A couple of weeks ago, a publisher asked for help. His advertisers were upset because they weren’t getting the kind of clicks that he was reporting. I’d like to put a campaign code on the links, he said. Is there a universal campaign variable that I can use — one that will work with anyone’s web analytics?

Normally, I wouldn’t have gotten involved but my friend Ian Houston asked me if I would write the publisher. (I can’t imagine why, he is a brilliant technical guy.) Since my expertise is in marketing, I succeeded in getting completely tangled up in the technical issues: I started to study the problem of which major WA packages could all accept the same campaign variables and which, like Google Analytics, had to have specific tracking code.

After I was done, I went up to 30,000 feet and wondered — did it really matter? After all, if advertisers are seeing fewer clicks than the publisher is showing, having them nicely tagged isn’t going to change anything, now will it? The advertiers won’t see more clicks, just pretty web analytics, right?

So I wrote Dylan Lewis of Intuit. Dylan wrote Hack #51 in Web Site Measurement Hacks, calculating click-to-visit drop off. He agreed that tagging won’t make clicks show up, but offered these pearls of wisdom:

Without making sure that everything is tagged, you will never properly diagnose the discrepancy because the argument of “not everything is tagged” will trump both the person serving the ad and the person serving the landing page.

Separately, Ian added at least one other reason that the effort was worth undertaking:

At least with the query string method they can use thresholds and averages to see when something truly unrealistic is going on that needs to be looked into.

So there you go — another lesson in web analytics that is completely unintelligible if you don’t understand campaign codes. (So if you have to learn only one thing about your web analytics, learn how to tag links with campaign code. It’s a great investment of time.)

Robbin Steif
LunaMetrics

I really want to spend more money with a pay per click engine that caters specifically to the techiest of techie markets. But I can’t (or at least, haven’t so far) because I am scared that the button I push will kill all my current work instead of letting me spend more money.

Although this isn’t Google or Yahoo or any well-known engine, it’s the same concept. So, imagine that you’ve got lots of Ad Groups and campaigns and keywords set up, you’re spending money, you’re doing well, and you want to push the envelope a little. And then the website gives you these two choices when you try to expand your campaign to areas that they suggest:

Maybe they are saying, add these new purchases to similar categories? Add them as completely new? Finally I gather up the courage to actually click on one of these buttons and sit and watch the screen do … nothing. At this point I think, Oh, it’s definitely deleting all my work, so I hit the back button and log out and at least my successful campaigns are still successful.

Sometimes, it’s really hard to spend money.

Robbin Steif
LunaMetrics

After my podcast with Eric Mattson, I spent some time talking to him about what he is up to. I am actually one of his listeners, and although I pick and choose among the podcasts, I have noticed that there is a strong analytics bias to them. (If you can call it that.)

Two of the topics that we talked about were really of interest to me:

1) The Long Tail. Eric is currently reading Chris Anderson’s book by the same name, and so I complained about the difficulty of managing the long tail. Everyone oohs and ahhs over how wonderful those long tail words are (my example is always, “Honda 5-speed 4-door red EX” — when you type in a tail word like that, you are ready to buy.) But I find that when I work with long tail words, the customer spends more on me than on the keywords. Eric’s take on that one is, it has to happen by itself. As soon as you try to manage the tail, it breaks. “Just crank out great content,” he advises, and visitors will find you in the natural search. (This echoes the advice I gave to Justin Cutroni about blog marketing — when your blog is big enough/has enough content, visitors start on the interior pages because they find you in the organic search.)

2) Relationships. Eric maintains that relationships are the ultimate competitive weapon online. Although he started off referring to mass relationships (like the relationship that Google AdSense advertisers have with Google), I think that we don’t focus enough on real one-to-one electronic relationships. The kind where you’ve never met someone, you start writing them professionally, they write you back, and they become your friend. They send you business, or recommend your blog, or recommend that you speak at a conference. Most important of all, they just plain old help you with what you don’t know while you help them with what they don’t know, and everybody wins.

For example,(as I pointed out in a recent post), I was reading David Meerman Scott’s excellent e-book (The New Rules of PR) on the plane to Portland. I finished the paper and scrawled on the front of it, “Write him.” And then I looked at those words and realized that I would never have done that two years ago. I would have assumed that David is a busy professional who has no time for fan mail (much less email that says, “you got 96% of this right, but let’s talk about the other 4%.”)

