Intro to Metrics

Part 1 of 4 of A KPIs Guide for Google Play Apps and Games

Alyssa Perez
Google Play Apps & Games
7 min readOct 22, 2019


This piece is part of a larger series of articles and business growth webinars about understanding Google Play app performance based on key metrics.

When talking to developers, we at Google Play often refer to a standard set of key performance indicators (KPIs) in our newsletters, public presentations, and articles like this one. We know you are regularly analyzing the performance of your app or game based on important metrics that come from your own data sets, marketing channels, and cross-platform visibility, so we want to support the quality of your analyses by offering strategic recommendations straight from our Google Play business growth experts.

This article will answer questions like the following:

  • “What metrics should I check if my user base is growing but my revenue is not?”
  • “How should the age of my app factor into my analysis of new installs?”
  • “How does Google Play define some of the most important metrics for my business?”

Behind any successful app or game is a team with a strong understanding of how, why, and when certain metrics can and should be used to make product- or business-related decisions. As developers come up with new creative ideas, it’s good to revisit the fundamentals and the contextual nuances around key performance indicators.

Start at the top

In this series of articles we will focus on an important top-level metric for developers: Daily Revenue. Below is a visualization of how the Google Play Growth Consulting team breaks down some of the KPIs that drive Daily Revenue and that top developers usually monitor. There are many ways to set up such a tree, but this is the one we will focus on.

A metrics tree built around Daily Revenue. Dotted lines signify child metrics that are added to equal the parent metric; all others are multiplied.

We’ll start with how to read this chart, then we’ll cover some basics pertaining to each metric.

Think of each ‘parent’ metric as the result of the relationship between the ‘children’ below it. This will help you remember how the parents are calculated, but more importantly, it also lays out the different levers you can pull when you want to increase or decrease the parent number (don’t worry, we’ll walk through this below with examples). Note that, generally speaking, the further down the tree visualization you go, the more important your app’s unique situation is in interpreting how to pull those levers; so we’ll keep it high-level for now, in order for this breakdown to be relevant to as much of our community as possible.

Let’s start from the top and work our way down the left side of the tree pictured above.

A metrics tree, built around Daily Revenue, with DAU-related metrics highlighted

Daily Revenue

  • Answers the key question: “How much money is my app bringing in each day?”
  • Calculated by: multiplying the child metrics, which are Daily Active Users (DAU) and Average Daily Revenue Per Daily Active User (ARPDAU).
  • Example ‘lever’ and analysis: I successfully grew the size of my user base, but my Daily Revenue stayed the same. Why? If your DAU increased but your Daily Revenue did not, check your ARPDAU. You may have more people using your app overall, but the ratio of paying users to general users went down, which means the growth in your user base was not coming from users who are spending. (To drill down more, see the example below for ARPDAU.)

Daily Active Users

  • Answers the key question: “How many total people used my app on a specific day?”
  • Calculated by adding the child metrics, which are New Installs and Returning Users.
  • Example ‘lever’ and analysis: I had a ton of New Installs, but my DAU stayed the same. Why? If your New Installs increased but your DAU did not, check your Returning Users. If DAU stayed the same, it means despite new people coming in, they are not coming back. (To drill down more, see the example below for Returning Users.)

New Installs

  • Answers the key question: “How successful are we at getting new users into our app?”
  • Calculated by: adding the child metrics of Organic Installs and Deep Link Installs
  • Example ‘lever’ and analysis: Are my paid marketing campaigns working? If there is a spike in new installs that correlates with the dates of marketing launches, campaigns, or promotions, then the answer is yes.
  • Pro-tip on nuance! If looking at the ratio of new installs to your entire user base, don’t forget that the age of your app or game and amount of marketing spend can have a big impact. Older apps will, naturally, have a smaller new install ratio, since a large chunk of your users will have had a longer time to hang around; this means the trend of your new installs may look like a plateau with occasional spikes. However, new apps are more likely to see a higher ratio of new users to entire user base, because they are just getting started.

