Every self-respecting startup must have a mobile application, and some startups are even totally focused on mobile. But for some reason many still have not learned how to correctly analyze the usage of their mobile applications. From which channel did the app user come from? Is he actively using the mobile app or just opened it once and never came back? These and other important questions can be answered with the help of mobile analytics.

All analytics applications can be divided into two types - those that track the conversion from external sources, and those that measure the events within an application.

Tracking conversions from external sources

The most popular tools for this kind of analytics are Mobile App Tracking, Appsflyer and Ad-x. Most mobile app developers seldom integrate these analytical services into their apps. This is quite strange because you can track a lot of essential metrics with these.

- Measure the conversion rates from each advertising channel

Moreover you can not only track installs, but also be able to see and analyze all subsequent steps. Also you'll be able to see it all in a form of nice looking graphs.

On the above picture you can see that the fourth and fifth sources gave roughly equal number of installs, but the fourth source produced much more conversions into further purchases and loyal users (those who launched the application 3 times). Of course, you can download these graphs as .csv files so that you could measure the ROI of each channel.

To measure the conversion, you just have to register your referral link for each source. And it works for any source - you can insert the tracking link into a banner in advertising networks, as well as in app reviews or Facebook posts and measure how many users you were able to acquire via any of these sources. You can also make a separate link of viral installations - when sharing a link to social networks or sending it to your friends use a special referral link. You will be able to see how many people are coming from this channel and how they behave within your app.

- Buy advertising on cost per install (CPI) terms

Paying for the installs of the application, and not for clicks is great opportunity to shift the marketing risks to the advertising system. But in order to pay for installs your app needs to report the fact of installation to the ad network. The aforementioned Mobile App Tracking, Appsflyer and Ad-x are are compatible with most of the CPI-networks.

- Do retargeting

Facebook recently rolled out a awesome tool for retargeting in mobile applications. In most cases Facebook would require that you use their SDK but it really is not necessary - the above trackers are among the partners which are integrated well with the social network.

Measurement of the events within the application

The most popular systems - Flurry and Google Analytics can measure many of the most important in-app metrics. These tools are also frequently used by many iOS and Android app developers. Still, few people use them to their full potential. You can also measure a lot of interesting stuff with their help.

- The daily number of active users in the application. You will be able to show investors some cool graphs indicating how relentlessly your user base is growing every day. Both Flurry and GA display these metrics without any additional tune ups.

- User Retention. Most apps loose their user base within one month without any marketing support. If you have a retention rate of 30% and more - you can freely show this metric to your investor as well. Flurry is capable of showing the retention rate without any additional settings.

- Screens browsing. This metric basically shows how your mobile app is being used. Google Analytics is capable of displaying this metric without any additional settings, while in Flurry - you would have to set an event for each screen.

- In-App Purchase. You can set an event for each payment in Flurry while GA has a wonderful ecommerce module included as standard.

- Any significant events within the application. For example, signing up on the start screen. You would be able to see what percentage of your app users registered immediately. Both in Flurry and GA you would need to configure each event.


Mobile analytics services from the first group measure the conversion rates of different user acquisition channels and allow to grow user base on CPI basis, but do not work that well with organic traffic. The services from the second group measure organic traffic while being not that good at identifying the source of the installation. Formally, both Flurry and Google Analytics are capable of identifying the sources of new users, but in practice it works not as accurate as you would think they would.

It is therefore best to use both systems of analysis - one of the first group and one from the second.

by Dmytro Bilkun