The Adobe and Omniture Acquisition: What It Means

For Sean Power’s predictions on the Omniture – Adobe acquisition, click here.

Adobe has a problem. They make great client software — Flash, Flex, Acrobat — that works with the vast majority of browsers. In fact, it ships with most of them. Despite attempts by Scribd, Silverlight, and others, Adobe’s technology makes the Web a more exciting place.

But Adobe only makes money when they sell the server side of all those tools. And they don’t have a monopoly on those sales. Plenty of software can save Flash or PDF formats.

With the acquisition of Omniture, Adobe may actually have found a way to make money from all those installed clients. It’s relatively easy to instrument HTML: just put in a snippet of code. But doing it in Flash or Acrobat is a lot harder, requiring some coding and instrumentation.

If Adobe makes it easy to track client behaviors in Acrobat and Flash, it can make itself the Google Analytics of the Rich Internet Application world. Done right, there’ll be widgets in the Eclipse-based Flash and Flex developer environments, and in Acrobat authoring tools like Illustrator. Imagine dragging a “Goal” object to a Flex view, or marking a text field as the “transaction value” for a session, or tracking how far down a document a particular reader has scrolled.

Then Adobe can offer tracking and analytics for video and RIAs to those who want it. If they’re smart, they’ll do it for free for clients that don’t have a lot of traffic, but charge for more volume. It’s a great Trojan Horse strategy, and it’ll work if they open up Omniture’s entire suite.

This goes beyond simple analytics, of course. Adobe is uniquely positioned to track the sharing of viral videos, Flash-based games, and forwarded documents, then to tie those back to conversions on the website. It’s the holy grail of Internet marketing, and it requires that a client be deployed across all browsers and embedded in the applications themselves.

There are some important security and privacy issues here, of course. If you thought tracking cookies were bad, imagine what they’re like when they’re inseparable from the document, video, or application itself.

Nevertheless, everyone making money tracking things — from bit.ly, to Doubleclick, to other analytics firms — is going to be watching this really closely. If you wondered how we were going to pay for online media and digital TV, well, now you know.

Places and tasks

I have a problem with web analytics.

The whole notion of a web visit as a rigid set of steps that users follow is incompatible with how we use the web today. Visitors browse around the site, taking their time, exploring and interacting. Occasionally, they complete some kind of action we want—inviting their friends, buying something, and so on.

For a couple of years, I’ve been thinking about web visits in terms of two fundamental building blocks: Places and tasks. If you look at your site as a series of places and tasks, you’ll think differently about how and what you should be watching.

[Read More]

[Web Analytics] My, How Things Have Changed

I’m currently in the middle of writing the Web Analytics chapter for the book, and my gosh – things have changed so much in fifteen years.  This screenshot is from the program “GetStats”, one of the first web analysis tools to exist.  I ran it using watchingwebsites.com logs (I had to parse them through sed & awk to change their log format to CLF for it to work).  Notice how it took 7 and a half minutes to process as many lines!

I was talking to the author, Kevin Hughes about GetStats and the state of web analytics when he first wrote it.  “Actually getstats wasn’t the first Web server log analysis tool, but it was very influential in terms of the way the data was presented and summarized.  Roy Fielding with wwwstat was the first as far as I can recall to present statistics in an easy-to-read paragraph summary form, that I think was written in Perl.  I also took ideas from Thomas Boutell (wusage) and Eric Katz (WebReport).

Web analytics tools began by telling us how many hits we had on the site, but that doesn’t do much today to tell us what’s really happening with our sites.  The tools went through many evolutions before they got to where we are today – simple metrics, a few KPIs and actionable information.  I’ll touch a bit on this in the book; we’ll also cover implementation methods, advantages, limitations and deployment impact of web analytics tools.

The book is days away from having a completed 1st draft.  I can’t wait to send the complete manuscript out to the reviewers!