Got an email this morning to let me know that my son has won an award @ school*. Hooray! But the award ceremony is tonight - nice of them to let me know so far in advance. Anyway…

Because I am not finished with my final read-through, and I insist on doing that before I publish… The release of the 2008 edition will be put off until tomorrow morning, instead of tonight. I'm sure we'll all survive an extra few hours without it.

The big burning question is: should you read the 2008 edition, if you've already read the 2007 edition? The answer is, it's up to you. But there's no rush.

The only changes that I consider significant are in the chapters on Site Structure (which I think is better written now) and link building (where there are some new opportunities worthy of discussion). Oh, and if you didn't grok the Keyword Strategy chapter, give it another try with the 2008 edition, I did a minor rewrite on that.

Your site will not burn to the ground if you continue to rely on the advice I gave in the 2007 edition. In spite of what you may have heard, SEO just doesn't change that fast.

*If this turns out to be a 'punctuality' award or something like that, I will feel seriously cheated. I want him to be "bad-ass of the year" or something, at least…

I'm sure the title of this post is going to make a few folks upset. Don't worry, the content is going to really make you mad. Not at me, but maybe at yourself, and especially at whoever taught you how to split test.

That's because, for most Adwords advertisers, the title of this post is very accurate. Most advertisers are doing split testing all wrong.

If you're doing  A/B, you're robbing yourself of profits at every step.

That's because, in a typical A/B split test, what you're doing is keeping your Control (the best performing ad) running, and creating a new test ad to run against it. Doing it that way is perfectly normal, but it's also completely, utterly, and totally wrong.

Most of the time, these test ads fail to beat the control… and predictably so. Your Control is the Control because it's doing well. After a few rounds of testing, it becomes less and less likely that your test ad will beat the control.

The "usual method" of ad testing delivers invalid results - here's why:

• The Control has a strong performance history, but your test ad has no performance history.
• As a result, your test ad may be ranked lower on the page, leading to lower CTR and (quite possibly)  conversion.

In other words, your test ad might have been better - but it didn't get a fair chance! This means that you may have already rejected many great ads, because you didn't test them correctly.

But that's not the worst part! "Control vs. test ad" also loses money:

When you run two ads in equal rotation, your Control is only getting half of the available ad impressions.

When a test ad fails fails to deliver good click-through and conversion rates, you've just given up as much as 50% of the profits that you would have had, if you had just left your ad group alone.

Wouldn't it be better if you could construct a valid test that didn't bleed money? Here's how it's done…

Step 1: Set Up A Valid Test

Instead of running your test ad against the Control, you create multiple copies of your Control.

How many copies you make is determined by how many of the ad impressions you want to go to the test ad.

In the example above, we've created 3 copies of the Control, left the Control running, and set up one test ad. Therefore, the test ad only gets 20% of the ad impressions.

This method leaves far less of your profit at risk.

It also allows you to run a valid test, by comparing the performance of your test ad against the copies of the Control only.

Since none of the ads in the "testing pool" has performance history, you can have far more confidence in the test results.

This method has reversed the results of many "close misses" I had with the old method of testing, which I was using myself until 2006.

Step Two: End Test (Fail) or Ramp Up To Full Test

In one of my many "hallway conversations" at the last Stompernet Live event, Andy Edmonds (Chief Scientist @ Stompernet and a real expert in testing) pointed out the real advantage of this method… which is that failed tests usually "fail early."

Scientific testing types have a much more formal language for this, of course, but I got tired head when he explained that part of it.

I'll just summarize the main points for you:

• You can usually kill your test ad pretty early because most test ads will be obvious failures.
• Running 3 copies of the Control actually accelerates your testing most of the time.
• If the test ad appears to perform well, you can progressively eliminate copies of the Control to ramp up the pace of the test.
• The final test is the validation test, where you run your Test Ad vs. Control - but now, you're doing it right!

Since I've started doing this "Andy's way," my ad tests are running even faster than before. Way cool.

But what's the real goal of all this testing? Higher click-through rates? Lower cost per click? These statistics are merely means to a greater end, my friend, and they can often lead you astray.

Indulge me for a minute, please. I didn't get to be the "Keyword Guru" for nothing, you know…

There's one other huge mistake that most folks make when they're split testing ads - they don't compare the business performance of their ads, just the click-through rate.

