labeled and if you need to, include a small box with short a explanation on what certain metrics are, how they are measured/calculated or directions for interpreting the report. Some clients won't know what CTR stands for or what the Conversion Optimizer is. The real mistake is letting the client read it and then make up their own interpretations.

Example graph, the stuff that should be looked at is circled in red. Missing a title, cut back on the '09 labels, who picked these colors(?!) and why are there 2 sets of numbers- one on each side?

Small changes to make big improvements

Face it, clients aren't big on change, particularly if you're taking something away. Unless they're changing SEM agencies in hopes of shaking everything upside down, back and around, you'll need to implement incrementally and subtly. If you're cutting stuff out, cut one to two things out at a time WITH an explanation. Do not simply cut something and think they won't notice. They will notice and they will ask. A lot.

The Pimp Your Reports workshop was super helpful and made a lot of sense. Many thanks to Jennifer Vessenmeyer for including Portent and providing materials that will not be leaving my desk anytime soon. You would be absolutely astounded to see some of the reports that are floating out and around today from SEM agencies that are considered to be standard reporting.

Google Suggest Links - A New Upgrade

2009-07-24T18:20:11Z

Like every morning, I began today with a sane dose of looking around my iGoogle dashboard, checking email, reading important news, and consuming my Calvin and Hobbes comic of the day. What grabbed my unexpecting eyes was the iGoogle search box. This is an important feature. It gives me quick access to all things Google. Today, when I used the iGoogle search box, something different happened, something unexpected. Today would be an exciting day.

Earlier I overheard my co-workers talking about a juicy post in FAILBlog. When you hear something like this and are sitting in front of a plugged-in powered-up computer connected to the internet, well, naturally you go searching for it.

As I typed my query into the iGoogle search box Google Suggest gave me the failblog.org/ URL as a suggestion.

Google Suggest also gave me a blue underlined hyperlink. This was quite unexpected.

Curious, I tried other variations. Let's try... s-u-c-c-e-s-s?

Zero. Zip. Zilch. Nada. Next attempt, n-y-t.

Link ahoy!

I fire out another reliable source of news. C-N...

Talk about confidence. I didn't even finish typing that query.

What other source of reliable information and news can I think of? W-I-K-I-P-E-D-I-A

No link. That's strange.

How is it that FAILBlog gets a link while Wikipedia doesn't?

This begs the question, what's wrong with Wikipedia's relationship with Google? Wikipedia is a juggernaut. Did it somehow make Google unhappy? Or could it be that Google just doesn't want to share all of its toys?

Google Suggest Links. What will they think of next?

Where is Your Social Media Marketing Headquarters?

2009-07-23T22:37:57Z

This post is about the nature of social media. The premise is simple. Your social media marketing needs a centralized location, a headquarters.

Notice I did not write that you should centralize your efforts. That would be too much like those classic cartoon scenes in which the characters disappear behind skinny trees. When you hang-out in one place and one place only, online no one will hear you.

Rather, I'm writing about one easy to find place on the web where anyone can catch-up on all your important information and find your company's other online hangouts.

Let's explore this together.

In real life--not that life on the Internet isn't real--locations don't define relationships. Sure, you may have met you best friend in school or your wife in church. You see some friends in some places more than others and it's pretty hard to go bowling in a boat. But, you can get together pretty much wherever you want.

I bet that you and your friends have a regular hang-out. It might be the local Elks club or an Irish pub or even your garage. And while you see each other there most often, you go to dinners and movies and concerts with your friends too.

I'm also guessing that all your friends don't hang out in the exact same places. They may all have your garage in common, but do all of your friends go to your favorite restaurant or your favorite theater with you? At the same time?

Finally, it's likely you have different circles of friends. The circles overlap, but not all your friends are a part of every group you belong to.

The Internet is a lot like this. Some people Twitter. Others like Facebook or MySpace or LiveJournal. I know lots of folks who hang out in forums or just read blogs. While most people do more than one of these activities, few do them all. This is why it's important to maintain a presence in lots of different places where your audience, niche or target market congregates.

To be seen, you cannot expect everyone to come to you; you must go to them.

Unfortunately, maintaining accounts on several different networks can easily become disjointed. You will have different conversations in different places. It's only natural. Keeping it natural is an important principal of social media marketing. Trying to include every little thing everywhere is not natural; it will put people off. For example, if your company announces quarterly earnings it will undoubtedly publish a press release. However, you will not repeat that same press release in every social media space. Instead you might find different highlights applicable to different social media accounts. You will post about those highlights and include a link to your blog or press release.