Ultimately, I didn’t write him, I just blogged about my thoughts — and then heard from him in the comments.

This phenomenon — professional electronic relationships — is a little bit like web analytics. We understand what takes place but we don’t know why. That doesn’t stop us from speculating…

*Do we create more electronic relationships because we don’t have time to go to conferences?
*Is it because we are on our computers all day (or at least, those of us in Internet Marketing are?)
*Is it because with broadband, our email is on all day long?
*Is it because more people work in smaller offices or at home, so electronic relationships are the Internet equivalent of talking at the water cooler?

One shouldn’t think that this isn’t about conversion. I’ve been having an electronic conversation for some time now with someone who is about to become my customer. And we’ve just been talking…

BTW, Eric is moving back to Seattle this fall and loves the analytics of CGM, so I hope those companies snap him up. (I told him he should work for FeedBurner, but they are in Chicago.)

Robbin Steif
LunaMetrics

Over at The Site is Dead, Matt Roche is having an interesting conversation about whether Google Analytics can be used by Google to raise your prices. Why, he asks, would you want to give your vendor all that inside info?

However, I don’t think that the goal of GA is to learn the inside scoop in order to raise prices. True, Google sets a minimum price for AdWords and I am sometimes appalled to see it come up as $5.00. But that is always for words that have almost no clickthrough. Busy, competitive terms have prices that are primarily (but, I will grant, not exclusively) market based.

I think we are forgetting all about the Google patent from March 2005. Specifically, let’s look at Claims 37 and Claims 45-47.

Claim 37 refers to ” an amount of time that one or more users spend accessing the document.” It is possible that they were referring to how slow your connection is, but I doubt it — they probably meant, web pages that visitors spend more time on get extra points in the Google search. And how do they know how long individuals spend on a web page, if not by looking at the site owner’s web analytics?

Claims 45-47 reference whether users do things to indicate that they like the site:

• Claim 45: Did the user “maintain or generate data indicating that the document is of interest to a user?” To my non-lawyerly mind, that translates to, did the user print the document? Export it to another format? Forward to a friend?
• Claim 46: Did the user add it to his bookmarks/favorites?
• Claim 47: Are users maintaining or generating data (i.e. doing the things they reference above in Claim 45) at a faster or slower rate than before?

If you wanted to know data like these in order to do a better job of ranking websites, wouldn’t it make sense to buy a web analytics company and then give away the software for free?

Robbin Steif
LunaMetrics

Here’s the link to my interview with podcaster Eric Mattson of MarketingMonger. Eric is on a quest to reach 1000 interviews with Internet marketing executives and entrepreneurs.

Robbin Steif
LunaMetrics

Last week, I read David Meerman Scott’s excellent and free e-book, The New Rules of PR. I was already pretty schooled in the SEO and PR thing (after all, I am a publisher). But David made a point that I really hadn’t thought about before. I always focused on the Dan Rather and Swift Boats episodes, i.e. a press release to a blogger can end up all over the Internet and eventually on CNN. David made a related but not identical point: under the old rules, only journalists saw the press release. Today, publishers on the web, like me, may pick up the press release in its entirety, and that means that regular people - not just journalists - are reading press releases.

Now that non-journalists are reading our releases — why are we writing them in the same old boring fashion? Why aren’t we writing for an audience that doesn’t want to read marketing spin? In fact, why do we even organize them the old fashioned way?

I think I was fated to have read his e-book when I did, because I started the next day with a phone call from Traci Hailpern, VP of Marketing at Feedburner. “We’re buying Blogbeat.net,” she explained, and asked if I would contribute a quote to their press release.

Both Traci and I were really tied up that day, so their PR agency wrote a quote for me, which they read over the phone while I was supposed to be eating lunch. Here’s what the PR agency wrote for me:

“This acquisition is a logical extension of FeedBurner’s valuable service,” said Robbin F. Steif, CEO of LunaMetrics, LLC. “I’ve been thrilled with each of these services separately and have wanted a more efficient way to determine how my feed subscriber trends relate to my blog traffic. I continue to be impressed with FeedBurner’s ability to keep its customers’ best interests in mind as the company grows.”

