… but the CrazyEgg guys do, and I wish them the best of luck. They have a very, very Web 2.0-y analytics and split testing program. Well, its not a full-featured analytics program — it basically tells you whats being clicked on, and thats it, with an extraordinarily cool (er, is that hot?) heatmap visualization of it. It is amazingly easy to get started on their free trial, so amazing that they could teach Google a thing or three about it. Put in email address, put in password, fill out two line form about page to test, copy one line of provided HTML into page, upload. 30 seconds or less to the shiny. If I get nothing else out of my free trial, I’ll have gotten a good usability example for Kalzumeus.
Here’s my concern: while they are much easier to use than Analytics (they will, for example, track clicks on my links offsite or to setup files without any special work on my part), I have already climbed Mt. Analytics Learning Curve and can now bask in the awesome view — such as dicing visitors down by where they came from, watching search terms trends, and tracking click paths around my site. For free. Their free trial (5,000 visits tracked a month) will work for a few more weeks for me, but then I’ll go over the traffic limit and it comes down to the “Do I really want to pay more than my web hosting budget for an analytics program?” question. Which would be “Oh, heck yes!”… if I didn’t already have Analytics.
Then they have this split-testing option, where you can have them run, say, the stats on 2,000 consecutive visitors. Then you twiddle the page a bit and rerun the test, getting 2,000 new consecutive visitors, and compare. That is both great and, again, inferior to the free Google Website Optimizer*. Now, if they could just bring their considerable usability chops to solving the biggest problem with Website Optimizer, which is that it is a freaking pain in the keister to use, I’d give them $9 a month and a gold star. I have been meaning to start using Website Optimizer for months but every time I do I get to the start of the process, figure its going to take me two hours to get anywhere, and then get distracted by other demands on my time.
* Note that the key problem here is that Website Optimizer does actual splitting — for example, if I get 4,000 visitors they get parceled out to the two alternative pages randomly instead of first 2,000 to one and next 2,000 to the other. That is fairly key for me, since my business has very definite cycles in it, and I’m doing all sorts of things in the background. If, for example, 2,000 visits happen Monday through Friday, and then the next 2,000 visits happen Saturday through Wednesday, then I naturally expect the “Download a Free Trial” link to get less activity anyhow in the second half because my best customers (teachers) aren’t typically on the Internet on Saturday or Sunday. That throws off the results of the test.
