Analytics, therefore, is the circulatory system of modern games, especially in the segment free-to-play (most free games that leave you the option to pay to improve). At the end of last year, Vasily Sabirov's book "Game with numbers" was published - the first Russian edition completely devoted to game (and product) analytics. Under the cut - an overview retelling of the book.
Friendly guide with minor flaws
Vasily Sabirov is a well-known and experienced game analyst, the founder of devtodev, a company that provides analytical services to developers. The main target audience of the book is those who are interested in the mechanisms of creating and promoting games, but so far do not have a sufficient theoretical and practical base: students, novice specialists, trainees and just game lovers. The author has set the course for a friendly “explanation on the fingers” and carries it through to the last page, equally lucidly and patiently covering most of the topics faced by the analyst in daily work - from career choices to cognitive biases.
However, the book is not without its drawbacks. One of them is that the material is not fully up-to-date. More precisely, it does not create such an impression. As you read, you realize that the Numbers Game is based on articles from the corporate blog, public reports and speeches of the author, which have been supplemented and collected under one cover. At the time of the first publication - and this is the period from 2016 to 2019 - this was obviously up-to-date information. But now screenshots, diagrams and examples with 3-5-year dates at least look outdated: everyone knows how fast the gaming industry is developing and changing. And the reader who comes across a not very fresh diagram can perceive the rest of the information as outdated (although this is not at all the case).
Another disadvantage is the slightly artificial inclusion of illustrative elements in the book. Vasily Sabirov undertakes to explain important questions of analytics using the example of a conventional game “about a hippo collecting coins”. For the target audience, this could be a great cross-cutting example that would unite all the chapters and all the narrative around it. But, alas, the author uses this game superficially, mainly as a plot for pictures. As a result, the book looks less coherent than it could be due to the implementation of a single end-to-end example.
Finally, another small drawback is the design of the links. In general, everything is in order with this in the book (there is a reference apparatus and a list of recommended materials), but there are also cases of negligence when links are simply not indicated in any way: there is a name of the source material, but it is not indicated where it can be found.
Nevertheless, we have before us the first original Russian publication on game analytics , which fully copes with its tasks: it gives a comprehensive view of this area and clearly tells how exactly the use of analytical tools helps to optimize the gameplay at all its stages.
Measure and Conquer!
The focus of the book is on “free-to-play” projects . Firstly, because the author of the book deals with such projects. And secondly, “it is the shareware games that require a special approach, it is they that imply analytics, and it is when applied to them that analytics can be revealed in all its glory,” because the life cycle and economy of such games require non-stop analytics.
The first chapters of the book explain in the most popular language what the job of a product analyst and his “standard working day” is. The author compares them to the work of a doctor who measures the temperature and takes tests every day in order to ultimately prescribe the most suitable drug and cure the patient. At the same time, it is difficult to find a candidate on the market who is ideal for the role of an analyst, even today. According to the author, the best way is to “grow” analysts within the company, guided, first of all, by the general adequacy of candidates, love of games and technical background.
The need for analytics arises as soon as a future game has a prototype.: what the future player will do, even in the early stages, should be transferred to the analytical system. But not quite “everything”, but those key events that are considered significant for the game, and “environmental events” (what the player does immediately before or after). For example, if you are tracking an “internal purchase” event, then it is advisable to include in the list what surrounds it: the entrance to the store, the choice of the product, the first use of the purchased product, the feedback about the purchase on social networks, etc.
The most important stage in the development of the project is soft launch (release of the game for a limited audience for testing and “strength testing”). The most popular metrics at this stage:
- 0-day Retention: the proportion of those who returned to the game within 24 hours;
- 1-day Retention: ;
- Tutorial Retention : , ( , );
- ARPU N : ;
- 7-day Retention: , 7 .
Map of popular metrics
All metrics, however, should ideally be based on a certain ideological foundation. And this is also a metric called the North-Star Metric ( North Star Metric , NSM) and is directly related to the level of user loyalty. Polar Star combines project profitability, user value and measurability. By defining their NSM, developers thereby answer the question “what is it all for”, what is key in their business.
The author not only analyzes in detail the types of retention metrics, but also gives a lot of tips and tricks to increase user loyalty . In particular, it is very important that already during the first session the player reaches the target event - the so-called. “Aha! -Moment” , which, for example, will be passing a level or defeating the first boss. This means that the user has figured out the application (they say that the player has been “activated”) and will return to it the next day.
One of the most important tasks of an analyst is to identify patterns of players ' churn rate. The reasons for the churn can be various factors - the quality of the product itself, high cost, attracting a non-target audience, a crowded market, problems with the level of complexity (too hardcore or, conversely, without a challenge), etc. Accordingly, the book offers several ways to reduce the churn rate, including:
- , retention ;
- , , , Net Promoter Scope (NPS, , “ , ”);
- , , , push- email-.
- , .
- , , , .
As soon as the project has a pool of players, the game activity metrics are used . They take into account how many active users (i.e. those players who have had at least one session) a game gets over a certain period - usually per day (DAU metric), week (WAU) and month (MAU). Additional indicators - CCU (oncurrent Users - users who are in the application at the moment) and PCCU (peak concurrent attendance).
– , , , , . , .
It is on an effective monetization system that the success of all shareware games is built - and, of course, The Numbers Game gives this topic a central place. There are no sensations here: all metrics are built around a "conversion funnel" that leads a certain percentage of players to pay. Nevertheless, the book provides many practical recommendations for analyzing and increasing conversions.
An example showing the amounts and number of transactions of different user segments
- Conversion to payment should be considered not as the total mass, but separately for the first and repeated payments, and also take into account payments by stages or levels of the game.
- — “” . , .
- — Paying Share, .. . — 1-2%, f2p- . — , -. , .
- : “” (whales, ), “” (dolphins, ) “” (minnows, ), . — ( ), ( ).
- “” , . , “” -, , . — “” , .
- — RFM-: (Recency), (Frequency) (Monetary) . , . , , ( ), , ( ), , , , ( push-, ).
- — ARPU, . ARPU , , , . ARPU — . — ARPPU ( ) Cumulative ARPU ( ). , .
- , , — FTPUE (First Time Paying User Experience). , . , , — .
One of the results of using metrics is a conversion funnel . In fact, this is a sequence of custom actions that shows how many unique players each took. The funnel is used to investigate user behavior and the “weak points” they fall off on. With the help of a funnel, you can analyze completely different processes in a product: from going through a tutorial to making a purchase, as well as study and optimize marketing processes: email newsletters, traffic attraction and, of course, promotions - the author tells about them in a separate chapter. No less important are user profiles that store information about purchase history, progress made in the game, data on installation time, device, etc.
Analytics culture
The final part of the book is devoted to the development of a data-driven culture - a data-driven approach to company management. It is not intuition and arbitrary decisions that are at the forefront, but the A / B test - a controlled way of testing hypotheses. There are several stages in the work of data-driven companies: preparation and analysis of data (this is exactly what the analyst does); making a decision based on the information received (and an experienced analyst should propose such a decision); finally, a solution implementation that re-starts the process loop from the beginning.
The following features of a data driven culture can be distinguished.
- Leaders are data-literate; they know they can't go anywhere without a report.
- A/B-. ( ) – -.
- , . , !
- . , .
- « , ». .
All in all, The Numbers Game is a great guide for the aspiring game analyst, a solid introduction to the analytical kitchen of a game developer. The book can be recommended to anyone who wants to get an idea of how popular modern games and applications are developing and earning, and how data from the fields of psychology, mathematics and economics is used to ensure that players get pleasure, and developers deserve it (or not very) profit.
