I know that lots of readers don’t use SiteCatalyst, so I will write a minimal number of posts on the Advanced Omniture training I did (and I’ll space them out.) One reader did ask for some Golden Nuggets, so here they are:

Golden Nugget #1) When you see an icon that looks like a broken egg, that indicates you’ve enabled traffic correlation. When you see an icon that looks like a spyglass (the kind that every other WA solution uses to drill down), that indicates you’ve got a commerce correlation. Most traffic correlations have to be created, which may just require a phone call to LiveSupport - that’s how I got a correlation showing me exactly which terms visitors typed into the organic search to land on each page, page by page. (So I suppose they were correlating their standard organic search term report with their standard entry page report.) However, lots of commerce correlations are already there, you can just see them in the data — just look for the spyglass. (You are wondering why this is in the classifications post? Read on.)

Golden Nugget #2) Classifications always have to be children of a standard variable, like product, pages, etc. However, not all classifications inherit the correlation abilities of their parents. So if you take a paths variable like pages and create classifications (sample classifications for pages in a media site: arts and leisure pages, sports page, business page), you won’t be able to do the same kind of correlations that you might with pages. On the other hand, some variables (I believe they are product, customer loyalty and campaign) are “fully subrelated” — that means that they can always pass their relational genetics onto their children (classifications.)

So, for example, I might own a website for women’s sweatshirts, but all my products are just SKU numbers — not very helpful. So I could create classifications and group the sweatshirts by Type (hoodies and non-hoodies. Now I can pull up Products > Type and see all my commerce metrics for hoodies and non-hoodies.) Everything that I could normally correlate with products can be correlated here. For example, with Products, I can look at not just the Big Three Sales metrics — Revenue, Orders and Units — but I can also look at cart additions. Product views. Finding methods. (Some of those get pulled in by clicking on the spyglass icon, since that’s the way that correlation is signified in commerce.) The point is that all the children of products, customer loyalty and campaign can have the same correlations as their parents, but the same is not true of classifications belonging to other parent variables.

Golden Nugget #3) Omniture works hard to call things by names that are meaningless — but then, if you’re still reading, you know that already. WRT Classifications, they have a neat tool called SAINT — probably some kind of backronym. However, they should have called in the SC Excel uploader. It’s pretty easy, even for non-techies — you just create your classifications in Excel and use SAINT to upload them. (It’s useful for other uploads, but we won’t go there today.)

Robbin Steif
LunaMetrics

PS I am hoping that all the people who work with SC day in and day out will get on an make additions and corrections. For example, I was unable to find a Path or Traffic variable that had any preset correlations, i.e. correlations that I hadn’t created myself. So that addition would be really useful.

Today, SalesGenius was launched, and when I understood that it it ultimately a form of phishing, I just had to try it out. (Like Mimi says in Rent, “I was born to be bad.”)

Here’s how it works: the salesperson, presumably, sends email to her prospective customer using the SalesGenius website (the Outlook plug-in is still in beta.) As part of the email, the sales rep includes a link to her own company’s website. And here’s where the phishing (and the intended value) come in. The link automatically redirects through the SalesGenius server, rsvp1, so the url comes up looking like this: http://www.mysite.rsvp1.com. However, the unsuspecting potential customer doesn’t realize that he was redirected unless he happens to look at the address bar, because the page is the same as the one he is expecting to see. The salesperson can now get on her account at SalesGenius and see which recipient of her email clicked to the site, which pages the potential customer looked at, in what order he looked at them, and how long he stayed on each page.

To test it out, I created a free 30-day account. Then I sent email from my new SalesGenius account to three web analytic/conversion rate bloggers - I just told them that I was testing invasive software and asked them to click around my site, on more than one page. When I created the email, I was going to create a naked link, www.lunametrics.com, but the SalesGenius software encouraged me to bury my site link in a phrase, like this.

Here you can see the first screen shot, where Pat McCarthy from Conversion Rater was on the site while I was looking, Clint Ivy from Instant Cognition had opened the email but not visited the site, and Eric Butler from Inside Analytics has already been to my site.