The problem should be obvious - you aren't trying to buy traffic, you're trying to buy customers. Actually, that's only sort of true. What you're really trying to buy is profit. Specifically, what you really want to achieve is the maximum possible profit per keyword.

At Stompernet, we call this statistic "Profit Per SERP"

Understanding how much profit you make every time one of your targeted keywords gets searched is a powerful idea. One of the most important metrics that any enlightened website owner will learn to measure and improve. It's the KEY to search marketing success.

You have a lot of control over every variable in the search marketing equation, except for one thing…

You can NOT change the number of times people carry out a given search.

During the month of May, some number of people are going to search for "mortgage rates," or "baby gifts," or "free seo book." We are powerless to change this number.

The number that we're trying to change is our average profit for each of those searches.

By doing Adwords right, you can change that number, and in doing so, change the rules of the game.

We'll talk more later.

For now, I'd like to invite you to watch the 50 minute instructional video that I prepared for you, with the help of master video editor Andy Jenkins. Watch online or download the Quicktime - but do watch… you don't want to miss this.

 It's been a while… here's what I've been reading:

Best of the week: The SEO Guide to Information Architecture, by Adam Audette
Goes a little farther than I would into the whole theme concept, which is a fine way to look at site structure but pretty meaningless for SEO.

Darned good, from April 16: Wake up, real estate bloggers, there is no money in the long tail of search for niche blogs so stop chasing it by Mary McKnight
Close to a rant, but just this side of the fence. Mary writes mainly for realtors, but this is good perspective for all of us.

Best of the rest…

Web 2.0 Expo - Creating a Social Strategy
Lee Oden
referencing Josh Bernoff and Charlene Li of Forrester (authors of Groundswell)

Economies of scale in the very personal groundswell
Josh Bernoff

Social technology marketers bullish in face of recession
Josh Bernoff

Targeting Conversions and Traffic with Long Tail Writing
Jennifer Laycock

7 Habits Of Incredibly Effective Bloggers
Sarah McHarry

Some Thoughts on SEO Ethics:  Matt McGee, Michael Gray, Jeff Quipp, Jim Hedger and more
MelissaF

The Web Developer's SEO Cheat Sheet
Danny Dover

Google Leaks Quality Score Variables (Pscore, mCPC and thresh) in Search Results
Eric Lander

Blogger’s to do checklist before hitting the publish button
Jennifer Slegg

Twitter Wrote This Column For Me
Chris Winfield

How to Analyze Your Site with Del.icio.us
Ann Smarty

Holy Mother of Linkbait
Jane Copland

How to get your Blog Traffic to Convert in 5 Easy Steps
Jennifer Osborne

Why Microsoft + Yahoo! makes sense – and why it won’t work (Deal's off, BTW…)
Charlene Li

The denizens of Internet Marketing Planet are hammering inboxes this week with another promotion, this time for Jason Potash of "Article Announcer" fame. I have to admit that I haven't been paying much attention to what he's selling, if it's even been disclosed, but I do know that he's been giving away some software called "Comment Kahuna."

Comment Kahuna allows you to search for blogs and posts on a topic. So far, pretty harmless. The feature that's being touted so widely is the ability to search for blogs that aren't using nofollow on their comments. I'm sure we can all see what might be disruptive about that.

Andy Jenkins posted his thoughts on the Stompernet Blog, about the right way for commenters to participate in blog communities, and I couldn't agree more. Comment Kahuna can be used for good, or evil… OK, maybe not evil, just unwise.

A cool thing happened after he posted - the first comment comes from Andy Beard, who does exactly what Andy Jenkins said commenters should do - he added value to the discussion. Kudos to both Andys for bringing some signal to the web, and not just with this one post.

Mr. Beard and I disagree as often as we agree, but I will argue to the end that he brings something important to the conversation.

Stompernet faculty member & "Chief Scientest" (sic) Andy Edmonds pointed out a very useful resource the other day:

Seven Pitfalls to Avoid when Running Controlled Experiments on the Web is a great white paper by Thomas Crook, Brian Frasca, Ronny Kohavi, Roger Longbotham from Microsoft. Check out the site for the MSFT Experimentation Platform while you're at it. Cool stuff.

