In the Web Marketing universe, qualitative data is often seen as the poor cousin of quantitative data. Information that can’t be reduced to a spreadsheet (or better yet, a graph or chart) isn’t trusted. It’s treated with skepticism, even derision.
It’s easy to understand why. Many practitioners have technical-leaning backgrounds, where precision is king. And so, “Data-driven” marketing is code for analytics and testing-driven marketing. (Where “testing” is defined as A/B and multivariate testing, i.e. backed up with statistics. Traditional usability testing just isn’t part of the game plan.)
Well, I’m a fan of data-driven marketing too. But there’s more than one type of data. Can’t we make room for others?
The Case for Parallel Data Collection
I appreciate the value of quantitative data. But it can only take you so far. Though it’s great for identifying problems, when it comes to finding solutions to those problems, it often falls short.
Take a simple example: A review of site analytics reveals that far too many visitors are bouncing from an important page. That’s definitely good to know!
But why are they bouncing? And how do you fix it? Sorry, but your graphs and charts and spreadsheets may not tell you that.
“Run some conversion tests”, you say? Great idea. But how do you decide what changes to test?
To gain the insights needed for solutions, don’t be afraid to look to qualitative data. Things like:
- Interviews with customers. (Yes, actually talk to them.)
- Interviews with CSRs.
- Surveys and other direct customer feedback.
- Most of all, usability tests: Watch customers try to use your site, and see where it lets them down.
“Data-Driven” marketing doesn’t have to mean “spreadsheet-driven” marketing. There’s room for a healthy sprinkling of qualitative data, and (dare I say it?) even creative hunches.