Whether you’re getting ready to launch a mobile app or you already have one on the market, the importance of analytics has never been more important.
Many people don’t realize is that downloads does not equal success in the app ecospace. Even if you get 10,000 people to download your app, within 1-2 days around 7,000 of those people will have either deleted your app or will never open it again.
What does this have to do with analytics? EVERYTHING!
You need to know where in your app people get stuck – can they not complete the login? Is it too difficult to figure out the user interface? What is causing users to leave the app and not come back.
Additionally, it’s very important to know the LTV or lifetime value of a user. In today’s crowded app marketplace it’s costing more and more to acquire users, whether via ads, video ads, paid downloads, etc. As it continues to cost more to acquire each user, it’s very important to know that you’re making enough to cover your costs. What’s more important though is how many of those users are good users – or users that will continue to engage in your app and more specifically – engage in activities that help you monetize them so you can keep the lights on.
Getting down to the basics
If you’re just getting into the world of mobile analytics there are a few core metrics you can start with:
- How many users are downloading my app each day?
- How many of those users are remain active after 3 days? 7days? 14 days? 1 month?
- What is the value of each download?
- What is the value of each active user?
- How do users interact and engage with the app?
- What features do users use most often? Which do they ignore?
- Which channel generates the most downloads?
- Which channel generates the most active users? The most valuable LTV users?
- Are users experiencing an friction? Any technical issues? UX problems?
- What actions encourage users to make in-app purchases? Tutorials that encourage spending content, giving away free content to encourage spending, restricting/limiting functionality, etc.
- What encourages more engagement? Achievements, goals, notifications, etc.
Tools You’ll Be Using
Fortunately, there many different tools available to help you answer those questions! Here are a few of our favorite measurement tools:
- Firebase: Google has now packaged up all of it’s tools like Google Analytics, AdMob, etc. into it’s Firebase SDK. We highly recommend that every app add this toolset, even if just for the power of Google Analytics.
- The App Store and Google Play Store: All apps these days should be both iOS and Android. Apple and Google both provide you insight into download numbers, in-app purchasing revenue, etc.
- AppsFlyer: AppsFlyer is great for apps that have multiple campaigns to track where their paid downloads are coming from and how to attribute how each campaign is doing. This is very valuable as it can help you determine that you are making $0.20 per download from your video campaign but only $0.12 per download from your paid CPI campaign – allowing you to spend your money where it makes the most impact on your bottom line.
- Appsee: Is a very cool tool that let’s you watch user sessions along with heatmaps of where users were clicking. This can be especially valuable when your app is in beta mode and you are trying to ensure its easy to use and encourages engagement before full launch.
- Amazon Mobile Analytics: Has recently gotten into mobile analytics. They let you track up to 100 million events for free and then just $1 per million events after that. We highly recommend you check them out if for nothing more than price and the tools that Amazon consistently offers.
- Flurry: One of the oldest analytics platforms. Additionaly because Flurry also runs an ad network, you can get metrics on your ad performance and detailed user acquisition reports. Their lightweight SDK is easy to install and you can start getting data right away. Get access to their user documentation here.
- Mixpanel: One really cool tool they offer is the ability to do A/B testing and notifications. This can let you introduce a new feature > see if it has the effect you want (e.g. leading to more revenue), following The Lean Startup approach that nearly every successful startup company does these days.
- Localytics: Is probably the best app notification companies out there. They allow you to do customized notifications when and how your users want which ultimately leads to better engagement and ultimately revenue. It’s been shown by Leanplum data that push notifications can increase in-app purchases by 9.6x! If you’re transparent, and make the user understand the benefit of the push, they’ll be more likely to opt in. More on VentureBeat.
- Apsalar: Apsalar is focused more on advertising attribution. This gives to a good look at your marketing ROI. Especially useful if you plan to monetize using Ads.
- Amplitude: Their tools focus more on behavioral cohorts and how different behaviors cause groups of users to act. Their goal is to bridge the layer between raw data and rich dashboards that you typically see to truly understand your users.
Setting Goals and Defining KPIs
For most developers, processing and handling all of the data available simply isn’t feasible. This is why many major studios have teams dedicated specifically to analyzing the data.
For the rest of us, we can still find value in the data if we take the right approach. Namely – we need to set specific goals and KPIs (key performance indicators) that will allow us to measure how we are doing on those goals.
Here are some examples of potential goals:
- I want to increase the lifetime average revenue per engaged user from $0.10 to $0.20.
- I want to increase the number of downloads to engaged users from 30% to 35%.
