We interrupt our regularly scheduled programming (which was Part 3 of the Avinash Kaushik book addendum) to hear from Princess Angela Brown. The princess (who points out that she is of relatively minor lineage) is also co-chair of the Standards Committee for the Web Analytics Association. Today, her Committee announced their 26 Standards (finally), at Search Engine Strategies - San Jose. We bring you here the exclusive inside story, also known as, “What took so long to write 26 definitions?”

Today the WAA Standards Committee released its second definitions document; eight months after our “Big Three” definitions were released. Eight months seems a ridiculously long time to define 23 new terms (the “big 3″ were carried over from our previous doc). After all, I could have written this document myself in under eight hours. Jason, our co-chair, could easily have done the same. In fact, nearly any of our committee members — more than twenty very competent people working as web analytics consultants, practitioners, and vendors — could have written this document in a day, blindfolded, with one hand tied behind their back and balancing on one foot (yes, I have a lot of confidence in our committee members!). What took so long?

Web analytics is not rocket science. Rocket science uses far more Greek letters and squiggly things than even the most complicated web analytics problem. The questions web analytics sets out to answer are really quite simple: WHO came to our site? WHEN did they visit? WHERE did they come from? WHAT did they do? HOW did they do it? Coupled with good marketing research and/or usability tests, you can even get a good idea of WHY your visitors do what they do.

As I see it, there are two issues that make web analytics harder than it looks. First, even though the concepts are simple, proper execution can be complex. To get a lot of value from web analytics, you really need to segment the WHAT by the WHO, and the HOW by the WHY, and the WHERE by the WHEN by the WHAT by the WHO. Second, a lot of our existing terminology comes from the tools we use. That’s not awful in and of itself: there are a lot of very good web analytics tools out there, and all of them deal with essentially the same concepts. But going from one tool to another is like learning another language, and no matter how well you know your stuff you are bound to misinterpret something because similar terms are used to describe different concepts. To use a cliché, the devil is in the details, and it’s that devil that took up so much of our last eight months.

We are fortunate to have a wide variety of Standards Committee members who have experience using different tools and analyzing different types of websites. This has led to some lively discussion about the meaning of the terms that so many of us use every day, and has underscored our industry’s need for precise terminology. For example, do you know the difference between a repeat visitor and a return visitor? How about a landing page versus an entry page? Single page visit versus bounce? Visit versus session? (The last one’s a trick question.)

For answers to these questions and more, download our document from the WAA site. We welcome your feedback.

In addition to her work as a guest blogger and her royal responsibilities, Angela Brown is the Web Analytics Manager for the MD Consult site at Elsevier. She has also been known to work for a large web analytics software vendor as a professional services consultant.

As I explained in Part I of this series, after I read Web Analytics: An Hour a Day, I had a lot of questions (and even some things I didn’t agree with.) So I wrote the author, and he sent me back nine pages of thoughts. That’s why I’m chunking my interview with Avinash into sections. Unlike the first part of this series, this one is very down in the weeds; I asked about some very specific best practices. You can see my questions below in boldface and his answers in quiet type, perfectly matching our personalities.

Moving from the very general to the very specific: On page 33, I scribbled, “First party cookies don’t talk to each other, and third party cookies get deleted.” Do you have any recommendations on choice of (vendor? technology) that deals with both of these issues? A first party cookie solution where the various sites in the enterprise talk to each other without lots of manual coding? A solution that you love?

In my prior role we had implemented first party cookies and “first party third party” cookies to overcome this challenge somewhat.

As an example let’s say my company was ZQ Insights and it had two sites www.zqinsights.com and www.webanalyticshour.com. I want to track each by itself and also the two pulled together.

I set a first party cookie on each (www.zqinsights.com and www.webanalyticshour.com). I am happy so far.

Now I also set a “first party third party” cookie on both, let’s call it tracking.zqinsight.com. The latter cookie I can use if I was looking at both sites as one monolith (to for example get true unique visitors).

