Itamar Rogel

How player data analytics help reduce churn

There are various ways by which companies deal with players who stopped doing business with them, namely with those who churned. Player churn is always a headache for businesses, and after they employ churn data analysis and measure what happened, they begin examining why it happened and then focus on efforts that would help reduce player churn.


But as the player churn rate is relatively easy to calculate, the reasons for it happening and the means to analyze customer behavior can be, and often are, a tough nut to crack. Sometimes it is the game itself that players complain about, other times it is the company’s service that is the cause they churn, and oftentimes it’s the competition that comes up with a better game or feature. And as they say, the player knows best.


How player data analytics help reduce churn

Connecting data analytics with player churn


In recent times, data analytics tools help companies deal better with player churn. When they choose to employ player feedback data analysis, companies begin to better understand how to run their churn prediction, reduce churn and increase player retention.

Data analytics helps both understand the problems as well as create possible insights and solutions. Many companies use data-driven methods to figure out why players are churning and even make an effort to discover those who are likely to churn. Using advanced analytics, artificial intelligence, and machine learning algorithms, helps them better understand users and perform a more accurate churn analysis.


By figuring out the data about what users think, need, and want, game studios can improve their retention rates and reduce churn. Data analysis of a player’s journey and user experience can assist a company to identify the segment and problem areas, leading them to plan how to fix them.


Player data must be tracked and analyzed for a better long-term focus from companies on true value solutions. But since the data quantities are enormous nowadays, the employment of machine learning and AI-powered solutions can bring about better business models, insights, services, and solutions.


Calculating churn


There are three methods of calculating churn, and companies usually use any one of them or a combination of all three (an integrative approach). These quantitative methods explain “what happened” in the world of churn rate.


As churn prediction is always on the table, companies are trying to employ whatever formulas and methods they can use in order to better prepare to deal with it. With volatile and very competitive markets in which companies find themselves dealing, each of the following calculations can expand companies’ understanding of the churn aspects.


1. Subscription churn

This is simply the calculation of the number of players that quit over the total number of players over a period of time. This method serves for a basic understanding of user churn rate and gives an overall description of how many players are churning.


The first step to understanding the churn usually begins with this formula. The calculation can be done over pre-defined time periods (month, quarter, annually, or year-to-year) and it answers the question about the size of the problem.


2. Non-subscription churn

Churn can further be figured out by measuring the number of players that have dropped below the threshold and are considered inactive, within a time period. It is up to the company to determine the exact threshold, as it can change from product to product or over time, but it still serves as the quota that must be met and obviously didn’t.


User attrition measurements are done by the above two methods, but there is another quantifiable way to figure out the churn rate and players who stopped doing business with the company.


2. Revenue churn

Revenue calculations in relation to churn rates can be tallied by the amount of recurring revenue lost over the total amount of recurring revenue at the start of the period. This is yet another method to figure out just how much a business is hurting.


Using those numbers’ methods is always the first step towards understanding players and churning better. Other than the obvious numbers, a thorough qualitative data analytics examination is required, in order to find out why users churn. Then a company can offer players value propositions in order to decrease churn.


The importance of reducing churn


With customer churn as a top priority on managers’ minds, they have multiple reasons to take proactive steps to reduce it as much as they can. Some of them use churn prediction methods to try and calculate in advance how many customers are likely to churn vs. those who are less likely to churn.


But whether a manager is able to predict churn or not, once he uses the above-mentioned methods to calculate the actual numbers of customers stopping doing business with his company, he must measure the existing losses created.


The reasons companies do all that they can to reduce churn vary. Apparently, player churn creates several problems for every business and reducing churn becomes a top priority that game studios must focus on resolving.

  • Avoid revenue loss

Business revenue loss is the number one problem for every organization. As profits and growth signify the aim of every business, when companies lose revenue because fewer players downloaded their games or stopped doing in-game purchases, the game must be hurting.


The next best thing in that type of situation is to measure the damage done and then figure out and have insights into the reasons for such a player churn. Something along the user journey isn’t working, so the company knows that it must find out the causes for such a problem.

Avoid revenue loss
  • Reduce player acquisition costs

Churn by players means also lost revenue of the money spent trying to acquire new ones. Every business model includes various marketing efforts, and once players churn, such amounts are lost too.


So every player churned means not only lost revenues but also user acquisition costs are gone. That is why many companies invest lots of energy, and money, in retention programs and other retention efforts.

