Are You Using AI In Your Job?

We want to understand the real-world applications of AL and ML in business and the impact it will have on all our jobs.

Want to help? Complete the survey, your insights could make a big difference. It will just take one minute.
You'll be the first to get access to the final report.-->

Different App Metrics to Help Understand User Behavior

| October 6, 2021

User behavior is one of the key tell-tale signs that your app is or isn’t doing what it’s supposed to – beyond that, app metrics provide granular details that help businesses understand why it excels, meets expectations, or falls short. Every metric from where the user begins their journey to where they end up and all of the things they do in between paint a real-world picture of how your app actually works.

Your app will generate a ton of data but as each app’s goals are different, measuring and understanding metrics will vary between every service. With that said, there are some common key performance indicators (KPIs) that are useful to most apps. So, let’s take a look and explain the most important metrics for measuring your performance and look at a few tools to collect and analyze this data.

Most useful KPIs for understanding user behavior

Before we move on to discuss the different metrics that businesses use to understand their users, one point to keep in mind is that there are few absolutes to most of these values. For each lifecycle development phase, each app will produce different values throughout time where the statistical significance will variably shift as well. As such, you won’t find many hard numbers throughout the following but instead, we’ll cover a handful of simple guidelines to help you identify and make sense of the numbers your app generates.

Active users 

Perhaps the best place to start is with the number of users engaged with your app over different timeframes. Most analytics systems revolve around the daily active user (DAU) and monthly active users (MAU) KPIs, both of which are usually calculated as a rolling average over a specified timeframe (e.g., twelve weeks, six months, a year, etc.) These 

Excluding apps that are only meant for a specific, controlled group of users like a POS app used by a restaurant franchise (such as one of the apps we built for Cirque Coffee) or some other internal business app, increasing the number of users on your platform is one goal that virtually all apps share. It’s a straightforward indicator that should ideally grow over time.

While it can be helpful to check these values at any given time, rolling averages plotted over different periods paint a better picture of performance. It’s important to understand that growth isn’t always consistent, periods of decline are common, and growth will almost certainly plateau at various points in a product’s lifecycle. From a calculus lens, math nerds would consider it wholly non-linear as composites of large timeframes are almost always piecewise in nature. 

In other words, both your MAU and DAU should look like a stock chart with stable growth if they’re doing well. 

Upward trends here reveal that more people are using your platform – while it’s not mutually exclusive with revenue growth, adding more users to your platform translates to more potential which tends to please stakeholders and increase attractiveness to prospective investors.

Retention & churn

An app’s MAU and DAU are frequently called upon in many of the functions and algorithms used to interpret performance and calculate retention (i.e., those who both download AND use your app) by revealing the number of users who are, in fact, interacting with your platform. Retention is a significant KPI as many businesses have demonstrated over the years that a loyal customer base is the main ingredient to long-term, reliable growth.

The other factor in “counting your users” so to speak is the stickiness of an app or the churn (sometimes called the abandonment rate) which is independent of other values, even though it seems like it’s just the sum of all users not retained. While this is somewhat true, churn is a separate calculation that conveys a unique metric. Most will look at apps at various intervals to see who is still using an app – right after an app is released, most will assess their user base every few days to see which users are still engaged and who has uninstalled or abandoned the app.

For example, let’s say you land 2,000 users your first day like the product we developed for the music discovery app, bopdrop. A week later, you’ve managed to accumulate 4,000 active users which looks great however, bear in mind that the devil is in the details. 

If these 4,000 users were comprised of say, 1,800 of your existing users plus 2,200 new users, this would be excellent churn. But, if said 4,000 users were comprised of only 200 of your original users, this could be problematic as the churn is quite high. Even though you have technically still doubled your userbase, you’re not retaining many users just yet. This can be common in an app’s early days, especially when you do a great job of building buzz, but over time you should ultimately see churn lower as your user base grows.

Just as you need to understand what draws people to your product, seeing what pushes them away is equally important (and in some cases, more important) and thus should serve as a call to action. It’s important to recognize that loyalty is valuable as returning users will generate the most revenue and bring the most stability to your growth in the long run.

Now that that’s out of the way, the next several items should be much simpler. 

Actual demographics of your users

When you first set out to build an app, market research will prove valuable in gauging the interest of prospective users in the market. It’s a great way to get a read on what the actual demographics of your userbase will be versus what you imagine.

However, once an app goes live, the actual userbase can end up being slightly different than the picture painted by your research. It’s a bad thing when demographics don’t align and you miss your marks but it can prove to be a blessing in other scenarios. For example, the language learning app, Duolingo, has become incredibly popular in India.

This knowledge should ultimately be applied or at least strongly considered when designing future iterations. Just because other demographics are using your app doesn’t mean it’s an ideal experience for them so by designing to better accommodate their needs as well as your original target audience, you should see better performance with your “bonus” demographics, much like we see with Duolingo.

