App download number is a delusive metric, as it does not reflect whether the application’s objective or key value proposition is being met or not. To figure out whether your app is achieving what it was built for (more revenue, user engagement or brand building), you need to go beyond counting app store downloads. The following post describes 15 metrics that matter the most for any sort of mobile application, why they are important and how to track them.----
Not everything that can be counted counts, and not everything that counts can be counted.~ Albert Einstein
Let’s consider the tale of Super Mario iOS game. It was announced during Apple’s annual conference in 2016, a huge marketing success indeed. Eventually, It was released in december 2016, with sky-high expectations.
The app received 2,850,000 downloads in the first day. In a span of 4 days, number of downloads rose to 40 million, a way faster than Pokemon Go.
Soon it topped app store charts, but only to knock billions off Nintendo’s market cap.
Surprisingly, it received more one star ratings than all other rating combined and an average rating of 2.12. What started as a successful marketing event, turned out to be a massive foray for Nintendo.
And hence, measuring app download number is a fool’s gold. What matters the most is usage and engagement numbers.
The real scenario is, most of the people never look at their app’s KPIs after publishing it on the app store, or they do not know how and what to measure. All that matters to them is the number of downloads, which in turn, keeps them in a state of deceit. Ignoring the performance metrics of an app is like building a ship and letting it sail without setting its direction.
Mentioned below are 15 metrics ( segmented by performance, user demography, user engagement and business attributes), that matter the most for any kind of mobile application’s success.
Performance Metrics:
Performance metrics are the most critical parameters as they define the overall performance of your app. These metrics define how your app is interacting with the hosting device. Speed of your mobile app is in direct correlation with the user’s waiting time, and hence performance metrics deserve a special focus. They primarily depend upon the code quality and architecture of your mobile app, and database and server configuration
1. Time to first byte:
TTFB is the time (in milliseconds) that your mobile app takes to receive the first byte of the response from your server. The ideal response time ranges between 1.9 to 3 seconds. If your app’s TTFB is larger than 3 seconds, there is surely something you need to ponder upon. A few proven techniques to reduce the TTFB are:
Useful Tools: Standard user testing tools, Developer tools in Chrome or Firefox
2. Average screen rendering time:
Once the data is received from the server, the mobile app renders it to show useful information to the user. The average screen rendering time is the time that the application takes to load the content, images, videos and other animation on the screen. It differs for different applications, depending upon their complexity.It may range from 1 second for simple apps to 5 seconds for complex applications. It is in direct correlation to the user’s waiting time; hence a critical metric to optimize. Improper screen dimensions, unscaled and unoptimized images, inconsistent fonts, excessive blocking scripts etc are a few factors that lead to slow rendering time.Take steps to eliminate these possibilities in case you observe longer page rendering time.
Useful Tools: New Relic APM, Rollbar
3. API Latency
Most of the applications these days use several third party APIs or web services in order to show real time information. API latency is the round trip time taken by various APIs to request and receive data. Since, for multiple API requests, the time to respond adds up with every API call. Hence, the apps technical architecture should be robust enough to handle parallelism and concurrency effectively. Also you need to make sure that the API’s architecture is well configured and optimized.An ideal latency time lies between 1 - 3 seconds, depending upon the number of APIs the application is using. http://nordicapis.com/optimizing-apis-for-mobile-apps/
Tools: Burp Suite, mitmproxy
4. App load per period
App load per period is a stress test that specifies how many transactions or events your mobile can handle in a given period of time. It is mostly done by clicking thing in rapid succession in order to overload the app with too much inputs. It is important to know the maximum payload of your application as the users may sometimes try to click a feature multiple times to get it loaded. To increase the app load, make sure your key events like network request or API call are done off the main thread, so that the UI elements do not get clogged. Heavy and time consuming functions should be only called when needed, and in the background.
Tools: New Relic APM, Rollbar
5. App crash Analytics
App crash is defined as abruptly closing of your mobile app during active usage. Frequent app crashes can lead to frustrated users and eventually high app uninstalls. There are multiple reasons behind app crashes like sloppy code, poor memory management, spotty network connection, poor exception handling and inefficient testing. App crashes can be analyzed using Google mobile analytics. You can also identify particular mobile brands, device name, or page where the app tends to crash. App crashes can be reduced with rigorous testing and continuously building test cases while the development process to identify issues right away.
Tools: Crashlytics, Google Analytics, Localytics
Usage Metrics
Usage metrics indicate the demographics of the users or devices that are using your application. They provide you a sneak peek into how your application is performing against a set of users or what sort of users your app appeals to, ie segmented by device, locations, operating systems, age group etc. Use this information to optimize or localize your app and provide a better user experience.
6. Devices/OS and Location metrics
When we say mobile devices, it also includes all its cousins like tablets, IoT devices, and other devices with different screen sizes and operating systems. Digging into the device and OS metrics will provide you an intel around the kind of devices, screen sizes or operating systems your users are using. Diive into the details like location, age group, network carrier etc and gain a trove of information to optimize your app apropos to the demographics.