Next, I could select any of the people who had clicked through to the site and see what pages they had visited and for how long:
Finally, I was able to “replay” their visit:

This last step seemed a little redundant, I had already seen it (albeit by page name) in the step before, but if the URL names are not very descriptive, or the clickstream was a long one (or the salesperson doesn’t know the website well) it might be helpful.

You can imagine how much I disliked the deception here. Here is an excellent review on CNET, The Case of the Spying Salesman. It’s also worth looking at the very interesting comments on the TechCrunch review earlier today.

Many thanks to Eric, Pat and Clint for their help.

Robbin Steif
LunaMetrics

It’s possible that I’m the only web analyst around who didn’t understand the interplay between Google Analytics and Urchin (but I doubt it.)

I have a customer with Urchin logfile software whom I wanted to upgrade to Google Analytics (and I even got them an invitation from GA while I was at the eMetrics Summit.) While I was at the Summit, their Urchin reseller set up their current Urchin 5.0 software for their new website, and told them that there would be an upgrade. I was sure that Urchin is going to die and will be replaced by Google Analytics completely; after all, Google bought Urchin for just that purpose. However, Tim Seward from The Unofficial Google Analytics Blog assured me that I was wrong — he even called me from his son’s ball game on Saturday morning to tell me how wrong I was. Server side Urchin is not only alive and well, but will be upgraded to Urchin 6 sometime in 2006.

The Google Analytics site says,

Urchin 6 software will not be free when it is released in 2006. However, if you paid for an Advanced Support contract that expired after March 2006, you will be offered a free upgrade to Urchin 6 software when it is available. If you are looking for a free analytics service, please sign up for Google Analytics.

And of course, if you want client side analytics, you have to use GA.

Robbin Steif
LunaMetrics

RevealSite is unlike any other WA software I’ve worked with, because it is not about trends - it is real time monitoring of who visits your website or blog. They don’t use tracking cookies - they match IP address strings, “but most people in the US don’t have dynamic IP addresses,” Addison Schonland from RevealSite told me on the phone. (I haven’t had a static IP address since I started with cable in 2002 — and now I have fiber optic service — but I haven’t taken a poll on this topic.)

The software enables you to watch as individuals (or more correctly, their IP addresses) wander around your site. If you know who they are, you can rename them. You can send a text message (”It looks like you’re having trouble with that software download, how can I help you?”) and you can block someone, like a competitor whose address you recognize, from your site. You can change the color of a specific IP address to quickly alert you when someone special or problematic is on your site. You can even create a Welcome, You’re Special page and send it to replace the page that the IP address is looking at (or for that matter, a page that says, “We don’t need your type of IP address hanging around here.”)

On the one hand, I’m fascinated, on the other, I’m scared silly. “What’s so scary?” Addison asked (and I paraphrase, since I wasn’t taking notes.) He said, “It’s my site, I can let in the people I want. If it were my store, I’d kick someone out if they were inappropriately dressed. It’s the same thing.” Well, is it? People like the perception that they are anonymous on the Internet. It’s true that an unexpected chat message might help someone who is pogo-sticking between the menu and the content, but will that person feel like Big Brother is watching? Some companies don’t even like to send email that says, “You left Widget X” in your shopping cart, because it breaks the illusion that the customer is anonymous. They find it more effective to write, “We’re having a 5% discount sale this week on Widgets X,Y and Z” and the customer feels that they just got a lucky break.” But you know, I have a special bias, I worry about conversion all the time. Other kinds of sites may have different concerns and so RevealSite may be just the ticket for them.

On the technical front, the monitoring dashboard runs on IE, and it seems to do a better job of picking up visitors who use IE than those who use FireFox (when Addison demoed it back to me from his site, he couldn’t pick up my visit until I opened his site in an IE browser.) After a three hour trial, I had to take it off my blog because it forced the entire page to scroll down to the bottom.

Addison and everyone else - you are welcome to comment and tell me how wrong I am.

Robbin
LunaMetrics

Last Friday, I wrote what I thought was a pretty nice review of StatCounter, a little free web analytics package. I included this line, “And one should never forget that, like many client-side analytic packages (i.e. the kind where you don’t need server logs), they don’t collect data about bots (like the GoogleBot), because most bots don’t read javascript.”

