A Brief Intro to Mobile Game KPIs

Introduction

As large scale online services, mobile games are often improved via analytics collected from users. Today I wanted to introduce several standard KPIs that can be helpful to people who are getting into the business of mobile games for work or simply for their own recreational game dev.

I’ll be talking specifically about free to play (also known as F2P/freemium) games that are free to download then monetised via in-app purchases or ads. The F2P model has proven extremely successful at scale over the last decade, where players are motivated to spend money (or watch ads) and receive virtual goods in return.

Much of the business of games is hidden behind acronyms but these metrics can be a valuable way to see where your game stands and what effect any changes you make have. I hope this removes some of the confusion you may have. I certainly could have used a guide when I got started. :)

Data Sources

There are a variety of ways to collect analytics from your users’ actions. Some are packaged with game engines themselves; others require implementing an SDK or rolling your own. Many of these come with dashboards that can be used to check a lot of metrics at a glance.

For smaller devs using the Unity engine, Unity Analytics is a solid offering. For larger teams that have dedicated marketing, product, or data staff, I have enjoyed using Firebase and Amplitude. For marketing and install attribution, I like AppsFlyer. There are a lot of good solutions out there that will work for most purposes.

Scope

This post will cover some of the more common KPIs: retention, DAU, ARPDAU, conversion, LTV and engagement metrics.

So, without further ado, let’s jump in!

Retention

Retention is simply the percentage of new players that remain active in your game a certain number of days after they started playing.

For instance, on 10th January you had 100 players who played your game for the first time. On the next day, 11th January, only 40 of these 100 players came back to play again. In this case, your day 1 retention for 10th January would be 40%. We would say the other 60 players “churned”, i.e. they gave up on the game.

On the 7th day, let’s say only 15 players were still playing. So your 7-day retention would be 15 out of 100, or 15%.

You will generally analyse retention based on more than one day of installs. A cohort (a group of players) may be the sum of weeks or months of installs and their respective post-install days.

A sudden drop in retention on some specific post-install day may indicate a problem or lack of interest at that stage in the game and should be investigated thoroughly. The only way is down from day 1 onwards so the early stages of your game are a good place to start looking for improvements to make.

Your retention will tend to vary a lot depending on the genre of your game. It is also heavily affected by the quality of the users installing your game, the country they are from, and whether they are organic (naturally discovered your game) or non-organic (acquired via advertising.) For this reason, we often use several dimensions to look at retention, including country (AKA geo), platform, and marketing acquisition channel.

Why is it important?

You put all that effort and money into making people try your game. You want to keep them in the game as long as possible. The longer you keep them, the more chance they have to become engaged players, the larger your player base will be, and the more chance you have of making money.

Where do I start improving it?

Many would argue that the core game itself is key to retention. A game that is not fun will struggle to retain users.

However, there are short term retention mechanics like early rewards that encourage players to come back on their first few days. Longer-term retention mechanics could be providing a clear path to content that players desire, social mechanics, and live ops events.

DAU (Daily Active Users)

DAU is the number of unique users that play your game each day.

It can be used to measure the general scale of a game. More is better, as each user has the potential to provide revenue and helps to support activity in the game. This is especially important if you have a multiplayer game where players can interact with each other. You don’t want it to feel like nobody is online.

It does tie in with retention - the better you retain players, the more of your installs will remain in-game long term as active users. Equally, it ties in with your install volume, as even a player on their first day counts towards your game’s DAU on that day. For that reason, you may prefer to filter out new users from your DAU metric, for example only looking at the number of active players that have reached level 5 onwards. Looking at this each day may give you a better idea if your game is getting bigger or smaller over time, and how many installs are needed each day to sustain and grow it.

In a similar vein, WAU is Weekly Active Users and MAU is Monthly Active Users. All of these are useful to monitor.

Why is it important?

It lets you know the scale of your game and assess whether it’s growing or shrinking.

Where do I start improving it?

High retention (above) will help you to retain the new users you get. To get more installs for money, you could run ads. For free, it’s always worth trying to apply to Google and Apple for features. You have a good chance with a solid game, especially as an indie dev.

