iMind Software

Since we deal more with our own products like joszaki.hu besides other special projects for our clients, we deal more with following the users in the systems. The most basic tool for it is Google Analytics, which now we use for not only web-based, but for mobile environments, too. The actions can be saved even in an offline app, and be uploaded later through API. It's an obvious advantage that we and our partners can browse a well-known and professional surface.

As we had become more and more interested in this stuff, we bought the second edition of Brian Clifton's Advances Web Metrics. Nowadays the books from England arrive here in no time, without problems, but we'd still prefer an ebook reader. The author is an ex-Google employee, he knows the inside part of Analytics, and nowadays works as a web-analytics consultant. The book is really interesting, it's worth to read even those parts that deal with the basics, because it gives a lot of background information. It's clear that the author has observed and analyzed the reports for many years. 

For me it was really entertaining to go deep into Analytics, because I had to think it over as a product developer, as a marketing person and as an engineer at the same time. It's a good exercise with the puzzles that are similar to programming. Many people use Analytics, but surprisingly few utilize its knowledge, although with a few clicks and a few lines of javascript, miracles can happen.

These are my favorites for now:

Intelligence

In the intelligence menu, it searches outstanding strangenesses (like the purchasing intensity of people living in Budapest has risen, or they spend more time on the site). We can also add filters manually (like the visit fell by 50%), and we can set to get an email notification when the condition comes true.  

Goals - transactions

That's the first thing which everyone should set, just to force yourself to mark the goal of the homepage. It's easy if it's an e-commerce page (we sell, we're happy), but it's also useful on other sites. E.g. sending a sales lead mail, or subscribing to a mailing list can also be useful. And you can also count its value. If we stick money to these things, too, (and it's worth to), we'll see the produced value of them in each and every report, e.g. splitted to referrers.

Virtual pageviews, events

We must use virtual downloads and events because of the spreading of AJAX systems, since the visitor is not necessarily moving between the sites. 

Custom variables

This is beautiful. :) With a function, we can label our visitor, and we can tell for how long the label should stay there (on page, session or visitor level). If there's any important feature that segments our visitors, we can distinguish between them very easily by this tool, and we can instantly see how the certain segments perform, and what profit they return. 

I wanted to survey my knowledge after studying from the book, so I went throught the Google Analytics Individual Qualification test. It brought back the university exam fun, but it was fine. My result is 87% which is not bad, maybe if I had gone through my answers once again, it would have been better.

What was interesting for me is that these online tests give you a lot of time, and you can even stop the timer. So theoretically you could look up every answer on the Internet, and what's more, they don't even write that you shouldn't do so. Fortunatelly, I didn't need it, I just checked a few little things, mainly in the already open Analytics account. I recommend it for everyone who's curious whether they can distinguish between _trackPagevies and bounce rate. ;)