Since I mentioned your blog, let's bring this full circle. For most companies I believe that your blog is the perfect place to create you social media headquarters. Foremost, it's a space that you have complete control over. You get to decide the look, the feel and the content. You can link to all of your social media accounts from the template. If something noteworthy comes-up on one of your social media spaces you can write about it on your blog and encourage readers to go take a look.

Unlike your media or press section, your blog can--and should--have a personality. You can be more casual and talk with people instead of making third-person announcements. On your blog you can show emotion. For example, if your earnings go up because your CFO found a few million dollars in cost savings, you can express your overwhelming joy and exuberantly tell people that you have the best CFO in the world. You can't do that in a press release.

Here's the check list:

Have a company blog where you talk about the cool things you do and how cool your company is.
Keep your blog conversational and friendly. Give it a personality. Have posts by people, not 'The Company.'
Link your blog to all of your social media accounts.
Link your social media profiles to your blog.
Mention interesting conversations happening in your social media spaces/accounts on your blog and link to them.
When you write something really interesting on your blog, write about it in your social media accounts and link those posts to your blog.

And here's the bonus suggestion. Use the press release section of your website to make announcements. Use your blog to write about what those announcements mean.

Fisher Plaza: Service Restored, Portent Response

2009-07-04T16:27:08Z

If you host a web site with Portent, you've been kept apprised of the situation that started Thursday night at 11 PM.

At this point, all sites are back 'live'. The Fisher Plaza East data center is running on temporary power from Caterpillar generators that the City of Seattle set up outside.

Here's a quick rundown of what happened, why, and how we can avoid it in the future:

Chain of Events

Thursday

11 PM: A fire breaks out in the electrical hub at Plaza Center East. The sprinkler system activates, but floods and shorts out the emergency generators. Plaza Center East goes dark some time before 12 AM Friday.

Friday

12 AM: We receive an alarm that our sites are 'down'. We attempt to contact our servers, can't connect, and then attempt to contact Adhost. Adhost's phones and servers are all down, as well, because of the fire. We try to contact them repeatedly over the next several hours.

5 AM: We get news of the fire at Fisher Plaza. Until 5 AM Pacific, no news outlet or other resource reported about it. We found the information via Twitter and by checking Seattle Fire Department incident logs. So much for Seattle's 'old media'. To be fair, though, two of Seattle's primary news stations are also knocked out by the fire, so they can't report any news.

5:30 AM: We contact all Portent hosting clients regarding the outage. At this point, it's very clear this problem will last the day. I review alternatives and decide we have to sit tight and wait for power to be restored. We can't get into the building yet; any repointed web sites will take 24 hours to 'repoint' across the entire internet; and then repointing them back to Adhost would take another 24 hours.

5:45 AM: We shut down all pay per click advertising for sites that are offline.

We spend the rest of Friday monitoring the situation. Where possible, we use client's Facebook and Twitter accounts to notify their customers.

Saturday

1 AM: Building power is reconnected. Power supplies are recharged, and HVAC is restarted to cool the building before servers restart. The building got up to over 100 degrees inside during the day (it's been over 85 degrees for the last several days).

4 AM: Servers come back online. Power restored to the building.

What Could Have Prevented This?

In the end, Portent had only one way to avoid this outage: Host all web sites in two physical locations. Then, when Fisher Plaza East lost power, we could have started up the sites in the other location.

The problem with this solution: It's expensive. Extremely expensive. It requires that we duplicate the hosting environment - hardware and software - and that we constantly keep both hosting locations in sync. For a once-per-decade power outage, the cost/benefits balance just doesn't work.

Future Plans

However, cost/benefits analysis doesn't feel as good right after your sites are down for 26 hours. So Portent will be planning and offering two additional services:

For e-commerce and other database-driven sites: Redundant hosting in multiple physical locations. If there's a major power outage, we can flip a switch to point at the backup location, and you only experience a few minutes of downtime.

For lower-priority sites: Hosting of 'static' files, such as images, video and pages that don't change, using Amazon S3. Backup of databases and other dynamic files using Amazon S3, as well. That way, moving to a new location in an emergency might take time but is still possible, even if all connectivity to Adhost is lost.

These services will cost extra: Portent will have to pay for data transfer, hardware and power usage for the additional storage and the 'mirroring' necessary to keep separate locations synchronized. Once we've worked out pricing, we will help you with the cost/benefits analysis.

I hope this information is helpful, and I appreciate everyone's patience during the outage.