Being a good cross referencing blogger, I then went to the StatCounter forum, where I have never written, and referenced my post in their “Do you like us?” section. I was really surprised to see the Master Member, Christine, write back this post:

“Compliment or Competitor: Forgive me if I’m wrong or paranoid. Bur aren’t you in fact a competitor of Statcounter’s? Your flagship website from your sig doesn’t use Statcounter…”

Wow. I was expecting, “Welcome new member” or maybe, “Thanks for the great review.” I never figured out how using their code on only my blog made me a competitor.

But she went on,

“BTW, you do have some factual errors when discussing Statcounter’s capabilities. The most strinking (sic) of them is that bots don’t get tracked by Statcounter. Statcounter tracks image enabled bots. The bots that don’t get tracked are not tracked because they are not image enabled, not because they are not javascript enabled.”

I wasn’t sure what she really meant, so I went back to Hack 23 in Web Site Measurement Hacks and read, “…a solely client-side data collection model (page tags) may not be able to collect all robot/spider traffic information, because some robot/spider agents do not execute JavaScript and generally do not accept cookies.”

Well, the author (Eric Peterson), wrote “may not” instead of “will not.” So I wrote Fred Kuu from HBX Uncovered. Fred is “the Web Metrics Technical Lead at Adobe Systems” according to the HBX website. Here’s his answer:

Hi Robbin,
Most bots (aka spiders or crawlers) cannot parse and execute Javascript. This is why all vendors (based on JS tagging approach) tout that they track and report only human activity. Granted, it is possible for a hacker to program a bot to parse the Javascript but it’s not easy and there’s not much of a gain by enabling it.

Now, regarding images, almost all bots (especially search engine ones) will be able to track if a page contains images but majority will not actually request the image. So in the logs, virtually all bot activity are to web pages and not to images or other binary files.

-Fred

Then I went to my SiteCatalyst user manual (as another reference point - it is also a client-side package) and it said, “SiteCatalyst does not track spiders since they do not load images.”

So finally I wrote Jon B. in London. The great thing about writing the other side of the ocean is, you write them at night and the answer is in your in-box before you wake up the next morning. Here’s what he said:

Hey Robbin, how goes it?

Strictly speaking, bots don’t execute Javascript. The javascript is responsible for loading the image hence there is some indirect truth in the sense that the bots don’t load images. Also bear in mind that bots can load images - that’s exactly how Google images sources its images database-index…

So, I put all this data into my blender, turned it on high and came up with this: If a bot reads a picture (like the Google Imagebot), a client-side solution like StatCounter can pick them up if the company decides to enable that ability. However, most bots are not about pictures, they are about finding your text, and those bots don’t get picked up by client-side solutions because they don’t talk to the javascript.

Robbin Steif, CEO
LunaMetrics

Let’s talk about free web analytics.

While Google Analytics are really awesome (I’ve worked with customers who already have them), I’ve been on their waiting list since November. So I don’t even bother to recommend them to customers anymore. If the website is mission critical but not a profit center, I recommend NetTracker or ClickTracks. (Note: I hear that NetTracker 8.0 is going to “do money.”) For e-commerce, SiteCatalyst is my weapon of choice, although I’m sure I’ll love HBX eventually. And there are lots of other great packages that I haven’t touched upon. But for customers who have ten-page sites, I really encourage the use of StatCounter. Here are the top reasons I like it:

#1. Most other free analytics are server side, i.e. you have to have access to the actual web server to be able to install the software and then run it against your logs. Ten-page sites are almost always hosted at some ISP, and installation of StatCounter is much easier.

#2. StatCounter makes it really easy to exclude your own computer from the analytics, even if you have a dynamic IP address.

#3. StatCounter provides rudimentary clickpaths, so I can watch how people come to the site and then where they go (and if they reach a “Thank you very much” page, how they convert. It’s not Omniture, but one has to keep the price/performance ratio in mind.)

#4. StatCounter has drilldown. So, for example, I can choose the referrer tab (what they call “recently came from”) and next to each referrer, I can click on the spy glass to see the IP address, which is often resolved into a company name that I recognize. It’s not NetTracker, but again, there’s that price/performance ratio…

#5. In addition to drilldown to IP/domain, StatCounter shows you where the visitor is physically, on a map. Or at least, where her computer is.