Session times, click-through rates & drop-offs

The amount of time a user spends logged in and completing various actions is the easiest way to interpret how large numbers of users are engaging with your platform. Along with metrics like page views, a composite of all users’ session times details if they’re engaging with a part of your app as they should be.

Here, longer isn’t always better as session time goals will depend on the amount and type of content or functions on a page. For example, if users are spending a couple of minutes reading a page with 500 words of content, that’s good. However, users spending 10 minutes to complete a form that only takes a couple of minutes would indicate an issue.

Knowing where traffic is originating helps you understand the click-through rate which shows where users are coming from. For example, if an app relies on timely notifications to alert users to take some kind of action, measuring and analyzing the click-through rate will help reveal what works and what needs attention.

Further knowing where users fall off or uninstall the app helps understand the drop-off rate which can stem from any number of reasons. You might have a transient app-breaking bug that’s weaseled its way past QA and stops users in their tracks. Or, they might just not like your app. In any case, seeing and addressing issues surrounding areas where users commonly drop off will help keep users corralled within the confines of your app.

Touch heatmaps

The areas touched on a screen and the gestures performed (e.g., a two-finger swipe, pinching, etc.) are compiled into touch heatmaps that provide granular details about how and where a user physically touches their devices. This is one area where you’ll see a ton of variation, especially in Android as every manufacturer offers a variety of different form factors.

Perhaps the most obvious examples are the differences observed between regular smartphones and tablets or iPads – for example, you won’t typically see one-handed operation of a tablet as you’ll see with a smartphone where many of us will simply use our favorite thumb to navigate an app. Moreover, you won’t see two-handed typing on a smartphone as you’ll see on larger devices. Beyond that, there are giant touchscreens like Wacoms and other large form factor devices (mainly used by businesses) which are used differently than that of their smaller counterparts.

Keep in mind, just because you can easily get the layout to look the same and “fit” every different device doesn’t mean it will have the same “feel” across the board. Touch heatmaps can reveal opportunities to make both widespread changes for the better as well as to get specific insight for different devices which can shed light on opportunities to prove a more comfortable experience for users.

User feedback

While not a metric, per se, user feedback is useful to help understand sentiment and frustrations directly from users. Not every review or piece of direct feedback will be useful but it will help serve as a safety net for sneakier issues that you can’t quite discern from data alone. Just like with your social presence, make sure to interact with users on the app stores, both the good and the bad. Some apps are designed to recognize the shake gesture which brings up a convenient feedback form for users to leave comments which can prove helpful in helping your team identify bugs and other undesirable quirks.

Tools to help you analyze your apps

As you can probably guess, the following tools are used to capture and analyze this data in your app. We’ve created a list to give you an idea of what to look for when seeking a tool for app analytics that best fits your needs.

Firebase/Google Analytics

Today, many apps are relying on better backends like those provided by Google’s Firebase. Technically, it’s an entire backend-as-a-service (BaaS) with a unique, NoSQL database at its core. Long story short, it’s better a solution than most relational databases in a growing number of applications because of how it stores and synchronizes data in JSON, essentially giving way to fluid data synchronization between users in real-time. It uses Google Analytics and a system called Firebase Crashlytics to help you better understand your users by giving you insight into the metrics above.

App Analytics from Apple

The cleverly named App Analytics solution from Apple is exactly what you would expect – it’s a quality first-party product for measuring and analyzing app data and metrics on iOS. This is perhaps the most accessible and least expensive solution for Apple platforms however, businesses offering a cross-platform service will either want to consider selecting a service that can track both mobile platforms or spend some time consolidating data from another compatible platform in a unified view. 


The Mixpanel platform is one of the most versatile and popular solutions on the market. It offers a ton of customization that allows users to tailor certain aspects like dashboards and alerts to make pertinent data as visible as possible. Mixpanel also offers some additional tools like an active user community where you can learn new tricks or share your own advice.


Though an outlier with respect to the metrics and products we covered above, it’s important to realize that useful data exists outside the app itself, such as the data and feedback from users, prospects, and even detractors on social channels. Our social media team uses Iconosquare to consolidate data from social media channels like Facebook, Twitter, Instagram, and LinkedIn to understand performance and engagement. In some cases, these tools can help expose the potential for some virility however, it’s a good way to help understand costs and simply get a broader view of your audience.

We build to help businesses make the most of their user data

It might sound hokey, but knowledge really is power. By taking direct cues from your users and understanding all the data in between the lines, you’ll be better equipped to grow and succeed. We lean heavily on different analytics systems for the various products we design and development. For more information or to discuss enlisting us to help with your product, get in touch.

+ posts