It’s really not bad. I give the PR firm credit for understanding Feedburner’s business. But if everyone on the web is going to read it, why don’t we talk like real people instead of like marketers?

Here’s what I dictated to them:

“I had been using each of these services separately and am so excited about this acquisition — the very best feed analytics company purchasing the best blog analytics company,” said Robbin F. Steif, CEO of LunaMetrics, LLC, a web analytics and conversion consultancy. “Now I can go to one place for all my blog data. Congratulations, FeedBurner, for making a great choice!”

Now, Feedburner being the incredibly cool company that they are, no one really cared about this particular release. FB created a special page with funny FAQs and a great cartoon and the whole world linked there instead. (Truly great marketing.) But most other companies don’t have that same sense of style and are still relying on their press releases — so why don’t we start writing them to be read?

Robbin Steif
LunaMetrics

Feedburner announced their acquisition of Blogbeat, my favorite blog measurement service.

I have been wanting to review Blogbeat for some time now, but instead, I have been sending email to Jeff Turner, formerly their CEO and now lead engineer at FeedBurner. My first email subject line, last April, included a laundry list of things I wanted him to fix or improve. “Soon I will be reviewing you in my web analytics blog,” I wrote, “but I would like you to fix all sorts of problems first…” He got right back to me with a note of appreciation, and I wrote,

You are so great to work with. I work with [fill in a big analytics company name here] where the software costs $50,000, they take a week to answer my questions, and they don’t answer in full.

So that’s one of the reasons that I love Blogbeat. The other reason is, it provides just the level of functionality I need for my niche blog.

I do run MeasureMap and GA on my blog, but I almost never look at them. MeasureMap doesn’t let me drill down to the user level. As user-friendly as Google is, it’s just too big for a small blog where the questions are primarily:

• How did people come to me: referral, search, direct? (This info comproses the Blogbeat overview)
• What were the referrals, what were the searches?
• Who came to me and what did they look at? What did they want to know? Did they look at more than one post?
• Did they subscribe?

In other words, I want to know, what kind of buzz is my blog generating and what do my readers care about? I also want to know if they converted.

I would never work at this level for a lead generation website. That’s because I want to know things like how much revenue was generated by marketing source (so, “people who clicked on a specific banner ad converted into a lead x% of the time, the cost per lead is $85, and a lead is only worth $45, let’s fix this.”) But most blogs - especially small blogs - are different. Conversion may be, how many pages did they look at? Conversion may be, did they click through to the rest of my site, or did they subscribe to the feed, or did they click on the AdSense? Blogbeat addresses all those onclick events in their Links section.

Certainly you can argue, it would be nice to be able to tie together more of the data in Blogbeat. So if Blogbeat already knows which IP address came in via which method and what they searched for and which pages they looked at, can’t they tie that to onclick events so we can see what causes a conversion? That’s one of the things I am looking to Feedburner to do.

Although the Blogbeat city cloud feature is lovely, I primarily look at the detail by visitor. It enables me to look at an IP address and see this (although I have deleted the actual IP):

Notice that I get all sorts of information here. I see that this is someone who lives in Coldwater, Michigan, they found my blog by searching for “causation vs. correlation” on Google, they landed on a post that I know will do a good job of answering their question, this is their first visit and they only looked at one page. (And if I cared, I would notice their technographics, IE6 and Windows XP.) If they had looked at more than one page I would have clicked on the “more…” to see all the pages they viewed.

Since my audience is not as big as John Battelle’s is, I have the luxury of scrolling through those visitor pages every day. Reports that show averages and medians are great, but this is the online equivalent of getting out and meeting the people.

So once again: Congratulations, Feedburner. You made a great choice. However, I’ll be sending you my punch list of improvements just as soon as the noise dies down.

Robbin Steif
LunaMetrics

ps Need more information on the acquisition and what it means to subscribers?

I spent 25 hours in Portland, Oregon. Not only is it a jewel of a city, but everywhere you go, you meet web analysts.

