User analytics are critical to understanding the success of your solutions and where they can be improved. Not all analytics are the same, and what you measure for one product may not apply to your others. Let us consider a few ways to make analytics work for your needs.
TRACK THE RIGHT INFORMATION
It’s common to think that pageviews or time-in-product are the most relevant stats to track, and while they are a good baseline, they are not the sole measure of success. Conversions are probably more important, as it's critical to know whether the value proposition you make is being received by your users or customers. This could be downloads based on viewing the app listing, sign-ups for your service after seeing a splash screen or if add-on products and services are being selected.
It is also important to look at how long your users have been a part of your service. A new user might have different preferences than a long-time user, so you may want to present different options at first. For example, the first time someone uses an app to order food, offer the most popular items right away and let them have a quick success. When they return, they might be more interested in exploring the entire menu since they enjoyed what you offered them the first time. By also tracking which users are most active, you’re finding your biggest supporters and the ones who will provide the biggest ROI. You’ll want to keep an eye on these users to see if you notice any trends or new behaviors.
BECOME A MASTER OF INTERPRETATION
When you’re looking at your analytics, you have to be careful to interpret the results properly. Some stats are easy, such as monthly active users, where higher is always better. The total time spent on your service or the number of page views can be good, too, especially if your product is ad-supported or correlates heavily with customer satisfaction.
However, more time spent ordering could mean that users are confused by the interface and less likely to return. Sometimes you want users to complete tasks more quickly, which could look like a problem if you only look at total time in your product. Now, if users who spend a long time ordering also order more (a correlation I’ve seen in some apps we’ve worked on), you’ve found an optimized solution.
It's also important to not look at your analytics in a bubble. Are the support channels on your website being used more than last year? This could be an indication of a problem getting worse and people needing to contact support more. But it could also be that more users are finding success with online support over their traditional phone support options.
If phone calls have dropped significantly, maybe you’re seeing a net improvement and are able to handle your support needs more efficiently. As new communication channels open up through mobile, voice, chatbots and social media, existing solutions may see significant drops. Very few people mail letters to express their feedback anymore, even though this used to be common!
USE YOUR METRICS CORRECTLY
Now that you’re tracking the right moments and interpreting the data properly, what do you do with the information you’ve gathered? In general, you want to make the actions that users enjoy easier to find and quicker to accomplish. Are there rarely used parts of the experience that could be removed? If there isn’t enough ROI for some features, it’s probably worth removing them in the long run. For example, some loyalty and ordering apps also include mini-games under the guise of being an entertainment option for kids. However, these features are almost never used over other forms of entertainment on the device and not worth the investment to keep updating them to support new OS and hardware revisions.
Conversely, you need to also look at where users are spending their time and lean into making the experience as satisfying as possible. Do your customers like exploring all the different options? If you have an e-commerce site and show three related products on your product pages but your analytics show that a lot of customers explore these options, perhaps you should try showing four or five. Another consideration is how the platform might affect the experience. Do desktop users perform more complex actions, whereas mobile users are more focused on simple tasks? Tailoring the experience to each user could have huge returns on satisfaction.
In the end, business goals have to be reconciled with user desires. Analytics measure what users are doing, but the interpretations and actions taken have to be thoughtfully considered. If the business goals can be distilled to their core enough, like “increase orders,” it’s easier to put the right lens on the data and take actions that help drive the right behavior or remove parts of the experience that cause friction. Consider all your options, but make decisions based on your data.