IA vs. marketing

From my experience it's a lot easier to justify IA to business people (aka decision makers) if you show them how it enables their business model. Start from understanding their marketing objectives, define success metrics, and explain how IA will help reach these goals.

Since acquisition costs are usually quite high, it's paramount to the profitability of most sites to retain users. This translates into metrics such as visits / period / unique user (ie. is the visitor coming back, if yes, how often). To use a rant familiar to Peter, adaptive sites increase the value they deliver to users each time they come back (through recommendations, wish lists, ...). So your IA loops back into your business model (more visits and sales per user, lower acquisition costs when balanced against the lifetime value of each customer).

The idea is to relate IA to what business types are trying to do:

- start from business strategy
- then define its tactical marketing implementation
- then flesh it out with adapted IA

At the very least, once you explain, most people understand they don't want to shoot themselves in the foot and undermine their grand strategic plans with a lousy site (which leads to IA but also quality control, user tests, ...). User scenarios seem a lot less like fancy methodology to "outsiders" once you demonstrate how they enable cross-sell, upsell, user recommending the site to their friends, etc.

IMHO The main value proposition should be supported by best practices. I think the industry has been playing a little bit too much on fear and failure. I like what Mark Hurst does, but is goodexperience.com the proper name, considering how much finger pointing he's done? Jakob Nielsen didn't help either to cast IA and usability in a positive light, however true it is that there are many crappy sites out there.

I wonder how you can be an information architect if you can't relate to marketing (unless you want to work for academic or non-profit sites, but even these have a some kind of "sales pitch" that should translate into the IA). From my POV, an information architect should be able to drill down into an business model spreadsheet, point to some underlying hypotheses (eg. transformation, retention, churn rates), and relate all IA work to them. Ideally, one should be able to point inconsistencies or unrealistic numbers (ie. "from my experience, this number of visits per month is too high, here's a more realistic goal we can set and here's how the site organization will support that goal".)

That doesn't mean everyone should be an proclaimed expert in everything, but site owners are not in the information architecture business. If they're in the widget industry, they want support from vendors and service providers to helps them sell more. IA, or for that matter IT or advertising, are means to doing business, not ends by themselves.

Olivier Travers
http://webvoice.blogspot.com