It is less likely that this cookie will get blown away by spyware (because it is not being set from known domains of web analytics vendors), though high security settings will still be a issue.

One last point, you have to have a web analytics tool that allows you to create “local” (site specific) and “global” (all sites) datasets with ease and mix and merge sites. ClickTracks is one such tool.

I hope I have answered your question (and the answer is not clear as mud).

Question: On page 37, you stress the importance of having the analytics code at the bottom of the page (”Customer’s first.”) But what’s an analyst to do when the other fancy things on your page don’t work unless the tracking code loads before them?

Let me share some context.

The reason for the tracking code to last is simple: Nothing should interfere with the customer experience.

The page that the customer has requested has to go back as fast as possible so that they can get on with their life (and convert for example). Just in case you have something “funny” going on in the code, just in case your analytics providers servers are under heavy load, or just in case….. we want the customer to get the page first and us to get the data second.

There are always exceptions to any rule. I would set the bar really high to ensure that decisions to load the tag first pass rigorous scrutiny.

I think you do novice analysts a disservice by focusing so strongly on bounce rate. To you, bounce rate is about time on page. But most bounce rates are calculated as (visits entering and exiting on the same page without looking at another page) divided by (visits starting on that page.) Someone can bounce after spending 15 minutes reading the home page of this blog. So when you associate bounce with ways your website is failing (p. 145), the new analyst will be confused. I am not sure I have a question there, but you are welcome to respond.

For blogs my recommendation is that analysts should not measure either bounce rate or time on site. Both metrics will paint the wrong picture, precisely for the reasons you have so correctly identified.

Regardless of how it is computed for most types of websites Bounce Rate is a excellent metric that helps identify opportunities for improvement in acquisition strategies or website entry points.

On P. 274, you wrote that one of the questions you should be asking of your clickstream tool is, “What is the most influential content on the site? How do we know what convinced people to buy?” In general, how do most analysts figure that out, and specifically, how do you like to get that answer with Google Analytics?

I refer to a specific example of using the ClickTracks “funnel” report to identify influential content on your website. I am not aware of any other web analytics tool that can do that (or as easily as ClickTracks does), even if they all have “funnel” reports. It is something unique, and built into, ClickTracks.

If you have access to Discover2 or MarketingLab or your own data warehouse environment I suppose you can construct a complex query to replicate the ClickTracks logic. If you want to understand content influence you should so that, it is amazing what you’ll learn.

You can also use page level surveys (described in detail in the book) to understand value and influence of individual pieces of content on your page (and do it at scale).

 

Coming next - Avinash answers my questions about testing.

Part 1  Part 3  Part 4

Robbin

Justin Cutroni will soon publish the Google Analytics Shortcuts book. And it will be the best ten bucks you will ever spend, if you care about advanced GA implementation. It’s hard to believe that the book is almost here - I remember standing in the lobby of a hotel somewhere in California just a few months ago, and Justin whispered, “I’m going to write a book about GA.”

I am not exactly impartial here. This is a book that I have read over and over and over again. I appointed myself Editor in Chief and rewrote parts of it. “Didn’t we fix this utmSetVar typo once already?” I wrote the author last week. When I read the penultimate (I hope) copy last week, I found out that this blog is in it. And did I mention that Justin is one of my best WA friends? Like I said, not impartial.

So I am a little like Bridget Jones. She loves Mark Darcy, even though his mother buys him awful gifts and she seriously believes he should rethink the length of his sideburns. I love the book, despite its imperfections.

Since this is a real review, let me discuss the imperfections. First, I think you need to be a pretty advanced GA user for Shortcuts. If you are already reading Justin’s blog, religiously, you have definitely taken a step in the right direction.

I think Justin goes to great pains to tell you why GA works the way it does, information that is badly needed. But I think he would be smarter to have put some of that in the appendix — it is just way too boring until you absolutely need it. (And then, of course, you are desperate for it.)