For those of you who don't know Andy, he used to work at Microsoft, on the Live Search team. He does research, actual work, tool development, and training for us at Stompernet, on analytics, usability, testing, and conversion. He is smarter than everyone I have ever met, combined. Andy's "Always Be Testing" blog is required reading for anyone working in any of the above areas. Seriously.

Andy's "Scrutinizer" tool is also one of the coolest free things any web designer, conversion/usability professional, or web entrepreneur could ever wish for. It lets you see your website the way your visitors do, by simulating human vision. This allows you to spot subtle weaknesses in design, usability, and conversion that would otherwise require expensive eye-tracking studies to understand.

Sorry folks, this one has nothing to do with search marketing… unless, like EVERY business, you use the telephone to interact with customers. Oh right, that's almost all of us.

Why? Why? Why?

I spent last Thursday-Sunday at Stompernet's "Live 5" Conference, so called because it was our 5th live event… really an amazing event, which I'll cover in more detail at another time. If you must know what happened there, new member Lynn Terry has some interesting perspectives.

What I want to talk about today comes out of a conversation that I had after the conference, while I was sitting in the hotel's coffee shop on Monday morning. I was chatting with Mark Benda of Faster Audio, creator of the soon-to-be-famous Benda Index. We got to talking about support costs for software developers, and that wound around to a question I have been asking for years…

Why Aren't Call Centers Profit Centers?

Some of the activity of your call center (wherever you answer your phones) is definitely costly. Customers calling to check on order status, that costs money. Customers calling about returns and refunds, that costs money. Customers calling because they can't figure out how to use your product, that costs money. In a moment, I'll suggest some ways to turn a profit, even with these calls.

If you're doing business on the web, of course, you just might be able to reduce those costs by doing a better job with your website.

Personally, I buy a lot of stuff online, and fewer than half of the merchants I deal with provide any kind of order status information online or by email. Of those who do, fewer than half give me a way to track shipments - most of the time, you don't even know that your order has been shipped.

Even those who provide tracking information, at best, only provide a tracking number, and *maybe* a link to the FedEx or UPS web site. Why? Is that all your "free" shopping cart software will do? Do you realize how much "free" costs yet?

Shippers can get detailed tracking information automatically, and send that directly to the customer. In fact, you could even send an email with a subject like "Your Order Should Arrive April 9, 2008 via UPS." Wow - wouldn't that be nice… and I wouldn't be calling you about it.

A call for returns or refunds is probably worth taking, but you could at least give your customers a way to initiate the process online, and call them back with instructions. I've seen good phone reps with a good product save many sales by talking to the customer, and in fact, I've even seen them generate new sales.

Most of the time, though, a conversation with a customer is an opportunity. A chance to learn what they want. A chance to collect testimonials. A chance to find out what your website doesn't do well enough. A chance to sell them something. A chance to regain their trust. A chance to create a raving fan.

If you can't turn a profit by answering your phones, you probably aren't thinking hard enough, and you may need a better business model.

Die, Bad Call Center Practices!

This isn't a comprehensive list, but it's a good sampling…

1. If you call your customers with a machine that puts them on hold, you should just go die. I know that it saves precious seconds of labor when your phone reps have zero idle time between calls, but it makes your customer and your phone rep less friendly, and leads to a much lower quality of interaction. You saved 20 seconds dialing time, and the customer is going to spend part of that time complaining about being called and put on hold. Let your reps breathe and compose themselves between calls, or one bad call will lead to another. You're already interrupting your customer, so try not to be a jerk about it.
    
2. If your call center that's trying to do outbound selling does #1, you should die hard with a vengeance, you idiot. I've had this happen so many times, and I seriously don't get it. Your robot calls me up, puts me on hold, and then your poor rep gets on and pitches your offer to me? Unbelievable. You're throwing away opportunity, and wasting customers' time, all with one bad practice.
    
3. If you ever tell your customers that they called the wrong toll free number, and tell them to hang up and call someone else, you don't deserve to suck any more of our precious air. Bonus death points if you made them sit on hold for several minutes, tell you their entire story, and then asked them to hang up and go get in another queue. A humane death by lethal injection if you at least told them to call elsewhere quickly.
    
