“Data” is not the plural of “anecdote.” (Brian Clegg)

As a qualitative researcher, I have to be really careful here. On the one hand, I do not have the data that the web analytics folks and the A/B testing guys and the data scientists have. At the same time, though, I’ve got a lot more than just anecdotes.

First, I do have some numbers. No, they’re not thousands and thousands, or even hundreds and hundreds, but they are more than your own personal data point, or what you heard from a friend or family member or your hair stylist. 

Second, what I have is indeed data. In particular, it is not just opinion. Now, that’s not to say I don’t have respect for your opinion. It might, indeed, be a very well-informed opinion. But it’s still an opinion.

What I can bring to the table is actual behavior, as well as verbalizations of thought processes. And with these coming from real, live users, trying to complete genuine, honest-to-goodness tasks. So, very different from some management types taking a gander at a screen in a conference room and wondering if the “green pops enough” or thinking that “the content doesn’t really speak to me.”

Third, the data that I have tends to be very rich. And that is, indeed, why I feel I don’t necessarily need hundreds and hundreds of data points. 

Now, if what I was trying to get at was more marketing related (How many people would be interested in this product? How much are they willing to pay? Which of these ads are people more likely to click on?), I’d be all about the numbers. What I’m dealing with, though, is whether something’s going to fly or not. 

And you don’t need a lot of numbers for that. You simply don’t need a ton of people to tell you that they have no clue what to put in field X, or that that description makes no sense, or that the CTA is impossible to find – when they actually show and tell you that themselves. If you didn’t have those rich verbalizations though, you might never figure out why you’re getting such poor data, or why no seems to get beyond screen X, or why people seem to be signing up for the wrong product.

Sure, you can speculate on why these things are happening and what the numbers might actually mean. Usability testing, though, will get you the real answers, giving you that very important why along with the how much.

Brian Clegg is an English writer specializing in explaining abstruse science to lay folks