Useful Tools: Google Analytics, Localytics
7. Active Users
Not all users who have downloaded your app will use it regularly. For some it will be just another app, neglected and waiting to be uninstalled. As per the definition, an active user is the one who has interacted with your app in a specified period of time. Active users are further segmented as daily active users (DAU), monthly active users (MAU), or weekly active users (WAU). Only active uses turn into customers. Hence, thrive to push your MAU down the funnel to DAU, and perhaps you can earn more customers. You can easily get the number of active users though Google Analytics
8. Stickiness
Stickiness is the ratio of DAU to MAU. It is a figure that tell you how often your users come back to use your app. More closer the daily active users to your monthly active users, higher the stickiness. Stickiness varies for different categories of apps; being highest for games and lowest for utility apps. Stickiness is all about the UI/UX and functionality of the application. You can increase the stickiness by using push notifications to users who have not visited your app after first eng
Stickiness = Daily active users/Monthly active user
Engagement metrics:
Engagement metrics provide you a picture of how uses are engaging with your mobile app. In order to build a lasting app, you ought to measure these engagement metrics. They will feed your strategy as in how to engage and raise the interest of potential users, retain them and convert them into customers.
9. Session Length
Session length is the interval between opening and closing of your app by a user. It quantifies the depth of a user’s engagement with your app. Session length varies for different types of applications. A 90 second session might be ideal for taxi booking app, however for a music streaming app that means the user didn't even make it through a Bieber song. Once you figure out the ideal session length for your application, you can identify where the user feels struck
Useful Tools: Google Analytics, Localytics, MixPanel
10. Visit frequency
As the term defines itself, it is the number of times a user has visited your application in a specified time interval. It is important to keep a check on visit frequency during the first week, as it is a plausible indication of what is going to happen in the future. Less frequent visitors are more likely to abandon your app as compared to more frequent visitors. Strive to make your application a part of user’s routine. Contextual notifications, offers, discounts, updates on new features etc. is an effective strategy to boost your visit frequency.
Useful Tools: Google Analytics, Localytics, MixPanel
11. Depth of the Visit
Depth of the visit or average page views per session is the number of app screens a users has interacted with in a session. Generally, more the number of page views, higher is the engagement rate. There is no benchmark for page views per session. A news application may want to keep their visitors engaged and hence have a high APPS, whereas an ecommerce app would want to hook their visitors to one page only and buy something - surely a successful session for such apps.
12. Behaviour flow
Behaviour flow is not a metric, but a complete roadmap of your user’s navigation through your app. It helps you visualize the path your users are taking to move from one screen to another or any specific event. It helps you discover what content keeps your users engaged, what routes they are taking to reach the objective, what road bumps they are facing and where do they fall out. With this data in hand, you can create strategies to re-engage dropped users and implement a clearer funnel. Merging behaviour flow with average screen-view time can provide you insights like how long a particular task takes and how to reduce the conversion time.
User Acquisition Metrics
These metrics indicate the performance of your marketing effort or in simple terms, how users are reaching your app. They highlight the visibility of your app on app stores, which marketing channels are bringing traffic to your app, cost to acquire a user and the revenue your app will generate.
13. Acquisition channels
The most important way to acquire new users is to analyze how existing users have found your app. Users reach out to your app through a number of ways; organic search, paid campaigns, social media, referrals, word of mouth etc. Measure your traffic channels that have contributed in bringing users to your app, most importantly, one the ones which have bought you customers rather than users. 65% of users download an app after searching for something on the appstore. Hence, organic traffic plays a vital role in bringing users to your door. Optimize your app store listing to get a higher rank on app stores. And similarly, revisit your strategy for other channels if required.
Useful Tools: Appsflyer, Appannie, AppSamurai, Google Analytics
14. Cost per user acquisition
Cost per user acquisition is the amount spent to acquire one paying user or a subscriber. To measure the cost per user acquisition, divide the total cost spent on your marketing activities by the number of users it has generated. Most of the investors consider active users and not all users to calculate cost per user acquisition. And hence it is first necessary to figure out out your active users. Cost per install is another important metric to monitor. The average cost per install in the US is $3.59.
Useful Tools: Google Anlaytics, AppSamurai,
15. Customer lifetime Value
Customer lifetime value is the amount of revenue you acquire from a customer before that user abandons your app. It is important to monitor this metric as it will help you figure out that for at least how long you need to retain a user to derive profit from that user. Customer lifetime value should always be greater than cost per user acquisition, or else you won't be able or scale (or even survive longer). There are many ways to calculate CLV, however, the most simplified equation is:
CLV = Average conversion value x Avg number of conversions x average lifetime of customer
Useful Tools: Google Anlaytics, AppSamurai
Concluding Note:
There is a difference in getting your app installed by any means and making your users joyous by the very fact its existence. In the long term, the latter is critically important. And otherwise, if the download numbers are not large as you expected them to be, that doesn’t necessarily mean your app is a failure.
Hopefully, this post has helped you understand these critical enigmatic metrics and the need to monitor them. The exact figures for each metric and its benchmark remains unique for every business or the kind of application. It will take a decent amount of effort and time to set up and analyze these metrics, but make sure you do not get analysis paralysis. Monitor only those metrics that are going to help you achieve your goals.