4. If you transfer customers to another queue, after they've already sat in one, because your reps can't handle the slightest exception, go die, and then sit in Purgatory for 999 years and think about what you did wrong. Do I even need to explain how bad this is? Figure it out yourself, call center guy. Go ahead, we're waiting.
    
5. If you make a man named Vishy tell customers that his name is Mike, you don't have to die, but please stop. Seriously. We know his name isn't Mike, and it's insulting to all of us, especially Vishy. He works hard for you, much harder than you deserve. He's probably smarter than you. He lives in a country where this is actually a halfway decent job. What's your excuse?

Now, Some Ways To Get Better

In addition to being whipped to death by bad call centers, I've also been pleasantly surprise by some astonishingly good practices, and even used some myself…

1. If you can't answer the phone immediately, collect information from the customer while they wait. Some call centers get all the information they needed from me automatically, and when I get a rep on the line, they know exactly how to take care of me. In fact, you might be able to collect some fantastic data on what your customers want. "We'll be with you shortly, and if you don't mind answering a quick question we'll move you to the front of the line." Ask them what their favorite color is, ask them anything that might help your business make better decisions.
    
2. If the wait time is going to be anything but brief, offer to call the customer back. This seems like such a no brainer. Humans do it all the time - ever called someone's office where their assistant takes a message and asks when they can call back? "This is a very busy time for us, and our current wait time is approximately 19 minutes. We'd be happy to call you back if you prefer. If you'd like us to call you back within the next hour, press 1. If you need to schedule a later time for us to call back, press 2. To continue holding, press 3." How hard was that? Your robot can call me back and connect us, and nobody has to sit on hold.
    
3. Why not make a little cash on the phone? If repeat business matters to you, offer customers coupon codes when they call. If they're using an older model, ask them if they know about the new one. If you sell software, why not offer the tech support customer a great deal on training DVDs, add-ons, or upgrades? Points off if you try to do this before you take care of the customer's need, bonus points if you actually start turning a profit in your call center. It can be done.
    
4. Staple yourself to a case! There was a great article in HBR about 100 years ago called "Staple Yourself to an Order." The idea was to see the entire process… if you run a call center, put together a list of common customer stories & cases, set it up in your CRM system so it will look real, and make the calls. Over and over. Don't just listen to recordings, experience the whole process just like a customer. Bonus points if you attach a blood pressure monitor and note what made your blood boil.

Thanks for listening, folks.. I feel better already. If you offer good call center or phone solutions that can help my readers fix what's bad and do more of what's good, I'm happy to hear from you in the comments.

PS - Quick thanks to Linda and Craig @ Kinko's who helped made the phone my "favorite machine."

Just a quick heads up for my readers here…

Adwords new display URL rules went into effect yesterday. Pretty simple stuff - the domain name you show in your ad's display URL must match the domain name where the visitor actually lands after they click the ad. Pretty simple for most advertisers, but for affiliate marketers who want to avoid looking like affiliates, a little troublesome… and even for honest affiliate marketers trying to play by the rules, it creates a challenge.

Affiliate marketing via PPC advertising is tough, and one of the ways you try to get ahead in the game is by split testing domain names, and using keyword-specific domain names to raise the click through rate as much as possible.

Bryan Todd, over at Perry Marshall's shop, wrote an article on how to get around these rules in order to split test domain names in ads. I appreciate Bryan offering a solution, but I wish he'd been a little more careful with his research. While the process Bryan outlines apparently worked with *his* hosting setup, it won't work for everyone, and there are SEO implications that weren't considered.

Bryan's plan (pointing another domain at the I.P. address of the existing web site) works if the web server is configured to serve up the same site as the default. Many servers will instead deliver something else, like a default page, if you do what Bryan recommends.

If your server configuration happens to work with Bryan's method, you'll still have more than one domain name serving up the same web site. That's duplicate content. If you must split test domains (affiliate marketers are advised to do so), and you care about SEO, you'll want to copy the site and use robots.txt to make sure that only the Adwords quality bot reads the extra domains.

Don't you just hate "that kind of marketing?"

Most of my readers, and in fact most search marketers, have at best an uneasy arms-length relationship with "internet marketing."