Periodically, Justin lapses into GA-speak. For example, he writes this about the Item line in the e-commerce hidden form: “There will be one item line for each distinct product purchased by the visitor. This usually means one item line per SKU or unique product ID.” When I read this I feel like I need to create the I: line 50,000 times if I have 50,000 SKUs. (And you don’t have to do that.) In a similar vein he says, if you have e-commerce tracking, you can just leave the goal value blank. But this drives users crazy, because there is no way to leave it blank - GA insists on zero.

In a couple of (very rare) instances, I think he is wrong. But remember, I got a chance to point out problems all along the way, and he didn’t correct them - so maybe I am wrong. I am mostly thinking about applying AdWords cost data — you really don’t have to apply it to all the Analytics profiles that are linked to that AdWords account, you can choose, even though he says you have to link to all, and GA says so too. (Or maybe I am the only person in the universe who is always able to make this choice when I set up AdWords and Analytics.) And I am thinking about his Count Me Out! hack, which works fabulously to take yourself out of the data - but he also has a workaround in the book that doesn’t work. He says, use Firefox, go to the website where you want to be counted out, type this into the address bar, javascript:__utmSetVar(‘foo’), and you will create a utmv (a user defined cookie) called foo for that site. But it never works. Maybe it’s just this blogger who doesn’t know how to do it? (OK, I figured this out. When Justin wrote the book, he did it in MS Word, and Word assigns “Smart Quotes.” That’s how it knows when to turn the quotes to the right and the left, even though you only have one key on your keyboard. Anyway those special characters were gunking up the works.)

And wouldn’t it be great if the .pdf used the power of html? So that when he says, “I’ll be covering that later in my section on…. ” you could just click to it? (Maybe that will be in the final version.)

So when I write that you should drop everything and then keep dropping, i.e. drop ten bucks on this e-book as soon as it is available, it is not because I am starry eyed. I do see little imperfections, but still…. It is an incredible resource, and no GA analyst should be without it. I sure wouldn’t want to work without it anymore. That’s one of the reasons that I wanted to give away two copies to winners of the GA contest - I knew it was the perfect gift. The one you don’t have but absolutely need.

So salivate. It will be here soon.

Robbin

Did you have questions when you finished Web Analytics: An Hour a Day? I did.

The book was truly amazing. But when I was done, I had written all over it. Sometimes my notes read, “This is awesome, we have to try it.” But sometimes they read, “I don’t understand.” And other times, they read, “I really disagree.”

So I got an interview with the author. It came to nine pages (count ‘em, 123456789), so I am going to reprint it in parts. Avinash, you are truly wonderful for devoting this much time.

So let’s get started. And in my usual “in your face” fashion, I’ll start with a question that most people wouldn’t ask the Guru of WA:

Please allow me a quick interruption. The word Guru is of Hindu (Indian) origin and having grown up in India I have to say that I do not consider myself a Guru. One has to meet an astoundingly high benchmark to get that title and I am very very far away from even the starting point of meeting that benchmark.

For more context on that word here’s the wonderful wikipedia [definition]

In the introduction - why do you write that your book is for everyone? Is it for my mother, who is retired and spends lots of time taking care of my father? Is it for my daughter — the one who can drill down in her Quicken, but refuses to do anything academic? No, of course not. But that’s what customers do. We ask them, “Who is your site for?” and they answer, “Everyone!” So – who is your book for?

You got me.

Perhaps that was overuse of the word everyone.

Here are the specific people / roles that are mentioned in the introduction of the book:

· Mr./Ms. Web Interested

· CEO

· C-level or VP-level or just No-level person

· Marketer

· Sales Person

· Web-Designer

· User Researcher

· Analyst

The introduction describes how the book will be helpful specifically for each role.

As an example, here’s the one for Web Designer: If you are a Web “Designer” then this book will share with you how you don’t have to compromise on the number of ideas you can put on the website to improve the site, that you can have all of your ideas (even the radical ones) go live on the site and measure which one solves the customer (or your company’s) problems most effectively.