ARPDAU (Average Revenue Per Daily Active User)

ARPDAU does what it says on the tin. It’s the average amount of revenue an active user generates in your game each day. This typically is calculated based on in-app purchases, but ad revenue could be included as well.

It provides a piece of the formula for your revenue on any day, which is just ARPDAU * DAU. Users typically cost money to acquire at scale, so improving ARPDAU is a goal when working on any F2P mobile game. You can really see this improve as a result of releasing strong game updates and having successful live ops events.

Because it’s an average across your whole DAU, it can be easily influenced by UA (user acquisition) efforts as if you have a lot of new users on a particular day, it’s likely they won’t spend as much as older players and so your ARPDAU would appear lower. To remedy this, consider looking at your ARPDAU for your core players, for example only your active players who have reached a certain milestone in-game that would take 3-4 weeks.

ARPDAU is also sometimes known simply as ARPU which can be confused with LTV (see below).

Why is it important?

It’s hard to add a lot of new users without paying for them, so you want to increase the amount that each existing user contributes.

Where do I start improving it?

Study and see if you can integrate some monetisation techniques similar to the high performing games in your genre. Blogs like Deconstructor of Fun can offer some insight on how other games do things.

Conversion %

Conversion is the percentage of players in your game who make a purchase in a given period, often viewed by day. For an ad-supported game, another type of conversion would be the percentage of players in your game in a given period who watched an ad. Of course, higher is better in either case.

One way to increase your revenue and ARPDAU number is to increase the amount of revenue each paying user contributes. However, another very effective way to increase revenue is to convince more players to spend money regularly. A bonus to having high conversion is that a paying player is more attached and therefore less likely to leave your game than a free player.

This metric is something you will want to monitor and attempt to maximise over time via game updates and live ops events.

Why is it important?

The more players that are contributing to your revenue, the higher your revenue will likely be.

Where do I start improving it?

You want to make it more appealing to spend money or watch ads in your game. Think of ways to do that. As above, it’s worth becoming familiar with the best performing games in your genre for some ideas.

LTV (Lifetime Value)

The average revenue generated per user over their entire lifetime in the game. If your LTV is $10, it means every new user who enters the game on average contributes $10. Of course, lots of users will churn (leave the game) early or never spend, so this is an average.

Because it is “lifetime”, it takes a long time to gather this data. For faster feedback, we often look at LTV on day 7, 30, 60, 90, 365. For example, day 7 would mean the average revenue generated by a user 7 days after install. In this context, it’s often referred to as “ARPU” (average revenue per user) as well.

Doesn’t mean a huge amount when taken as an average across all users in all countries. It will differ from country to country, audience to audience, and platform to platform.

Why is it important?

It determines how much you may be able to afford to spend on buying installs while remaining profitable. This is one side of the equation for how scalable your game is. (The other side is the cost per install.)

Where do I start improving it?

It will be decided by your success at retaining and monetising users (see previous KPIs)

Average Daily Engagement: Sessions and Time

Activity will be really important to most games. Convincing your players to spend longer in-game each day and check in more times per day is generally a really good thing. To that end, we often look at metrics like average sessions per day and average engagement time per day.

The definition of a single session differs, but it would be reasonable to use a meaningful amount of time clearly separated from the previous gameplay session. For example, a user spending 10+ seconds in-game but only counted if there have been 15 minutes since the last “session” ended.

Engagement time per day would be the total time spent in your game each day, usually expressed as an average time per player. It may surprise you how long some players spend in-game daily!

Again, with these, you may want to analyse new players and longer-term players differently as they may well have very different results.

Why is it important?

If you can make people spend lots of time playing your game and open it lots of times per day, you’re probably winning.

Where do I start improving it?

Once again, these metrics are often improved with effective game updates and live ops events. (This is starting to feel like a common theme…)

Glossary

Wrap Up

That’s all for now! Note again that these are just a few of the most standard metrics - it isn’t supposed to be exhaustive but I may add another part in the future.

Metrics can be a really useful way of looking at your player base as a whole and assessing the impact of changes you make to your game. I hope this brief intro helps!