  • Lower marketing and sales costs

Another key reason why game studios make every effort possible to reduce player churn is that they try to decrease marketing and sales costs. Since such costs appear on the expenses side of every budget, they become more problematic when more players churn.


A careful user journey analysis and insights are called for in order to better plan future marketing and sales expenses. Evaluation of the existing player base and more data analytics regarding how to better match expenses vs. results are also required if a company wishes to reduce such expenses.

  • Improve player lifetime value

The player lifetime value (LTV) refers to the measurement of the average player’s revenue generated throughout their entire relationship with a specific company. So for existing users who decide to churn, game studios can figure out the income loss from their stoppage of playing, downloading, or in-game purchasing.


In order to reduce player churn, companies go out of their way to make sure users continue buying from them. The next best offer, special features, discounts, and a variety of other exclusive deals are offered to players in order to increase retention and reduce churn.

  • Upgrade quality of player experience

Player experience support serves as another key element in business-player relationships. Users value the service they get from organizations as they evaluate just how much those businesses care about them. So the whole process of handling complaints and then using player comments to ensure quality service means a lot to businesses.


Since there are ways to measure player satisfaction from the service they receive, a company can compare its score from period to period and figure out how to further improve this aspect of its business. Sentiment analysis also contains how a user views player support, so every business can receive a pretty good idea of how it is performing here too.

  • Increase revenue options from existing players

Up-sells and cross-sells mean getting more income out of a company’s existing players. This can be done by offering them extra features and building upon the trust they have for the business which is based on previous ones.


By getting more revenue from existing users, a business reduces the risk of losing income from players that churned. So a user loss may give the company somewhat less pain if the business manages to cover such loss by getting more value and making more money out of a regular player.

  • Maintain brand and game reputation

The final key in the importance of reducing churn relates to keeping the positive sentiment and brand reputation of the entire business. After all, players do not like to continue doing transactions with a company that suffers from a large churn due to the underperformance of a game.


A sentiment analysis, when done on a regular basis, can give a company a glimpse of what the overall reaction players carry for its brand or game. Data analytics, when calculated properly, can help businesses identify their current position as well as compare such brand reputation scores to previous time frames.


The types of quantitative churn analytics


Businesses use player data and other analytics to calculate churn. There are predictive models and post-churn models, as organizations try to figure out and analyze their data regarding churn risk.


While churn can be difficult to predict, since many factors must be considered for such a calculation, it still can give businesses ballpark figures regarding possible revenue losses.

While the data gathered from the numbers’ methods is not sufficient to comprehend the full situation, and a player’s voice and opinion data analysis is required for comprehensive knowledge, there are a few analytics models and tools which organizations use to try and measure their churn risk.

  • Predictive analytics

When businesses wish to predict the future outcomes of their sales picture, they use predictive models, statistics, and modeling techniques. Predictive analytics takes current and historical players’ purchasing data patterns, in order to identify if such patterns are likely to repeat themselves.


Data insights derived from predictive analytics make companies use a form of personalized marketing called “Next best offer”. They offer players products based on their habits and past behavior. In doing so, businesses predict their users’ needs and wants, and later analyze whether such efforts materialized.


Predictive analytics is an interesting method of figuring out the player experience and trying to repeat it. Insights from previous purchasing patterns are used to analyze, create and predict future player buying moves. it also falls under the user retention efforts made by game studios, intended to also reduce customer churn.

  • Prescriptive analytics

Yet another category of predictive analytics, this analysis model combines machine learning algorithms and artificial intelligence to simulate several approaches to possible player behavior patterns. Eventually, such a model suggests the best action to take in order to optimize practices.


This type of data analytics puts the focus on “what should happen.” It differs from the predictive format by diving deeper into the user journey and employing much more data and a broader perspective.

  • Descriptive analytics

This type of data analytics relates to understanding “what we know” and collecting all current known player data information. It often serves as the basis for the above-mentioned two predictive models.


As soon as describing what an organization knows regarding its player’s journey and current buying habits, such data can serve multiple functions such as predictive models, churn prediction efforts as well as customer retention analytics.

  • Consumption analytics

The process of how data insights are leveraged within an organization is termed “Consumption analytics”. Such a process follows the recommendations to act, and then the actions an organization takes.


Later there is a measurement of those actions, leading managers to a “feedback loop” of repeating those steps over and over again. As players keep buying and then commenting online, there is more data to track and analyze constantly.


The types of qualitative churn analytics


Lots of machine learning, artificial intelligence, and natural language processing (the ability of a computer program to understand human languages as they are spoken and written) are employed by managers in order to complete the other side of data, namely the qualitative segment. This side complements the quantitative side of data analytics.