 

Question: I love the idea of surveying, continuously. However, it has not worked out well for me or my customers. We figured out how to delete the pop-up blockers, only to find out that customers hated it. And shouldn’t it be “customers first?” Any advice about the best way to *administer* exit surveys?

In my experience surveys that are shown at the right time with the intent of allowing the customer to express their opinion go ok. Typically we cram so much into a survey that only we care about that a customer looks at it, pukes and exits.

If you ask customers “nicely” they want to tell you about their experience.

My advice:

1) Experiment with different invitation types (pop up, pop under, on exit etc) and see what your customers prefer. And you only have to do this a few days each to get a feel for it.

2) Start with small number of questions (remember the “golden questions” post?) and then expand.

3) Put a really large close button. Make it apparent, clearly, that the survey can be closed. In a very subliminal way it works very effectively and actually gets closed less (if you have done #1 and #2 above first!).

From a mindset perspective you want them to share with you what they think rather than do a quick little interrogation with a battery of questions. Fine balance.

While we are on surveys - how do you feel about surveys that force the visitor to answer certain or all of the questions? It is infuriating to me when I answer a BizRate survey for the chance to get a “free” magazine, and they force me to answer questions (so instead, I just lie.) Thoughts?

I skip it.

I also rarely do any other surveys. I often read them to study them from a knowledge / awareness perspective, but I don’t fill surveys.

Here is the thing. I am not the customer and it is irrelevant what I think about surveys.

The first time we did a survey it was 20 questions and I was positive it would bomb, after all who in God’s name has that much time. Turns out that it had a consistent 18% response rate (compared to an internet standard of 1% response rate for surveys).

My lesson was that I should try not to impose my views and opinions and check them in at the door. Because I am not the customer, no matter how much I think I am. It is a tough pill to swallow because we tend to think we are “experts” because we have so much knowledge and data.

Experiment, it is cheap, see what works and what does not, refine and try again.

Why do you have a Trinity? I really see a duality, clickstream data and qualitative data.

It is qualitative (Experience), clickstream (Behavior) and the third prong is Outcomes.

Some people mix clickstream with outcomes. I choose to break it out for two reasons:

1) I want people to outrageously focus on outcomes. It is easy to be hypnotized by all the clickstream data and reports and forget to set goals or measure in a very hard core way outcomes. Yet the thing that drives action is not all your clickstream analysis, it is the tie to outcomes.

2) I want people to think of outcomes more than conversion. I am not big on obsessing about conversion, which will almost always lead to solving for a minority of your site traffic. Outcomes are improved customer satisfaction numbers, increased task completion rates, increased depth of visit over time, problem resolution rates on support websites, better recency trends on non-ecommerce website.

By putting outcomes as a separate part of the Trinity I am trying to emphasize the importance of understanding outcomes, and different types of outcomes beyond just revenue/conversion.

Do you think that makes sense?

 

Notice that I didn’t answer his last question (”Does that make sense?”) — I hope some of you will.

Coming next: Part 2, where I ask Avinash about first and third party cookies, where to put the code on the page when you are torn between conflicting needs, and other really “down in the weeds” Q&A.

Part2   Part 3   Part 4

Robbin

I just got my copy of Web Analytics - an Hour a Day. I am taking it on vacation to Hawaii. (I’m not taking a laptop or a Blackberry or anything, so I figure, I can spend some time reading.)

The first 300 people who send in their pictures with the book in their hands get an autographed bookmark from the author, Avinash Kaushik. So I gave my daughter a choice - did she want her picture on my blog, or did she want it on Avinash’s blog, where more people would see it, but no one would know who she was? Well, you can see which one she chose. BTW, I think letting people send in their kids’ pictures with the book is a brilliant idea.

Avinash, since you asked for all feedback, good and bad - I was so disappointed that the included disk said “CD-ROM” on it, but doesn’t work in my car (and what was I expecting? you wrote that it includes video, so it must be a DVD.) If it were a CD, though, I really could do it in an hour a day, half an hour each way to work and back.

