What Your Analytics Aren't Saying

Not every piece of information matters.

It’s a borderline heresy in today’s data-driven age, but data is only worthwhile insofar as you’re learning something useful from it. The vast proliferation of analytics means that you need to know what to pay attention to, and to know what you aren’t seeing. Sure, I can see how many likes, shares, and pageviews I get. I can track bounce rate and time spent on a page, email opens, email clicks, and unsubscribes. I can see my PageRank, how many impressions my ads are getting, and how many times a form has been filled out.

And that all matters.

But to crib a line from The Incredibles, when everything is measured, nothing is.

You need to cut through the noise and recognize what the limits of your web analytics are. Because they can’t tell you everything.

1. It can’t assess the quality of your content.

It’s really easy to look at pageviews, shares, and bounce rate to determine how successful your content is. But here’s the thing: none of those measurements can tell you if your content did its job. The length of a session, for example; how long is the “right” length to view a webpage? How does the number of shares tell you whether or not someone was pushed further towards a purchasing decision by the content?

Assessing the quality of your content is much more than counting the numbers associated with it, and it’s something that grows over time.

2. It can’t tell you why your content performed the way it did.

Even with numbers that matter – email open rates, for example – analytics can’t provide definitive answers. Was your email sent at a bad time? Was the subject line not compelling enough? Was there a big game on? Was there a power failure? Did you send to the wrong kinds of people? Was your messaging simply off?

What about the content of your emails? If you had a low click rate, it could be for any number of reasons; the email was too long, the link was in a weird place, the messaging didn’t click, the picture was unappealing, it was sent on the back of another email from another company that scratched the same itch, et cetera et cetera. Your analytics platform can’t answer these circumstantial questions that often sounds the death knell of a content piece.

Analytics can show “more effective” versus “less effective,” but it can’t ever demonstrate “good” versus “bad.” That means you need to be willing to keep trying, using the data iteratively over time to build up more and more effective content.

3. It can’t tell you what you should do next.

Your marketing strategy can’t be a plug-and-play formula dependent on specific numerical outcomes. It’s not a number-crunching machine that dictates your next steps – but lots of people want it to be. You wouldn’t believe how often we hear things like “we want to increase our unique sessions by 30% before we launch this marketing initiative” – as if your website’s overall traffic was necessarily the right people for the campaign, as if the numbers alone indicate readiness or strength.

In the same way, simply having a large number of Twitter follows doesn’t mean you have a receptive audience; ignoring the fact that most of your tweets will only be seen by a fraction of your followers, your “audience” might ultimately just be a bunch of randos, following whoever they stumble across. There’s no actionable next step to take with that number.

Your analytics are ultimately descriptive and not prescriptive; they can’t show you the lay of the land in a few usefully-quantifiable ways, but it’s up to you to make sure you take the right numbers and use them in the right way to determine your path forward.

What marketing metrics and analytics do you depend on? Let us know in the comments below!

{{cta('582ea345-ef1e-4885-8b0f-d5d86fedaf1d')}}