It's been said that "the numbers don't lie" and that may well be true but sometimes the numbers can mislead.
Today’s digital marketers are fortunate in that they have ready, no-cost or low-cost access to a wide range of data analytics that can help them determine what works, what doesn’t and how they could modify their marketing tactics and techniques to get better results. Unfortunately, sometimes the data can be misleading—or, better put, the way we interpret the data can cause us to be misled. Here are some common examples of data interpretation missteps that even digital pros can make.
Many website managers value organic traffic to their sites—traffic that represents people who were looking for something you had to offer but hadn’t previously heard about you or your company and didn’t have your URL. They did a search and that search brought them to one of the pages on your site.
Not all organic traffic is really organic, though. Consider this example:
Your name is Jacarious J. Jarinta and you run a consulting firm called Just Jack’s Results. As you look through your organic search results you’re very likely to see searches like:
Each of these searches suggests that the searcher already knew of you or your company. They weren’t searching “organically” based on general terms describing the types of products or services you offered. These results, therefore, could more accurately be placed in the “direct” traffic bucket to give you a better idea of what’s driving people to your site.
Analytics will report to you the traffic from all sources, or IP addresses, unless you tell the program you use not to. Why would you not want to learn about all traffic? Because not all of that traffic represents actual visitors and potential customers. Depending on the size of your organization and/or your circle of friends and acquaintances, these numbers could be significant. If, for instance, you have a company of 100 people it’s likely that many of those people are visiting your website on a daily basis to find information, or as part of their jobs. If that’s the case, you’ll want to filter out those IP addresses so you don’t have an over-inflated view of your website traffic.
In general, the lower your bounce rate (a measure of the percentage of people that landed on your page, did nothing and left your site), the better. But bounce rates can be misleading at both the high and low end of the spectrum.
High bounce rates, for instance (while generally an indication that your content is capturing the attention of visitors and, consequently, is causing them to leave), aren’t always bad. For instance, if you use landing pages for many of your inbound promotions, it’s highly likely that visitors only stop at that page which they reach after clicking a link on an ad or email. A certain percentage of these may go through with placing an order or making a request on that landing page, but a high percentage likely doesn’t. And, because a landing page generally represents the penultimate stopping point, your bounce rate is likely to be high. The same is true for many blog posts. Visitors may come to a blog post through an organic search, read that post and then leave.
Even low bounce rates, though, may be subject to misinterpretation. While, as we’ve said, low bounce rates are generally good and what most digital marketers are striving for, in some cases a low bounce rate may indicate that something is wrong with the way your analytics are being tracked.
When we design our websites we start with the homepage, viewing that as the entry to the rest of the content we make available. That presumes, though, that people coming to our websites will automatically land on this page first. If they enter the homepage URL, they will. But, they won’t land here, necessarily, through organic search or backlinks.
The bottom line. Don’t take your analytics at face value. Think carefully about what the data is telling you—or may not be telling you—before you use it to make important decisions. Numbers don’t lie. But humans often do misinterpret the numbers.