You know, the kind of marketing that uses long sales letters with with big red headlines, bold claims, and gushing testimonials. The kind of marketing that bombards your inbox with the latest and greatest "new" thing.

Most of us naturally recoil at "that kind of marketing." We don't want to use it in our own businesses, because it just makes us feel kind of slimy. I get that. I'm with you… but the truth is, "that kind of marketing" does work.

With a little understanding of the principles behind it, you can make it work just as well without all the hype. When you understand how it works, you can use it to sell more of whatever you sell, not just information products.

In fact, if you're reading this, there's a good chance that I used "that kind of marketing" to get your attention in the first place.

You probably didn't think of it as an "internet marketing" type of promotion, but when I relaunched SEO Fast Start last May, the process was the same. Same principles, same methods, the only difference is that I used my "inside voice" instead of shouting… and it worked.

How well did it work? Well, we've had over 20,000 downloads since last May, and a huge portion of that audience actually read the book and took the time to register for my newsletter.

It's Not About The Hype & Claims, It's About Building Anticipation

I've been getting a lot of emails about Jeff Walker's "Product Launch Formula" for the past couple weeks. I'm sure some of my readers have as well, because it seems like every "internet marketing guru" is pushing it like mad.

Some of you may already be jumping out of your pants with excitement about it. Or, if you're like me, all that hype makes you skeptical.

In this case, though, the product merits the attention. Having purchased and used the first edition of Jeff's course a couple years ago, I can tell you that it has been tremendously helpful, and not just in selling information products about online marketing.

Jeff has distilled what's true out of the hype-laden world of internet marketing. You can use Jeff's formula to sell more of anything you sell.

You don't have to launch a "product" at all - it could be a special sale or promotion, an end-of-the-month inventory blowout, anything. The point is to get your customers and prospects interested, and make whatever you're doing into an event.

How Does Selling Out Amazon Sound?

The most recent email I got about the PLF was from Andy Jenkins at Stompernet. Andy's a good friend, and a business partner, but make no mistake about it, he's a lot closer to being one of "those types of marketers" than I'll ever be. He just happens to be one of the honest ones.

Jeff Walker helped us out (a lot) with Stompernet's most recent launch of SMARTS, a social marketing course & coaching program that I helped put together. He was deeply involved in Stompernet's initial launch in October 2006, which set all kinds of records in the IM world.

Jeff has talked about those launches in depth… but Andy's email didn't mention those "internet marketing" launches at all.

Andy tells the story, in great detail, of how he used Jeff's product launch methods last year. Not to promote Stompernet, but to save a movie that he produced from dying on the DVD racks.

Even though they had a major studio involved in distribution, it was the producers' own online promotion that really made the DVD release into an event. Such an event, in fact, that they caused Amazon to sell out their entire inventory in 36 minutes.

How Do You Hype A DVD?

This is where it gets interesting. "That kind of marketing" was used to sell a movie, folks.

They didn't use a big red headline. They didn't tell everyone that they'd made millions of dollars just by watching the movie. How could they? For one thing it wasn't true, and unlike some internet marketers, Andy doesn't just make stuff up. More importantly, that kind of hype wasn't what the audience wanted!*

What did the audience want? They wanted what movie geeks always want - insider information, behind the scenes stuff, trivia… so that's what Andy offered them. People watched the trailer, signed up for the mailing list, and got a PDF chock full of that stuff for their trouble.

When it came time to "launch" they had a huge list that was eagerly anticipating the release… and it sold like crazy.

Chances are pretty good that if you walk into a Blockbuster store today, you'll find Andy's film, Altered, occupying prime shelf space. Internet marketing, done the right way, has a lot to do with that.

What does your audience want?

I don't know your customers like you do, but I do know that they want… something. I know that Jeff's course can help you identify that something, and show you step by step, how to use that knowledge to sell more stuff.

I don't promote or recommend products that I don't believe in. I wouldn't suggest that you run out and buy something, unless I would buy it myself. I have no idea what the price of Jeff's course will be, but I do expect it to be worth every penny.

Hey Dan, Is That An Affiliate Link?

Yes, it is, and I think the product actually goes on sale this week. If you buy Jeff's course through my links, I will earn a commission, and *you* will earn a free pass to my next online class.