#6. On their summary page, StatCounter lets you choose the time period you want to measure.

It also does all the things you would expect of a free package. Entry pages, exit pages, search terms, etc. Of course, there are lots of things StatCounter doesn’t do. Other than letting you choose a time period, you can’t customize anything - no customized dashboards, no campaign management. And one should never forget that, like many client-side analytic packages (i.e. the kind where you don’t need server logs), they don’t collect data about bots (like the GoogleBot), because most bots don’t read javascript. So for the first time today, after I had a new client put up a “coming soon” site, I wrote the developer and said, “Please load AWStats. [A log-based, not-as-interesting-but-still-free-package.] My only goal right now is to see if the bots/spiders are coming to visit.”

Robbin
LunaMetrics

WebTrends recently published a 34 page report on analytics (which you can get here.) This finding really amazed me:

In our own 2005 Web Marketing Confidence Report, we surveyed over 250 marketing professionals and learned that only 5% are currently “very confident” in the measurement of their results. In fact, 26% admitted to “flying blind”. Why? Because less than one out of four had a complete view of their performance metrics—including conversion, revenue, ROI—to gauge the success or failure of their marketing initiatives. The majority, a combined 52%, are still relying on clickthrough rates or have no campaign metrics at all.

For those of you who aren’t looking at your metrics every single day — don’t you wonder if you are shouting into the wind? (No one can hear you?) And if you respond, “Well, we have e-commerce so we can see that people are purchasing,” you are leaving lots of money on the table by not having analytics.

When I first got this blog, I was sure there were web analytics attached, and I just couldn’t find the right button. It drove me slightly crazy to blog and blog but not know — was anybody listening?

Eventually I realized that there was no button, I just had to install my own analytics. I chose a small, free, client-side package because a) I only want to measure traffic and marketing referrals and b) I need a client-side package in order to be able to measure this particular site.

Analytics are magical. They are instant feedback. When I installed analytics here, I learned that I wasn’t shouting into the wind. More interestingly, I learned that more of my audience wanted to hear about analytics, as opposed to just great ways to increase conversion rate, and so, I’ve started to concentrate on that topic a little more (as you see!)

Robbin
LunaMetrics

I do an incredibly amount of work in analytics in order to increase conversion rates, and one of the things I do is help the tech-types create better reports. (Well, I help them because it helps me, but that’s teamwork, right?) So, for example, I helped one customer take his thousands of products and put them into about fifteen categories. That way, we can create a product report with the click of a mouse (and not get a list of meaningless part numbers.)

The report defaults to Product Category and Revenue. This week, one of the company’s employees created a third column, product page views. “Look,” he said in an email, “Product Category A gets hundreds of thousands of page views and very little revenue. Product Category B gets few page views and lots of revenue.”

In general, this is a great kind of analysis to do with one’s analytics — for each product, what is the purchase to view ratio? Often, we find that “hot” products do well only because they are seen often (e.g. they’re on the home page.) Notice that I wrote, “For each product.” However, the employee’s analysis was way off base. What he didn’t realize is that if a page includes 47 products, the number of category page views increases 47 times when a visitor looks at it once.

There is a more important point here, though. Whenever I have some big analytic pronouncement to make, I do two things. 1) I ask, does this make sense? In the case I described above, the numbers were way out of line. 2) I triangulate. If the data leads me to a strange or unexpected conclusion, can I find it in a different way? In this particular case, I looked at page views of individual products and saw the problem right away. But triangulation can take other forms. For example, maybe you do have a really awesome analytics package. Have you thought about using a little free one at the same time? That enables you to say, is this a problem with the package I am using, with how I am interpreting the report, or is this truly something that matters?

Recently, I did some keyword work for this same customer. They have one of those big functional analytics packages, SiteCatalyst by Omniture. But I still installed Google’s conversion tracking and waited a month (to collect data) before making any drastic decisions. The SiteCatalyst report gave me lots of data that I couldn’t get from the Google AdWords conversion tracking, such as conversion over multiple visits. Nonetheless, having the Google data made me feel really good about making an important AdWords decision, because I could tell from the Google data that the Omniture data was in the right ballpark.

But I guess that to truly triangulate, you need three sources of data….

Robbin
LunaMetrics