• Why the first ten minutes are crucial if you want to keep players coming back
  • By Adam Carpenter
  • The Nuggets translation Project
  • Permanent link to this article: github.com/xitu/gold-m…
  • Translator: Cherry
  • Proofreader: hanliuxin5, ryouaki, baileilei

Post 1 of 3: How do you analyze your mobile game retention data

As a mobile developer, one of the most powerful tools you have is data. Using game data in the right way offers incredible opportunities to identify problems, optimize performance, provide value to players, and ultimately grow the business.

In this article, I’ll discuss first-time user experiences, which can help you determine where in your game you might lose players. I’ll also share some meaningful data from Google Play to better understand your game’s performance and identify opportunities for improvement.

Average player retention in Google Play games

Retention is one of the key performance metrics for an install, along with buyer conversions and average revenue per install. In many ways, retention is the main indicator, because if you can keep your new players, you can always figure out how to make money **. ** If you can’t keep any players, you can’t make money.

The retention rate itself is calculated by dividing the number of installs by the number of active users at a given retention date. If a player downloads a game for free on Google Play, the average 2-day retention rate is 38%. A retention rate of over 46% after 2 days will get you featured on the first page of the store, which means you’ve already passed 75% of the charts.

(Second day retention)

However, it is also important to think about the implications of the data. A second day retention rate between 22% and 52% actually means that 48% to 78% of users who play the game one day don’t play it the next day. If you want to build and market your game, do everything you can to improve it.

What should you focus on in order to improve retention

Many developers focus on metrics like the level the player reached on day one or the tutorial checkpoints the user passed. However, these metrics are game-specific and don’t tell you how they compare to similar games.

One key metric compared “play time on day 1” to “retention time on Day 2.” This metric, day 1 playtime versus day 2 retention, is more of an Apples-to-Apples comparison, making comparisons more valuable and comparable, and Google uses these metrics to help partners identify early flaws and improve performance for new users.

Day 1 Data from first-time players

This graph shows how long new players played the first day, and what percentage of players remained the next day. This trend suggests that the longer they spend playing the game on day one, the more likely they are to keep playing.

We can feel it intuitively. Let’s say someone spends a lot of time playing the game on the first day, and the more fun they have, the more likely they are to want to come back.

What’s really interesting, though, is that we split the top games into four tiers based on day 2 retention. The first tier, the best performers, had an average retention rate of 52 percent the next day. Their retention rate starts to rise strongly around 22% and steadily increases from minute to minute of play. The second tier retained an average of 42% the next day, and the third tier retained an average of 32% the next day.

What we can see is that most echelons of four show very similar trends after the first ten minutes; They all curve up and to the right with a decreasing slope. However, it is in the first ten minutes that the most interesting pattern is visible.

The first ten minutes are crucial

This chart zooms in to the first 10 minutes, and that’s where we can see a very distinct pattern emerging.

For the best players (the green line), their retention starts to rise strongly and steadily, and the second tier of players (the blue line) shows a different pattern, so we can see retention leveling off for the first minute and a half before steadily increasing. For the third tier four players (the orange line), retention remained essentially unchanged for the first four minutes and then began to increase. But at a slower pace. The last tier 4 player (the red line) actually saw a drop in retention for the first two minutes, and it wasn’t until the fifth minute that retention passed its starting point.

Now, let’s take a look at how the earlier model affected our users. The chart below compares the cumulative attrition rates of users in the top four tiers (green line) with those in the bottom four tiers (red line).

The worst performance was losing 46% of new users in the fifth minute. By 10 minutes, they had lost 58% of new users. Basically, more than half of new users don’t even play for ten minutes. By comparison, the best performers lost only 17% of users in the fifth minute and 24% in the tenth minute.

The first five to ten minutes are critical, they change retention the next day

These top games managed to retain twice as many users as their lower-performing counterparts. But there’s still a key question: If you can retain twice as many users as you retained on Day 2 or less, how does that help your daily active users (DAU)? Will that increase your income? Using data from Google Play, we identified two key patterns.

Don’t make retention “flatlands” and “canyons”

The first mode is called flatland. The antipattern remained largely flat for ten minutes, before the percentage rose meaningfully after five to ten minutes. The second is the “canyon,” which stays down minute by minute for the first five minutes or so, and then starts to rise again.

As a broad estimate, it is likely that between 25% and 50% of games exhibit one of these antipatterns. Using data from your own data warehouse, you can generate these charts to see what patterns are emerging in your game. If you see flatlands or canyons in your own data, there are a few things to check:

  • Can your game be downloaded from other sources? This can cause players with poor wi-fi to quit.
  • Was your tutorial interesting? Will it increase player goodwill?
  • What is your loading process like? It’s hard for new players to accept a long wait because they’re not in the game yet.
  • Is your game intuitive for novices? Does the person coming out of the tutorial know how to operate, how to build their foundation, and how to start having fun again? All of these considerations are important to ensure that users stick with the game.
  • Did you make a lot of sales on the first day? This strategy may yield some short-term gains, but will reduce overall retention. Consider running a test that cancels offers on the first day, but makes players feel rich and maximizes their fun.

One of the great things about optimizing the first ten minutes of your game is that you can quickly iterate through experimental changes and get results from A/B testing within A few days.

Start increasing your app’s retention

The first few minutes of a game are critical moments in an app’s life cycle. At this point, their only investment in the game is the time and effort it takes to download it. Any negative experience will cause the player to quit, or move on to another app.

If you find an anti-pattern of “flatlands” or “canyons” in your game, or low retention at the beginning, then look at what might be causing it. Consider whether secondary downloads, long load times, or other factors could have a negative impact. If you can connect the negative retention in the first ten minutes to any particular factor, test moderately changing or eliminating the cause, and you should see an improvement in your game’s retention, daily active users, and ultimately revenue.

What do you think?

Do you have any issues with anti-patterns and player retention in game data? Continue the discussion in the comments below or use the hashtag # AskPlayDev and we’ll reply @Googleplaydev. We often share news and tips on how to succeed in GooglePlay.

This is the first article in a three-part series, but look out for my second next month, where I’ll discuss further how to use data to understand and increase player engagement. Then in my third and final article, I’ll explore whether players and “payers” are happy, and how to use these insights to drive conversion.

The Nuggets Translation Project is a community that translates quality Internet technical articles from English sharing articles on nuggets. The content covers Android, iOS, front-end, back-end, blockchain, products, design, artificial intelligence and other fields. If you want to see more high-quality translation, please continue to pay attention to the Translation plan of Digging Gold, the official Weibo, Zhihu column.