If you bounce over to Andy's blog and buy through their links, then Stompernet will earn a commission, and you'll get whatever they decide to offer as a bonus. Naturally, I like my bonus better.

BTW, like Jason Calacanis, I do believe that affiliate links should be disclosed in some fashion… and I think we're all better off doing so before the FTC forces us to do it. It's kind of a no brainer, and if anyone but Jason "Hate Bait" Calacanis had said it, I think the reaction would have been very different.

Go Watch Jeff's Videos For A Free Education

The  good news is that you don't need to buy anything to learn a lot. Jeff Walker has produced several case study videos that are themselves an excellent course on his methods. The first video is here, but I recommend that you start with this case study instead. You can watch a short sample and then opt in to his list to get access to all of the videos.

Since it's easy to unsubscribe, I recommend just opting in first, and then watching them all. The sample video is just the first part of a longer video, so if you watch the sample and then opt in, you get to watch the same part twice. Seems a little silly to me, but hey, I never said he was perfect!

Trust me, whether you buy Jeff's course or not, these videos are worth the time.

You may not like "that kind of marketing" as it's practiced by most internet marketers, and I don't either… but  it does have some real value when you understand the principles behind it.

* note: sometimes, red headlines and bold claims are exactly what the audience wants… knowing their audience is why a lot of these guys are successful. What makes so many of us uncomfortable about "that kind of marketing" is that we're often exposed to messages that were written for an audience that we aren't part of.

Google has announced that "landing page load time" will "soon" become a factor in "landing page quality scoring."

Okay, so what does that mean to you? Well, if you listen to the blogs & forums since the announcement, it means, of course, the end of civilization as we know it. You'll need to buy a dedicated server, stream every bit via Akamai, blah, blah… Hold on. Stop panicking. Use your inside voice. Think for a minute…

First of all, if you don't advertise on the content network (Adsense), this shouldn't be a big deal, unless it actually creates a minimum bid issue for you. Why do I say this? Well, because Google says so - "landing page quality" is not a factor in ranking search ads.

• Just ask Google.
• Then ask them again.
• And again. This last one, by the way, is a great read for those who don't understand the many meanings Google applies to the single phrase "quality score."

Is it possible that Google could change their policy, and let landing page load time influence search ad rankings? Sure they could, but to date, they haven't said so.

Otherwise sane people are insisting that they've already been slapped over landing page load times. I'm sure some folks have noticed higher minimum bids this week, but that happens every week. I'm not buying cause & effect with this announcement, because that idea makes no sense. Some of them are even prescribing fixes based on nothing more than complete speculation. Fixes that, if done wrong, could leave your site offline until a professional fixes your mistakes… Relax, people!

According to Google's announcement, this hasn't even rolled out yet. When it does happen, we'll get another announcement and we'll be able to see the load time score in our Keyword Analysis Page. We will then have an entire month to make adjustments, if it's even necessary. Until then, take a chill pill, and just wait to see if this means anything to you. My bet is that very few advertisers will be affected.

A friend just tipped me off to this offer they received from Google Adwords:

I'm excited to tell you that you have been selected to participate in a beta for our new Automatic Matching feature which will be starting on February 28th.

Automatic Matching automatically extends your campaign's reach by using surplus budget to serve your ads on relevant search queries that are not already triggered by your keyword lists.  By analyzing the structure and content of your website and AdWords campaigns, we deliver more impressions and clicks while maintaining your current CTRs and CPCs.

For example, If you sold Adidas shoes on your website, Automatic Matching would automatically crawl your landing page and target your campaigns to queries such as: "shoes" "adidas" "athletic", etc., and less obvious ones such as "slippers" that our system has determined will benefit you and likely lead to a conversion on your site.

Be assured that automatic matching will try to never exceed your budget. If you're already meeting your daily budgets, automatic matching will have a minimal effect on your account.

The broad match feature of Adwords is bad enough, folks. Now they're offering you the exciting opportunity to bleed every penny of your budget every day, advertising against keywords that you didn't want to bid on. Sure, if I sell Adidas shoes, why wouldn't I want to get some traffic from people who searched for slippers? I mean, it's not like I'm trying to turn a profit or anything, right?

This is pathetic. Don't get sucked in.