A visit from Omniture...

I met Brett Error, the CTO of Omniture today. I also heard him give a presentation about the company, what they do, and the implications for marketing and 'ebusiness'.

(If you don't know why that's cool then you probably don't know what Omniture does. Therefore, you should look at this summary page and this article. In case you're lazy, here's the short version: Omniture does web analytics (tracking, measuring, testing, reporting, the whole nine yards), and they do it very, very well. How well you ask? Well enough to merit a $10,000 a month (and up, based on traffic) price tag and to have companies like eBay and AOL paying for it.)

So, needless to say, I'm not going to be signing up for Omniture's service anytime soon, but I wish I could. I have seen just a bit of the inside workings of their product, SiteCatalyst, and it is amazing. It allows you to see where your visitors are clicking, where they're coming from, where they are going, where you're losing them, etc., etc. The over-all product is not just fancy software, its a complete, tailored solution. Beautiful.

Now then. What does this mean for me and/or you who can't afford this thing? Well, here's an old quote (probably said by an accountant) that Mr. Error mentioned during his presentation: "If you don't measure it, you can't manage it."

New technology, especially that of the internet, is allowing businesses to take that idea to a whole new level. If Mr. Error (and a lot of internet marketing gurus) is right, this type of measurement and analysis represents the future of marketing. It also represents, at least in part, a turn away from a creative design type of marketing to a more scientific type.

For example: Your web design team says: "I think such and such would make our site better because its sleek, its edgy, and because we're awesome." SiteCatalyst says: "You get more sales when your site is like this than when its like this." Which of those statements would you, as a manager, rather hear? Whose advice are you going to take? Yeah. Somebody is fired.

Moral of the story: Measure and analyze whatever you can. Don't try stuff, test stuff. No more Russian-Roulette-style, "this seems cool" site upgrades.

Comments?