- I want to increase the number of engaged users to paying users from 10% to 20%.
By having goals that are more specific, you can setup tools and analytics that allow you to focus specifically on your goals.
Next you need to define KPIs that help you track your goals. Let’s say I want some KPIs to help me track my first goal of increasing the lifetime revenue per user:
- Average revenue per user per day/week/month (this is important so that you have a baseline value to work against).
- Average revenue per spending user per day/week/month.
- Average amount of content spent by user per day/week/month (this is important because if they aren’t spending the content they don’t have a reason to earn more).
- Number of users that go to the screen where they can earn/purchase content (this is important because many of your non-paying users may not know how or where they can get the content from).
- Number of users that engage with notifications from your app (this is important as notifications can lead to more revenue and engagement if done right).
Let’s say we’ve decided that we want to increase average revenue per engaged user from $0.10 to $0.20 (or more!).
First we want to take a look at our current app. As an example, let’s say we have an app that let’s users text and call for free (assuming they have minutes available). When users run out of minutes, they can make an in-app purchase to get more minutes.
By using our basic analytics tools we’ve determined that 10% of our engaged users or ~3/30 engaged users or 3/100 downloads make a purchase for $1 for 10 minutes.
There are several things we can do to see about increasing revenue:
- Consider implementing VIP levels that can only be earned through making in-app purchases. These VIP levels unlock additional features in the app, give bonus minutes for every purchase, perhaps allow you to choose your phone number, etc. In many apps VIP levels are becoming very popular as it gives users a goal to stride towards (which they can only achieve through making in-app purchases).
- Consider adding multiple layers of in-app purchases – e.g. $1 = 10 minutes, $5 = 55 minutes, $10 = 120 minutes, $50 = 600 minutes, $100 = 150 0minutes. This encourages the spenders that spend to spend more.
- Consider adding a daily reward – maybe for $3, each day you get 5 free minutes for 30 days, so in the end you get 150 minutes for $3! This encourages users to make the same $3 purchase repeatedly (or possibly make it a subscription).
- Consider adding a rewarded survey SDK like TheoremReach. Our SDK allows you to specify an exchange rate (e.g. $1 = 10 minutes), so if a user completes a $3 survey, they earn 30 minutes and you earn $3! This allows you to monetize users that will never make an in-app purchase or to monetize the users that very infrequently make an in-app purchase but still want the content. Typically survey eCPMs are in the $200 range (eCPM = earnings per one thousand users).
- Consider adding a rewarded video SDK like AdColony or Vungle. While videos pay much less than surveys eCPMs in the $5-$15 range, they require much less effort from users (watching a 30 second clip) in exchange for 1 minute. Additionally you can use tools like Hypermediate that will monitor performance and eCPMs of each of the different video SDKs and only display the highest paying video ad for each view (also known as the waterfall approach).
- Consider adding an offerwall like Tapjoy or Fyber – these let your users download other apps under CPI (cost per install) campaigns in exchange for content in your app. So a user can download Game of War for $0.30 of content or 3 minutes in your case. Typically eCPMs are in the $60 range for offerwalls.
- Consider adding “flash” sales or timed offers where users can earn double the minutes if they make an in-app purchase for $1 = 20 minutes in the next 7 hours. Companies like Supercell have used this technique to encourage users to make in-app purchases quickly to earn the bonus rewards.
- Tying into number 8, you can consider flash sales via notifications by targeting specific users when they are most likely to buy. As an example – by tracking that a user has just spent all or most of his minutes, then went to the in-app purchase screen but didn’t actually make a purchase. I can infer that this user wants more minutes, but ultimately didn’t feel like it was a good deal to purchase them at that time. In an hour or so I could trigger a notification to this user and other users like him, letting him know that there’s now a special sale going on where you triple the number of minutes for every minute purchased (e.g. $1 = 30 minutes!). I can then track how many users opt into this notification and ultimately make the purchase. By tracking “hot” users and notifying them when it counts, I can increase overall revenue, and help users to feel like they are getting a bargain!
We generally recommend adding and testing all of the above campaigns. It’s especially powerful to give users many different options to choose how they earn their content whether by making an in-app purchase, by completing a survey, watching a video, or downloading another app. When a user feels they have options and are in control of how they earn their content, they do so.
For each of these that you add, you’ll want to utilize the appropriate analytics step. This will help you to track revenue per user. You’ll also want to try different user interface implementations of how you display and advertise the different options to earn revenue. By testing different combinations you should be able to increase your revenue from $0.10 to $0.20 and beyond!