Lies, Damn Lies, and Analytics

We have a huge crush on data. We don’t just like numbers, we like like numbers. A lot. Data can help us assess the value of advertising tactics, build holistic strategies, and measure results—what’s not to like?

Oh, right. Misleading manipulation of data.
—We don’t like that.

 

There are a lot of ways that data can be misused in advertising, but you don’t have all day, so we’re going to keep this brief and focus on three that really grind our gears: use of inappropriate metrics, unequal comparisons, and ignoring the big picture.

Digital advertising offers a ton of information. Make that several tons. In general, being able to capture that information and use it in a responsible manner is very helpful for brands and advertisers, but with that bounty comes a burden. Specifically, what are the right numbers to focus on? Depending on the analytics platform you’re looking at, you may have access to very basic or very sophisticated metrics. Let’s look at Facebook’s advertising analytics as an example. Currently, Business Manager allows users to track over 200 metrics in 6 broad categories. It would be easy (and tempting) to share a report that includes all of that information, but it would be overwhelming and not all that helpful. Without relevant knowledge and experience (or a deep, unwavering love of the data) a lot of those metrics will be meaningless. So we don’t do that. Instead, we synthesize and analyze the data and present it with our best recommendations on how to proceed and improve.

Currently, Business Manager allows users to track over 200 metrics in 6 broad categories. It would be easy (and tempting) to share a report that includes all of that information, but it would be overwhelming and not all that helpful.

Remember when we said that metrics can vary from one analytics platform to the next? That’s another way data can be abused to bamboozle. Google Ads differentiates between impressions (your ad appeared onscreen while somebody was browsing) and viewable impressions (at least 50% of your ad was shown on screen for one second or longer). Isn’t that tricky?  When you look at those numbers side by side, you’ll usually see a significant difference between them. And if you’re gathering data from multiple sources to report on a holistic campaign, it’s crucial to make sure you know how different platforms define their metrics in order to make fair comparisons.

So far, we’ve been talking about data in the digital world. Now, we love digital for a lot of reasons, but we know that it doesn’t exist in isolation. Even if your campaign is 100% digital, your audience isn’t. Your audience is a group of people who live in the physical world and are influenced by their friends, family, environment, and a whole pile of other things. Don’t isolate your digital data from the rest of reality. Before you give yourself credit for all the brilliant results or blame yourself for all the disappointing shortcomings, take stock of what else was going on in that great big wide world out there. Your brilliant email campaign may have convinced a lot of people to buy your umbrellas in Q2, but before you assume that, take a look at how much it rained.

So, what’s the takeaway? Be thoughtful, transparent, and realistic where your analytics are concerned. Because if you like like data like we do, you’ll treat your data right.