tecosystems

Untapped Intelligence? Mind Your Web Traffic Logs

Share via Twitter Share via Facebook Share via Linkedin Share via Reddit

When speaking with those who don’t know us that well – and occasionally those that do – one of the questions we hear most often centers around enterprises. Specifically, to what degree we can – or cannot – speak for them, about their interests, their adoption, their buys. The answer is, as so many things are these days, complicated. It is not, however, a surprising one given a.) our predominantly vendor customer base and b.) our focus on technology adoption rather than buying trends.

But while we’ll happily acknowledge that we spend more of our time with developers, architects, and engineers than their IT manager or CIO bosses – by design – we’re usually quick to note that we also spend a significant portion of our time on the line with customers of all shapes and sizes. These conversations can and often do blur the lines between briefing and consultation, but they’re of significant value to us.

As are, increasingly, the implicit dialogue enterprises have with our site, as observed by our traffic logs. For all that we’re at a disadvantage to some of the larger firms with large enterprise customer bases, when it comes to web intelligence we have an inherent advantage: open content. I see far more web queries from enterprises, governments and educational institutions across the globe on a daily basis than any analyst could possibly service. It is, after all, always going to be easier to ask questions of Google than it is of the analyst firms.

I mentioned last week that someone from a large investment bank in New York was looking for information about Solaris package management. Interesting datapoint, I think. As is the fact that one of the largest healthcare providers in the US showed up looking for information on compliance and service oriented architectures. Or the query from a law office concerning WordPerfect to ODF migration. Ok, maybe the one from the defense contractor on corrupt iTunes libraries is slightly less important. But what about the fairly sizable SI that wants to know about open source economics? Or the state governmental agency interested in a OneNote equivalent for Linux? Or the problems that the UN is having with Google Analytics? Or the Spanish university interested in mashups and SOA? Or…you get the point, I’m sure. And that’s without getting into what we see from the vendors themselves – you should see the traffic on Q&A’s after a product launch. I’m frankly a bit surprised that organizations don’t anonymize themselves via proxies to a greater degree than they do.

It’s necessary, of course, to consider context when looking through traffic logs. You (generally) don’t know who within a given organization is querying, or why. And the query itself could be offhand, of no general significance. But even so, the datapoints themselves can be of interest, possibly on an individual and certainly on an aggregate basis. My only regret is that the traffic analysis tools themselves are still relatively primitive, delivering little in the way of business intelligence on the available data. Even Google Analytics, which goes further than the likes of Statcounter in that department, does little to provide me the type of intelligence I’m looking for. Mining the data, then, is still a regrettably manual process (yes, I think there’s an interesting reporting/BI software opportunity therein).

But if you doubt the value, just look at your own logs and take a look at what your competitors are searching your site for. In our case at least, it’s often worth a chuckle or two.