Returning Users

  • Answers the key question: “How successful are we at getting new users to come back?” Seeing high values here is indicative of a strong-performing app, since it means you can continue to grow your active user base as long as new users are coming in.
  • Calculated by multiplying retention rate on day [x] with the Install Cohort (new installs from [x] days prior) and summing over all [x] days. We’ll dig deeper into this particular formula in an upcoming article.
  • Example ‘lever’ and analysis: Are more new people returning because of the improved onboarding experience? If your retention rate is higher in the month following your updates than in the month before it, for example, the answer is likely yes. (Even though the Install Cohort size may be different, the retention rate will tell you if a higher or lower percentage of them responded well to your improvements by returning.)
  • Pro-tip on nuance! All users that return to your app after their first day will fall into this bucket, but this metric and its implications are highly dependant on what day or time period you’re evaluating and how big your cohort is, so don’t forget to look at the context as well.

Now let’s jump back up and work our way down the right side. For simplicity’s sake, let’s choose in-app purchases as our example app’s monetization method (we will also cover subscription-based monetization later in this series).

A metrics tree, built around Daily Revenue, with ARPDAU-related metrics highlighted

ARPDAU (Average Revenue Per Daily Active User)

  • Answers the key question: “How well am I monetizing my entire user base?”
  • Calculated by multiplying the child metrics, which are Daily Conversion and Average Revenue Per Paying User (how much spend on average is coming from each user who is paying).
  • Example ‘lever’ and analysis: My Daily Conversion has gone up, but my ARPDAU stayed the same. Why? If a higher percentage of users are spending and overall revenue per daily active user is flat, check your Average Revenue Per Paying User to learn more. You may see a broader range of spending levels — meaning you likely have a mix of high-value users and smaller spenders, which affects the average.

Daily Conversion

  • Answers the key question: “What percentage of my active users see enough value to spend money?”
  • Calculated by adding the child metrics, which are New Buyer Conversion and Repeat Buyer Conversion.
  • Example ‘lever’ and analysis: How can I increase overall conversion? You can focus on two strategies: drive more new buyer conversion or drive more repeat buyer conversion. When focusing on repeat buyer conversion, you can create strategies based on their prior conversion — however it’s important to note that the end goal is to increase the overall buyer base, so focus on growing both if you can.

ARPPU (Average Revenue Per Paying User)

  • Answers the key question: “Of the users who see enough value in my app to spend money, how much are they willing to spend?”
  • Calculated by multiplying the child metrics, which are the Average Transaction Value and the number of transactions an average buyer makes (# Transactions per Buyer).
  • Example ‘lever’ and analysis: Why has my Average Revenue Per Paying User gone down, even though the Average Transactions per Buyer is the same? Given how ARPPU is calculated, you can conclude that the average transaction value must be decreasing in this scenario. A good way to keep abreast of opportunities for growth is to consider your pricing strategies. Remember that if the size of your buyer population is changing, that can also have an impact.

Congratulations! Now you know more about some of the fundamental KPIs that our Google Play Experts recommend app developers get familiar with for informed decision-making.

What next?

If you enjoyed stepping your way down these branches in this high-level overview, good news — this is just the tip of the iceberg. There are more levels of ‘child’ metrics and levers to pull beneath these, especially when it comes to monetization (for example, in subscription-based businesses, which look slightly different than IAP-based portion of the tree above. We’ll cover that difference later in the series).

There are also a variety of ways to look at how changing your app design around engagement can influence retention. There is so much to unfold, discover, and dig into, especially if you are open to trying new strategies with your product. You may even want to start experimenting with levers designed to affect a particular segment of your audience, if you aren’t doing so already.

If you’re hungry for more, be sure to sign up for one of our upcoming webinars or follow us on here to catch future articles in this sustainable growth series. Happy analyzing!

What do you think?

What metrics are you most or least familiar with when analyzing app performance? Let us know in the comments below, or tweet using #AskPlayDev and we’ll reply from @GooglePlayDev, where we regularly share news and tips on how to be successful on Google Play.