Endnotes: Thank you, June Li, for telling everyone about the Google Analytics Documentation contest. Thank you, Bob Mutch, for teaching me how to do a better job with my Wordpress pictures (although I thought someone had set up the CSS and in the end, I just added padding to my picture, which achieved a similar goal.)

Robbin

… go on over to Dylan Lewis’s new web analytics wiki. The coolest part of the wiki (IMHO) is that the Google Analytics are public — and here is the username and password.

Now, he only gives read access, not administrative access. I went over to fix all the things I wanted him to have (and to figure out why his data is “accepting one goal” when there are no goals under the goal tab), and realized that I wasn’t an administrator. (*Sob* — I am so used to being an administrator.)

On the other hand, it is a great opportunity to see what they are like, if you’ve never seen them before. Plus, tonight, a reader was complaining about some v2 functionality; it was a complaint I had never heard before. “Show me,” I wrote him. “Go to Dylan’s wiki and use his public analytics.” What a fabulous opportunity — the reader can post the information to the world (well, at least to the 650 people who subscribe to this blog), and there are no NDAs required.

Now, if you have used GA, you really want to enter The Contest, right? (I never worked so hard to get a list of problems, but when I see the entries, I can see that it’s worth the effort.)

Robbin

If we had a contest, I might win with that photo of Ian and Dylan in the 3D glasses. Steve gets all the credit for that Photoshopped piecture of Avinash and Judah. Shareen, here at LunaMetrics, gets all the credit for bringing order to my PhotoChaos. Anway, click to enjoy - Robbin

This is a two-topic post.

First, Batman rules. When I originally post this link to LunaMetrician Taylor Pratt’s article about Web Analytics and SEO, he got 30 unique views (that’s how they judge this contest, by unique views.) But over 500 people read this blog, and while 30/500= a 6% conversion rate, still, I just need a little bit more help. If you already went to his story, you don’t need to go again (unless you are on a different computer — how is that for gaming the system?) But if not, please check it out, even if only for a moment or two.

Also, I got into SF way too late last night, and refused to go out with everyone for sushi (because they felt like it was 9 m and I felt like it was midnight.) I did manage to capture a few pics with my new digital camera. I haven’t figured out all the flash settings yet.

So here is Eric, with his new haircut. I offered to give him $15 and take him to Supercuts to get a really awesome cut, and he was offended, he said all the other women love his cut.

(To be fair, this is a really lousy shot. Like I said, I am still learning how to use this little camera.)

I have a great shot of Rene that I can’t get up here, for some reason, and I lost my picture of Scott Baldwin.

Finally, here is June. It’s a nice picture of her, but by this time, I had figured some stuff out on my camera, plus, I wasn’t sitting in the dark outside with 50 other people, balancing luggage at the same time.

When I am not desperately trying to get down to the workout room, I will get Rene’s picture back. And another picture of Scott.

And here’s that link again to Taylor’s article.

If you’re going to the Summit, please don’t forget the offer I made in February - come up to me if you don’t know anyone, and let me introduce you. Web analysts are the nicest people around. Plus I love being a connector; let me connect you. So here’s my picture and you’ll know which one I am.

(Notice how I still haven’t written about custom advanced filters. But Oh! am I using them.)

Robbin

Some quick notes: if you are going to be at the Emetrics Summit, we are having a table for WA bloggers on Monday. Any WA bloggers are invited, although we might need more than one table…

IE6 can be pretty bad for blogs with lines that go over the margins, I learned. Many thanks to Chris at Moody’s Economy.com and Joe at More Visibility for pointing out the problem to me. I was just kind of lazy when Chris wrote me (because I go home every night and work on rethreading almost 300 posts on Wordpress, it is hard to worry about this blog too.) So Joe, you were just the fire I needed to fix the blog.

Coming soon, the next part of the multi-part series on GA filters. If you are having trouble with a filter, send it to me, I am doing so many tests that I might as well include yours too.

Robbin