Heavy Coastal Fog with Inland Sun (byline San Francisco)
Predicting the future is mugs game – not for the predictor (who gets a risk-free chance to pontificate) but the reader. Of course, the reader really ought to know better.
Why? Well, weather systems are the classic example of Chaos theory – and we all know how well weather forecasters do – and the real future is chaos theory writ large.
Even in islands of stability where all appears to be predictable (and what good is a prediction when everything is predictable?) the world will play tricks on you – I always think of Steve Martin’s turn as a weatherman in LA – he videotapes his forecast so he can play hooky. Naturally, since we are talking Los Angeles, the forecast is Sun, Sun and more Sun. But a freak storm hits and he ends up fired.
All of which is simply a way of warning you not to take my predictions for the future seriously – because, god knows, I don’t. Nevertheless, I was tagged by Eric Peterson (http://www.webanalyticsdemystified.com/weblog/ ) who was tagged by Avinash Kaushik in another of these "tagging" games invented by Bloggers to, I suppose, give us all something to write about. First, a new year’s resolution - I hereby promise that this will be my first and last "tagging game" post of 2007!
Prediction 1: The practice of SEO will change dramatically. I’ve written elsewhere and in more detail on why I think the current focus on #1 position is evidence of a deeply flawed system (I have an article upcoming in my regular ADOTAS column that goes into a little more depth on this). I don’t know if 2007 will be the year this finally changes, but it has to come soon. And with Ajax tools becoming common, I think the Search Engines will finally deliver sub-settable search results in a user-friendly fashion that will dramatically reduce the value of #1 position. It can’t happen too soon – and this change will make SEO at once less technically demanding, more in tune with other marketing efforts, and much less mysterious.
Prediction 2: Agencies will finally emerge as a significant factor in the web analytics space. Maybe. There are plenty of reasons why interactive agencies haven’t been fully on board with measurement and there are going to be lots who will screw it up pretty royally. But in a way, that won’t matter. Agencies are so well positioned to sell measurement to many clients that they are bound to have an impact on the market – whether they do it well or not. And, at last, they are seriously trying to do it.
Prediction 3: The pendulum will swing back – at least a little way – from ASP solutions to internal solutions for web analytics. Notice that I didn’t say tag-based to log-based. Web Logs can be a mess and were never designed for measurement. So it isn’t unusual to see companies pointing tags back to themselves or implementing special server logs. I don’t except this shift to be all that dramatic – ASP is enormously advantageous for many companies. And companies like Omniture and WebSideStory have done an amazing job handling truly staggering amounts of data. But for organizations that want to integrate web data into larger systems in seamless fashion or have access to additional analytic tools, the ASP process may not always be optimal. Sure, those companies can send you back your data and still provide the interface – but I think there is a distinct and growing market niche for systems that begin and end with the data on the company servers.
Prediction 4: Tags will get universally lighter. There was never much sense in the idea of tagging tons of variables based on page name or URL. All this accomplishes is the addition of complexity and the creation of barriers to change and analysis. Here’s a basic rule tagging systems should live by – no variable should ever have to be created in a tag if the data it captures is represented anywhere else in the system (like in the URL, Referring Site, etc.). There are always going to be variables available only in the session – and these are necessarily passed in the tag. But the days of forcing users to build tags to capture data from parameters in the URL are – I think – ending rapidly. No doubt, widespread adoption of this rule will come just in time for Web 2.0 to make it irrelevant.
Prediction 5: A year from now, when people once again write their predictions, we’ll hear the same laments as always about the lack of productivity in web analytics, the problems with tools not delivering real value and the difficulties in replicating web analytic success. Why? Because while change is a constant some of the essentials are much less tractable. We are all, fundamentally, lazy. Companies are generally productive only where and when they really need to be. No tool will ever transform a data geek into a strategic thinker for the business. And, for most companies, measurement will be the stuff you do to prove that whatever decisions you actually made were right. And this will probably all still be true when we are whooshing around in jet cars and have super-computers in our cell-phones!
All of which makes me think that maybe this predicting the future is a mugs game for the writer as well. I’ve rarely worked harder on a post – and I think it’s because it’s just a lot easier to write about something you actually understand.
Which is why I’ll be going – thankfully – back to tool evaluation in my next post! I'm also hoping (if I can only find the time) to comment on Eric's excellent and very interesting posts on measuring engagement - well worth checking out by the way!
Comments