In my last blog post, I talked about how important it is to listen—whether it’s cues from a friend that we’re not interested in talking about the same things, or signs from a customer that says since I’m a guy, I’m not really interested in shopping for bikinis. But what’s next? Now that you’re listening, you need to know what info’s actually worth acting on.

Since I’m a big sports guy, I’ll use a baseball analogy to explain what I mean. The other day, I was listening to sports talk radio and the hosts were talking heavy about betting. One big talker was saying things like, “He hits .300 when there’s men on second and third, and it’s a cloudy day, in a field with an even number of letters, and they’re wearing white uniforms.”

All I could think was, “Wow, you really want me to make a bet based on that?” You know what I care about? The fact that this pitcher throws high fastballs and this guy likes to swing at high fastballs. Now that’s information I can really do something with.

This is what some call Analytics Theater. It’s a term that means while some facts (or data) seem pretty interesting, they’re merely just wild-ass coincidences. Lots of dramatic appeal, but nothing really rooted in reality.

So what’s the reality? Sometimes, the data that really matters is rather mundane. The fact that a batter has a propensity for swinging at high fastballs isn’t nearly as dramatic as when the stars align for that seemingly ideal weather/uniform/stadium name combo, but exciting doesn’t necessarily equate to actionable.

Gathering data is important, but it’s crucial to be able to differentiate between data that’s merely interesting and data that actually has leverage. Why? Leveragable data means something—you can do something with it, it can be acted upon. Interesting data? That’s merely the stuff the sports pundits fill air time with.

And that’s what happens so often in marketing. We obtain all these little obscure findings, and we waste time and effort figuring out what to do with them. But just like everything in life, there are priorities. And to be successful, you need to be able to differentiate the facts that matter from the facts that are merely fun to talk about—and then take action only on those that can make a real difference.

So how do you start looking for useful, not interesting, data?

Start with customer behavior and work from there. Ask what behaviors you want to change and then what you can do differently to make those changes happen. Try to change the behavior that matters. If you’re a retail website, that’s conversion. If you’re a publisher, that’s content consumption. And if you’re a travel website, that’s bookings.

After all, if it’s data you can’t act on or optimize against, then what’s the point? Those little nuggets are fun to learn about, but they won’t improve the customer experience, customer satisfaction, or your bottom line.