Testing our instincts

I have a theory that we place too much importance on instinct in journalism. There’s a good reason for that – back in the print days, we didn’t have much else to go on. Defining what elements of the magazine were and weren’t working was largely a matter of instinct, as there was no clear and easy way of doing on-going market research. The instinct of the editor lived and died by the circulation figures. And there’s a mythology in our trade around the editor’s snap decisions and the instinct of the reporter. 
But we don’t only need to work like that now. Even free analytics tools can give us clear pictures of what is attractive on our site, and what was a waste of our time and money. But again and again when I ask “who/how many/how often” about parts of our site, I’m met by blank looks. We have headline traffic stats, but very little analysis of how people move around, or how much impact a certain box on a homepage or list on an article page is actually having. I was in a meeting last week where two really good, really eye-opening pieces of analysis were presented that materially shifted our thinking around a particular website. I really wish we saw that more often. 
Now, to be clear, I’m not just talking about hits and traffic measurement here – that’s a woefully crude way of measuring success – but figuring out where the busy plazas are on our sites, and where the deserted, neglected back alleys are. What value is that box, or that header area adding? How does the traffic it drives or its usage compare to the development costs or the site speed?  We think stories of that type are important – but if people aren’t reading them, why is that? Are we wrong? Are we telling the stories the wrong way? Have we made then too hard to find? Start with instinct, by all means. But test if it’s right, and if you find that it’s not, figure out why it’s not and learn from that experience. 
I’m not saying that we should do away with instinct (because, honestly, I think instinct is often a better way of launching things than market research). But I do think we should be testing that instinct more rigorously once it is codified into our websites.