Why I need to write a book called metrics for designers

I will not be the first to comment on Doug Bowman’s post about his departure from Google, nor will i likely be the last. But the fact that so many of us feel inclined to say something regarding his experience with his now former employer, makes me question and at the same time acknowledge that perhaps the are many Google’s out there.

When i first read his post, I was pretty much swept off my feet. He was describing in a very accurate way, something that has become my current environment at my job. And it struck me that I was not only stupid for thinking it was only my company that worked like Google does, but that the feeling its designers have about the methods of getting things done were being reciprocated if not mimicked by me.

I have no complains about my job, but I also know the realities of being a designer in an engineering environment. I was speaking not too long ago with a UI Engineering Manager at Apple, and i was telling him about how hard it is to get a design point across some times, and he responded quickly without hesitation, “welcome to silicon Valley”. This comment hit me across the face like some kind of giant slap. Had i been so naive to not see it coming, or was i really facing a completely unique development process for the first time? It turns out it was a little of both, and to my incredible surprise, i was completely immersed in the roots and depths of what a metric driven environment is all about.

Less than 4 months ago, i wrote a post about how metric driven design was the elephant in the room designers struggled to acknowledge, and it seems now almost funny, because back then i thought it was my inexperience and my newness to the idea of testing that made me think that it all felt like a very unnatural component to be placing in design. I thought i was in a bubble of idealism floating in a sea of delusion, and that soon enough the system would ware me down to the point of simply taking the numbers in stride and becoming a vocal supporter of them.

Four months later I am neither a hater nor a follower. Mr Bowman’s eloquent writings of the ill fate of metric driven design is not without merit, but he too acknowledged the strength behind the numbers. Google like my own company have managed incredible amounts of success as a result of being able to use numbers to focus the resources of the company to develop ideas that will most likely work. Innovation while being a novel concept, is but a byproduct of being able to deliver on the improvement in performance of things that have already been produced, and refining them till they reach that sweet spot where no more can be done or a newer idea with better results can replace it. And this is the safe path, the path of least resistance, the one that lets you get from point A to point B because you’ve measured the possible outcomes of every option between c-z and you are left but with one choice. Its the tendency to lean rather than lunge forward, in the hopes of not investing too much effort or focus and get enough of a signal to then cast yourself forward with full strength. I equate it to Indiana Jones and the last crusade; engineers need to see the path before they can build for it, weather that’s by tossing sand on an invisible rock bridge or by testing the hell out of a feature on a site, designers are more likely to take that first step as a blind leap of faith and hope to find something to hold them up at the end of that initial stride. Mind you the key factor is that both start with the same amount of information, but only one is utilizing the gut feeling that Mr. Bowman speaks off. It is almost part of the designers process to act this way, because its this instinct that generates that human component that allows people to connect to something designed.

I believe strongly that any sort of product development can’t be devoid of either approach, but it seems far more of a greater challenge to convince people to spend the time and effort on ideas that are less safe and less predictable, and this is no surprise at all. It just seems like a very impossible task to ask designer to do what they do best while also trying to apeace the non designers and simultaneously justify why taking the risk matters in the end.

I guess i should stop trying to re write what he wrote, after all he probably explained it better than i could ever dare to try, but amidst all the conversations back and forth, it seem inevitable to think that designers should ultimately learn to accept the numbers fully, it seems almost like a betrayal of the core factors that makes us creative. I don’t think for a second that metrics are a bad thing, but i am also smart enough to know that there needs to be a line clearly drawn in the sand, that separates or at least clarifies how much of these numbers are to be trusted as suggestions or advice and how much of it is gospel. I for one think no number should ever be the reason why a design decision is done, as its far to isolated to convey what the “best ” thing for any site could be, metrics tend to be so focused that one number could help one aspect of the site, but is that really tackling the overall user experience if that one factor is looked at so closely.

A metric should help reinforce the practicality of an idea not, validate it, it should help look at its progress but not dictate its direction.

I have come to understand the need for numbers, and i have learnt from many engineers the need to keep these present in our lives, but I’m not about to relinquish the fight of what it means to design for the best user experience as a whole, and i think this is where metrics can be most damaging if let loose, as they make us loose focus and the ability to see the all to often forgotten bigger picture.