Design at the Board Level

Thursday, March 18, 2010

Design at the Board Level

Whether you’re a VC, independent board member, or member of senior management, one subject that’s particularly tricky to discuss at the Board level is interface design.

Some board members will say, “I don’t like X about the site’s design, and we should change it to Y.” That’s a fair comment to make; however, it should be taken for exactly what it is: a N=1 sample size of user feedback. But, because the single user in this case is a board member, this one opinion can have a significantly out-sized effect on design decisions.

The same is true for the children of VCs. The NYTimes had a great article on this subject a few years ago, about how VCs’ children are the taste-makers of future companies because they are the opinion panel sample of “what kids today enjoy.” Again, if the kids of a board member don’t like a particular design, this is just a qualitative poll with sample size N=2 (or 3, depending the number of kids) and shouldn’t be treated as anything other than that.

The danger of designing from the Board room is that you end up with a product that’s optimized for the Board members instead of your target audience. This is especially dangerous for products that are designed for mass-consumer appeal. Board feedback represents a tiny fraction of the population, and is fraught with sample bias in terms of socio-economic status.

Qualitative and quantitative polling are terrific ways to get the feedback necessary to fuel an iterative design process. Pretty much everyone company I have ever met should do more of it. But, increase the sample size and make sure your sample accurately represents your target audience.

Jon Steinberg and I were talking about the subject of designing from the Board room as we shared a cab back from JFK on Tuesday. Jon quoted an excellent remedy (and I’m hoping he’ll comment with proper attribution for the original source of the idea, because I forget who he was quoting — I think maybe it was from Disney): “You can only use yourself as a data point once in a conversation about design.”

This is a great remedy because sometimes there is often good reason to speak about your own experience. But, a far stronger point can be made if you’re speaking from research created with sound methodology. If you only get one shot per conversation to speak about product design from your own experience, then you’re going to make sure it the best possible comment you can make. And then, the rest of the time everyone can focus solely on the needs of the target audience.

Disclaimer: I am absolutely guilty of volunteering my (unsolicited) opinion on product design regularly to management teams, especially for services where I consider myself a target user. I definitely do it more than once per conversation (though I’ll try to restrain myself in the future). I hope any team on the receiving end of this advice takes my opinion as one user’s feedback and nothing more.