Also called diagnostic analytics, this advanced analytics format tries to answer the question of “why something happened”. So the “what happened” (the numbers side) gains the much desired “why it did”. By having data from both sides of the equation, managers are now able to take fewer risk decisions and more data-driven actions.


The way to figure out “why something happened is by tracking and analyzing the player feedback and voice. It is the task of every team within an organization to figure out the user’s voice and take proactive steps to improve their games and features as well as ensure churn reduction.


The responsibility of each organization’s team can be described in the following examples. By taking care of their tasks and actions, each team also contributes to the overall organizational effort to increase player retention and reduce churn rates.


1. Product

Product people can contribute to the churn rate reduction efforts by analyzing their player feedback and improving the designs, features, and contents of the products they are working on. Insights from player experience tell product people what works and what does not work, with the ability to break it down by category or segment, for a specific fix or improvement.


Once analytics of the player feedback is delivered to the product department, it enables this team to better focus and takes actions based on such insights for an improved user journey. Such improvements can later be measured too, as players react to them, and then the product team can further learn how its fixes were perceived by the users.


By improving products, organizations help increase player retention and decrease churn. By tracking, monitoring, and analyzing feedback from players, product people can rely less on guesswork and focus their efforts on data-driven analytics for more accurate decisions and actions.


2. Marketing

Reducing churn rates is always on the minds of marketing teams, as they consider which steps to take. Brand awareness efforts, public relations initiatives, and paid advertising are some of the actions taken in order to drive in more new players and prevent existing ones from churning.


But as with any effort, all considerations must be based on data and analytics. As players react to various organic and paid campaigns or in-game efforts, either on social media or on special forums, managers can gather lots of insights and learn what customers are thinking about such actions.


And after a manager is able to identify what worked and what didn’t, and what the organization’s various player segments want and need, he can have a better focus when trying to create future campaigns.


3. Player success

A player’s journey does not end with the download of the game, and as it is the responsibility of the player experience team to handle user requests and complaints, this team serves a key role in the organization’s overall effort to avoid player churn and increase their engagement.


There are various machine learning algorithms and AI means to measure player feedback regarding companies’ player experience efforts so that the team can figure out how users are perceiving the treatments they are getting. Such data and analytics, based on tracking and analyzing huge amounts of player conversations, figure out the churn prediction and conclusions that an organization evaluates.


4. The management

Executives dealing with increasing customer retention and churn avoidance must view each team’s effort as well as the total brand strategy and goals in order to maintain growth and profits. They too rely on machine learning analytics and usually take a bird’s eye view in order to improve overall performance.


Once the churn rate is realized, the actionable insights from the data tracked apply to each team’s effort separately but also contribute to the entire organizational churn rate handling and resolving. Player churn is then met with specific actions, relying on player feedback data analysis and intended for long-term better user journey and improved retention.


Affogata player data analytics helps reduce churn


Affogata tracks and monitors feedback data from players, as found all over the open web, as well as internal sources. The real-time data collected is then used to analyze a variety of topics and data segments that assists game studios to make actionable insights with regard to their brand and games.


The “why” completes the “what”


Analytics of data can apply to each department specifically, as the platform is cross-organizational, and each manager can take advantage of it for his own purposes. But the common denominator for each analytics piece is that it always covers the “why” aspects in support of the “what happened” (number figures) in relation to the product.


AI-powered data tools, including machine learning and natural language processing, manage to track real-time huge volumes of feedback from players. Such unstructured data then goes through the process of analysis and structuring, so that companies can figure out what players converse about regarding the games, features, and brand.


As players make comments on many online platforms, including review apps, forums, or social media outlets, their voice forms opinions about the different aspects of a product. As the data is analyzed and processed, the platform is able to create reports and summaries of every key segment and category which is relevant to companies.


Such qualitative analysis, or the data explaining the “why it happened”, helps companies greatly to figure out “what happened”. They can now evaluate their previous decisions and plan their next steps, based on accurate feedback data from their players.


More tools to figure out churn


Dealing with churn is always a delicate and difficult matter. Using the AI-driven player feedback analytics platform to figure out why players are not happy and what led them to churn, is the first step toward fixing the problem.


The digital tools promise fast collection and analysis of the players’ voice, and that puts companies in a position to fix a product’s category, feature, or segment almost immediately. Ultimately, they can handle churn better and minimize its damages.