Measuring App Install campaign’s effectiveness
We know how smartphones are on everyone’s purchase list in India and almost 100mn users have already shifted. One of the reasons for its success is availability of wide variety of apps to make your phone experience better. These 100mn users in India are doing almost 500mn installs every month and approx. 1/5th of these installs are paid installs where App developers are paying ad-networks or other publishers to get their app installed.
While its always good to get organic installs, paid installs are desired by most app developers to acquire inorganic growth.
So how does one find out if such app installs (organic or paid) are proving effective for app developers?
Answer is not very simple — effectiveness can be defined as:
1. Increase in Rankings
2. Increase in mobile transactions/users
3. Increase in revenues from mobile
4. Reduction in user’s attrition
5. And several such KPIs defined for your business
So let’s find out how do we measure if your app installs are effective.
Your first source of information is App store dashboard itself. Both Apple & Google will provide you basic information like new app installs, uninstalls, total installs and so on.
So a graph like below will give you an overview of how is app performing. But frankly that’s not enough and analytics provided by App stores do not go beyond this point.
Image 1: App Install Growth Chart (Image Source: Google Play store dashbaord)
Does increase in app installs means increase in active users or increased revenue? — The answer is Not always.
So unless your answer to above 2 queries is yes, the increased installs have gone waste and are not a good indicator for your mobile business.
While you look at daily installs, don’t forget to look at daily uninstalls, as that’s the first indicator of the health of your app install campaign. You can easily find this information from your respective app store’s dashboard.
Biggest issue in app installs is that it’s not easy to find out the source of each installs without using any 3rd party SDKs like Hasoffers MAT, Apsalar or Facebook (only for FB driven installs)
Why do you need to find the source of each app install? — answer is simple — you promote your app across Twitter, Facebook, Blogs and Ad-networks and its good to find out which channel is more effective.
Also you may experiment by writing different communication messages across various channels so it helps there as well.
This becomes even more important if you are running paid campaigns across various networks and you pay based on installs (CPI — Cost per install campaigns)
Image 2: Sample App Attribution Chart (Image Source: Flurry)
So now that we have established the importance of tracking app attributions, we now need to see how to measure the success of your campaign.
Again to go beyond mere install numbers, App developers need to install analytics SDKs like Mixpanel, Flurry, GA for Apps, Localytics or Apsalar (I am not delving into which tool to use as that would require a dedicated post)
Indicator 1 — App Rankings — Most app store ranking algorithm will not just increase rank based on daily installs, they will measure uninstalls, active users, reviews & ratings and multiple sources driving the app install. So if your rank goes up then celebrate because it means you are going in the right direction
Image 3: App Rank Graph (Image source: AppAnnie Analytics)
Indicator 2 — App Reviews & Ratings — For most new App installs 1–2% users will review or rate the app on store and it goes up to 3% in case of app update release. Check if recent reviews & rating inflow is inline with your app install rate. Its an indicator if people are actually using your app or not
Image 4: App Reviews (Image source: Play Store App description page)
Indicator 3 — Active users — Once your app is installed, its important to know how many users are active. This is a real test of your app install campaigns (Organic or paid). Global averages say that Install to monthly Active user ratio should never be below 10% of your install base. Desirable range is 12–18%. Anything above that signifies great traction for your app (Eg. Apps like Facebook are exception and get up to 60% Install to Active ratio in some countries)
Note — Tracking active user requires one of the analytics tools mentioned above
Image 5: Active users growth graph (Image source: Flurry)
Indicator 4 — Repeat users — Key to increasing Install to Active ratio is to get users coming back to your app. Most analytics tools will provide you cohort analysis and the most basic function you can perform is to find how your newly acquired users are returning week on week. Higher the %age of repeat, better it is.
Image 6: Sample cohort graph (Image source: Apsalar)
Indicator 5 — Revenue Traction — Apps can have multiple revenue models like Advertising, IAP or direct sale (ecommerce).
Advertising Revenue — If you are selling advertising based on impressions or clicks then you must keep an eye on daily/weekly session in your app. All analytics tools provide this information. Session = It starts when user accesses the application and ends when the user quits the application. So more sessions means opportunity to serve more advts.
If App is using any ad-network to serve advts, then this data can be co-related with Ad-network’s impression data.
Image 7: Session in app chart (Image source: Apsalar)
IAP or Ecommerce Revenues — It is important to setup revenue tracking using your analytics & attribution tool. While you can find cumulative revenues from App through your own reporting, good analytics tool will help you find ARPU & revenues from each source. So you will be able to make a better decision in selecting a promotion medium/network based on revenue earned through installs from a particular medium/network. Note — Revenue tracking is usually an optional integration in each Analytics SDK often missed by App developers. Ensure that you follow the guidelines to track actual events.
Indicator 6 — Source wise Conversions — Conversion is defined as the achievement of end goal you expect a user to hit inside your app. It could be purchase of an item, post an advt or click on an advt. For every user that installs your app there are bound to be certain conversions. If the conversion from each source matches with your business average then that network/medium is trustworthy. To see conversions, you can setup funnels in your analytics tool which is a pretty common features available across all tools. Note — you need to setup event tracking in your app to create & visualize funnels
Image 8: Funnel visualization (Image Source: Google Analytics)
If you see positive signals from all the above indictors then your app install campaign has proved effective. These indicators will help you eliminate bad sources and focus on better sources for app installations.
You can connect with me on twitter @deepakabbot
P.S. Most of the indicators mentioned above should be available in almost every analytics tool but their form of representation may be different. Do proper research before integrating any SDK in your app as it can be time consuming.
Useful links:
Apsalar — www.apsalar.com
Localytics — www.localytics.com
Mixpanel — www.mixpanel.com
GA for Apps — www.google.co.in/analytics/mobile
Google Play Campaign Measurement — https://developers.google.com/analytics/devguides/collection/android/v2/campaigns#google-play-campaigns
Tracking Adwords conversion in apps — https://support.google.com/adwords/answer/1722054?hl=en
Facebook SDK for Android — https://developers.facebook.com/docs/android
Facebook SDK for iOS — https://developers.facebook.com/docs/ios
