Apple-picking season is over in the Northeast, but no matter where you are, you can improve analytics before the end of the year by working on just a handful of common failure points.
Some of them may seem obvious, but in my experience, the number of organizations leaving this fruit unpicked is greater than those that fill their baskets and have apple pie for dessert.
And it’s more important than that.
Because while we can call these items low-hanging, we can also call them foundational. Leaving them unaddressed exposes you needlessly to more long-term analytics ills, and possible failure. Pushing the metaphor one more step: does an apple a day keep the doctor away? Maybe. But taking care of the following four items will certainly bring noticeable improvements in analytics.
- Audit. Do you have trouble trusting your data? This is one of the most common signs of a misaligned analytics architecture. Numbers “don’t make sense.” They drop off abruptly. They gyrate inexplicably. No one relies on them to make decisions. Fixing this problem is fairly straightforward. Perform an audit of your tags. Create a list of tags and test each one to see if it returned an expected result. In a large percentage of cases, bad numbers are the result of bad tagging. Once you find the trouble, fixing them is usually a simple matter.
- Kill zombie profiles. Often, multiple administrators are, at one time or another, “in charge” of analytics. Worse, no one really is; and the result is that numerous well-meaning marketers and business users decide unilaterally that they need to configure their own set of reports. Often these are grouped together in what may be called “profiles” depending on what tool you have. A typical scenario has the profile in use for one or two viewings, then being abandoned forever. The result is a long list of undead profiles that suck bandwidth without providing anything useful. No one wants to look at them, but no one is brave enough to kill them either, because they’re afraid they might break something else, or that someone somewhere must be using them. But good governance would create a process for killing zombie profiles. Ask around if anyone is using the profile(s), and remember to ask why (“because I like to” is not good enough). If the profile is not being used and for a good reason, destroy it now before it sucks more bandwidth and causes further confusion. Then you’ll have a clean, new, more useful profile.
- Be who you are. Self-definition is hard in any walk of life and it’s not much different in analytics. Do you know what kind of data will help you improve business? More traffic? You need to know this trend. But if you go even one step beyond this, you’ll discover some easy ways to align measurement with the type of digital business you have. For instance, do you sell advertising on your site? Then overall traffic to your premium pages is probably a very important metric to understand. Do you perform inbound marketing? Then you need to know abandonment and throughput rates on key conversion pages (landing pages or calls-to-action, for instance). It will pay big dividends to ask what you really would fix if you knew it was broken, and perhaps simplify your reporting to focus on that. Sure, your tool can drill down to the core of the earth, but you just need a few apples for that pie. It’s almost Thanksgiving, and company is coming!
- Get trained up. Simply put: how well do you know how to use your digital analytics tool? How about the other marketers on your team? Let’s say you were in Google Analytics and you needed to find out a particular value for a particular parameter relating to an affiliate program. Quick, where do you find that? The typical interface on an analytics tool is fairly complex, offering lots of information and choice. But you need to know how to navigate to this one metric. Even if an expert created it for you, you may still need to know how to find it, share it, and understand it. Is it in a drilldown of a drilldown and across at the seventh column of the table? Would you know how to find it right now? Get trained in the tool. Online courses are OK and general classes are better, but the best and most targeted training is achieved when a trainer can come onsite and work with a small group of need-to-know users, hands-on, with your data and your configuration. Sometimes wishes do come true.
No doubt there are more easy achievements than I have listed. I’ve tried to show some that seem both common and not very difficult to solve. If you’ve already solved these problems, no doubt there are higher rungs to climb and juicier fruits farther up the tree. But my guess is you have at least one of these apples right in front you today. Enjoy!
Ripe Lemon image on home page via Shutterstock.
Emily Ma, product director of Tencent’s advertising platform products department, was a keynote speaker at ClickZ Live Shanghai where she discussed the ... read more
The terms that customers type into your site search function can help you to gain an understanding of user behaviour and can be used to optimise ... read more
Google Analytics comes with lots of standard reports and settings, but with a little customisation you can extract much more value. One way is ... read more