I finally got to meet Brent Hieggelke, my Web Analytics Association co-chair. Brent used to be CMO at WebTrends but is now CMO at a London-based start-up, TouchClarity. TouchClarity uses neural networks (and probably genetic algorithms) to look at each visitor’s click stream and dynamically change the layout of the screen. That way, for example, the individual who is always clicking on sports stories will get served up ads for Sports Illustrated, and the person who is clicking through on beauty stories will get fashion ads. Although this might sound vaguely like split path testing, whereby different people see different options, it’s not the same at all: the goal of split path testing is to determine which one design works the best for everyone, whereas the goal of TouchClarity is to always customize the website for every individual.

I finally got to meet Loren Hadley from L.Hadley & Associates. Loren also works with Brent and me on the WAA Marketing Committee. At lunch, the talk turned to Google Analytics and I pointed out that the new invitation wait time is down now to about three weeks. Loren then told everyone that he had purchased an invitation from China for $.99 on eBay and received it instantly.

After corresponding with him for over six months, I then finally got to meet fellow analytics blogger Eric Butler. We went out for dinner with two of his colleagues from WebTrends, Kevin and Clay. Kevin gets all the credit for the title of this post (although just about everyone I met in Portland is in love with it. Brent practically dragged me over to a window to see Mt. Hood.) And then Greg Drew, CEO of WebTrends, magically appeared in the restaurant so that he and I could finish our WAA business. (I use the word “magic” because he had no idea where I had gone to dinner or that I was going out with Eric. Or that I even knew Eric. Am still not quite sure how he figured it out.)

I even got to meet with my friend Andrea Hadley (no relation to Loren, but they both like meeting other Hadleys) of NetSetGo from Vancouver, who offered up the observation that the WAA is like Canada — multicultural. (I think Switzerland and “neutral” would have been a better analogy, but hey, she’s Canadian, let’s give it to her.)

Finally, if there are any WAA types out there who would like to be on the Marketing Committee - send me email, steif at lunametrics.

Robbin Steif
LunaMetrics

Here it is: an easy way to remove yourself from your Google Analytics.

Two months ago, I wrote about how hard it is to remove yourself from the data when the web analytics solution only allows you to filter on IP address. My issue is always about having a dynamic address but, as was pointed out to me today, maybe you only want to remove certain computers (not certain IP addresses) from the data.

Since then, this blog gets more hits on “remove yourself from GA dynamic IP” or some variation than any other keyword. I did write the guys at Google and they said, “Really?” But Google is always up to something so maybe they will work on this.

This is why I was so excited today when Justin Cutroni did a post in which he shows you how to create a cookie that takes you out of your GA. You copy his file, paste it into an editor like Notepad, save it with a name that ends like your website does (I had to save mine as filename.asp since I have an asp site, but yours might be .htm or .php or something else) and then use your FTP editor to drag your new file into your website. Then follow his directions. It is really easy, especially after I did it and wrote him twice and he edited his post and now even non-programmers can take themselves (or more correctly, their computers) out of their Google Analytics. Thanks, Justin, for all the help (and of course, thanks for the cookie-generator.)

I just want to end with a plug for blogs that give really concrete “here’s how you do it” advice, like Justin’s and ROI Revolution. (ROI just did a fabulous piece on differentiating links in the GA overlay. And did you all notice that Google had to name their blog, “The Official Google Analytics blog,” because ROI’s is the unofficial one?)

Robbin Steif
LunaMetrics

I don’t usually beat up on my friends, but today I just had to take issue with a comment that June Li make in her excellent blog, Share the Genie’s power.

June’s overall premise is a good one, site owners should design for measurability. However, I disagree with one of her points: one-page shopping carts should be broken up into multiple pages so that website owners can do a better job of measuring where people abandon their carts.

First, there is a technical solution: some analytic packages will let you create events in each field of the shopping cart, so that you can see exactly at which field the customer jumped ship. However, not every package is so sophisticated.

So, if your package doesn’t allow you to create an event for every field in the shopping cart, should you “chunk it up” into multiple pages, as June says in her blog? This takes me to point #2: Absolutely not. MarketingSherpa’s recent study of e-commerce sites addresses the value of a one-page shopping cart. While they do point out that they have limited data, we all know that clicks are precious. If you have a one-page cart, you are probably performing much better than you would with a multi-page cart. True, you can’t measure abandonment of the multi-page cart as well, but which would you rather have: the money or the measurements?

Robbin Steif
LunaMetrics