Mobile game metrics: what do they mean and how we use them

All services

Mobile game metrics: a vital part of game developing nobody talks about

Mobile game metrics research is a process of finding what chances do the game have to earn profits. And this is the part of game development that nobody talks about because it’s boring and far from “real work” like character drawing and coding.

Most gamers think that game development is an act of art from start to finish. Yes, most mobile game developing parts are artistic processes, but market research and metrics analytics are nowhere near. In this article we will tell you what we do in the first weeks of development, how we analyze the market and why a good game can fail because of bad metrics.

mobile games metrics research

What is mobile game metrics

Every sane company does market research before releasing a product. Usually market research is done by quizzes, sometimes — by giving testing products to the potential buyers and noting their reaction. Testing is needed to predict how many potential buyers will leave a product they used to and buy a new one.

Mobile game market research is done with the same goal: to predict how many gamers will install the game. Research is made before the actual development to save time and money on projects that won’t work in the current market. Also, the research can predict how much money the publisher will need to spend to attract new players and how much profit the game will generate.

By doing the research we collect such data as:

  • CPI — cost per install;
  • CTR — click through rate or how many users will install the game after seeing it’s ad;
  • D1 — day one retention rate or percent of players who will install the game and play it on next day;
  • D7 — day seven retention rate or percent of players who will play the game after one week.

Ideal game has low CPI and high CTR and D1 numbers. Every popular mobile game you’ve played or heard about had them. Concrete numbers are:

  • CPI less than $0,3 on iOS and $0,2 on Android;
  • CTR more than 5%;
  • D1 more than 40%.

How the mobile game metrics research is made

The research is made with a so-called D1 prototype. The prototype is not a full game, but it has to provide the same experience as the complete game.

The D1 prototype must have next features:

  • complete art style;
  • complete core gameplay;
  • no monetization;
  • at least few levels, randomly generated levels or infinite gameplay;
  • set of icons for publishing it;
  • complete ads which show art style and gameplay.

Development of the prototype with all of this features can take 2-3 weeks for hypercasual game or 1-2 month for games with more complex mechanics.

metrics research process

Testing is simple: we publish the prototype under misleading or work title. For example, if we are testing some type of clicker, we will publish it as a “Tile clicker” or something similar. Then we will buy some ads for our D1 prototype and measure the player’s behavior: CPI, CTR, D1 retention. For more complex games we will also measure D7.

If the game is part of a well known franchise, e.g. it has a Mario or Pokemons as main characters, we will make the prototype available only for small particular countries.

After 1 day or week of testing we will delete the game from stores and make a decision: to make a complete game out of the prototype, to improve mechanics or to throw the prototype in trash.

Reminder: testing is needed only to analyze game mechanics and art style. We spend a small part of the budget just to see if the users will play it for some time.

How we use metrics to improve the game

The first thing we look at is the CTR number. Basically, the CTR shows us the percent of users who saw ads of the prototype, clicked and installed it. Out of these stats we can see the effectiveness of the ads we were planning to use for the actual game. Conclusions are simple:

  • low CTR means great advertisement video;
  • high CTR means that our ad doesn’t interest users.

The second thing is CPI. It shows us the efficiency of our marketing. The key to marketing is to advertise the product to the right audience. More specific and less-valuable key phrase means lower CPI. The trick is to find balance of CPI and CTR.

d1 game prototype

The third thing is D1 and D7 retention. If D1 is low, we know that players get bored quickly and don’t want to play the game again. That means that we need to improve gameplay and make more content.

But we do not want to look at the metrics separated from each other: we need to make decisions based on complex images. For example, if the CPI and D1 are good and the CTR is bad, we can draw conclusions that we created bad advertisement materials. Maybe, the icon isn’t catchy, or the gameplay video doesn’t show the player all the fun in game. In this case we need to improve ads and test it again.

By doing the testing we can be sure that we are developing not just a great game but a successful product.

Why the mobile game metrics are so valuable

The key to profit is to make the game popular. We can’t do it by ourselves, even with your help – it needs a lot of money, 70-150% of the overall development budget. The best thing to cut marketing expenses is to publish a game with the help of a professional publisher. But they won’t work with every game: first of all they look at the D1 prototype metrics.

If the metrics pass their requirements, the publisher will do all the work related to advertisement and traffic in exchange for part of the profits.

If the metrics are bad, there is no point of further work with this prototype. Bad metrics means that the players are not interested in core gameplay mechanics and artstyle, which are the only two things how people decide which game they want to play.

The failed testing means only that we need to try other ideas or to advertise the game to different audiences.

How a good game can fail without proper testing and how to do the testing right

Imagine that you have developed a good hypercasual game but forget about D1 prototype testing. You’ve published it and bought some traffic, but the game got almost no installs and earned no money. Why? Maybe players didn’t like the setting or art style? Maybe they didn’t like the gameplay at all? You can’t know this because you didn’t do the testing. And even if you know why the game has failed, you have no money left to make all the changes needed.

Testing is a lot cheaper than full mobile game development. We use some tricks along with basic testing to collect more data for our money:

Test multiple hypotheses. Roughly speaking, release one prototype with knights and dragons and the other one with aliens and spaceships. Don’t change the core mechanics, just the setting. By doing this you will know what setting player’s do like the most.

prototype style variations

Try different types of ads. Maybe the puzzle game does not need a CGI video and the plot-driven RPG totally needs one? Maybe it’s better to advertise your puzzle game among first person shooter lovers? You never know until you try.

Add no monetization systems in D1 prototype. In the D1 testing we check the effectiveness of advertisement and core gameplay, not the ingame economy system. The prototype must be a perfect and absolutely free demonstration of gameplay, and the testing is needed to prove that the gameplay is addictive.


Code, art and music are useless if there is no good advertisement behind them. The good game won’t become a successful one without proper marketing because nobody will know about it. And you’ll never know that your game is good until you spend money on publishing and advertising.

mobile game profits

Most developers and investors are trying to create a perfect game to fill their needs but not the market needs. No one can surely know what the crowd would like tomorrow. But prototype testing can give us a hint what the crowd does like now.

If you want to develop a mobile game with a high chance of success — contact us. Otherwise you risk to spend money on a well developed game which will attract few thousands users because no one has done a proper prototype testing.

Keep in touch:
+44 7 44 55 3 66 31

Send us a message via messenger or email