The phrase “data-driven” has been thrown around quite a lot over the past several years. Google famously made tweaks so small to their search page that most users never noticed them (like changing the font size of a single word from 12pt to 12.3pt), yet large enough to drive incremental revenue increases. Apple, just as famously, partially ignored data when creating their products, relying for many years on Steve Jobs’ intuition to drive product direction.
So which approach is correct? The answer, of course, is a mixture of the two. Intuition gets a bad reputation, because many think of it as data’s opposite, a sort of witchcraft in which one spirits a divine solution. Intuition, in reality, is only made possible by the sort of data that even Google’s engineers have a hard time replicating, the data one collects through years of life experiences processed by the (yet) unrivaled human brain.
I put a lot of stock in intuition. The strongest ideas often come from thinking creatively about a problem and intuiting a solution. Let’s say I’m creating a signup flow for a website that lets users track their craft beer collection. A pure data approach might lead me to design as plain an experience as I can imagine, then test into an optimized funnel with hundreds or thousands of tweaks over months of testing (assuming I have enough user flow to drive statistically-significant results).
A better approach would be to take my best guess at what works. I’ll allow my intuition to lead me to a solution that will almost certainly outperform a purely data-driven methodology. But what factors guide my intuition? Really, the factors are actually data:
This is all a long-winded way of saying that intuition is, in its own way, data-driven. It provides a strong starting point, and a product owner shouldn’t be afraid to trust her or his gut. Once the signup flow is live, then it’s time to deep-dive into where the users clicks, who’s dropping off where, and all the other fun stats that’ll help optimize the conversion rate over time.
