The question of whether to trust data or intuition sounds especially acute in an industry where even small fluctuations in behavioral metrics cost companies a lot of money. And to answer it honestly, it is worth starting not with philosophy but with numbers that show how difficult it is to retain a user today and how misleading obvious conclusions can be.
What Player Retention Data Actually Shows
According to the GameAnalytics 2024 report, the average D1 retention for the top 25% of mobile games ranges from 26.48% to 27.69%, which is worse than the 2023 results (28-29%). For the weakest projects, this figure drops to 10-11.5%. The picture with long-term retention is even more telling: the median D7 retention across all projects ranges from 3.42% to 3.94%, and 75% of projects have D28 retention below 3%.
Other sources give somewhat different figures; for example, one report puts D1 retention at 28%, D7 at 13% and D30 at only 6%. Such a discrepancy between sources is an important lesson in itself: even objective industry benchmarks depend on methodology, sample and market segment.
Interestingly, the overall picture of player engagement looks the opposite of the alarming retention figures (mobile gamers spent a total of more than 444 billion hours in games during 2025, and the number of gaming sessions grew by 12% year over year). In other words, the audience is shrinking, but those who stay spend more time in the product.
Why Raw Data Misleads Teams
If a product team looked only at overall D1 retention and compared it with the industry average, it could draw the wrong conclusion about the quality of its product. GameAnalytics data shows that iOS demonstrates higher D1 retention (31-33% for top projects) compared to Android (25-27%), meaning the platform alone shifts the benchmark by several percentage points.
The product's genre matters just as much. According to the report, the best mid-term and long-term retention rates are shown by board games, card games, puzzles and casual games, while other categories by their nature have a lower baseline.
An additional complication is how differently the same metric looks depending on the benchmark source. For example, for D7 retention, one analytical report sets the industry average at 15-20% and treats anything below 10% as a signal of problems with the product's core loop, while the median values from another report look significantly lower.
This means that whether a result is good or bad is a question of which benchmark you are comparing against, not an absolute truth.
Where Data Ends and Expert Interpretation Begins
This is exactly where the intuition of an experienced product expert plays the role not of an alternative to data, but of a tool for reading it correctly. Data can show precisely that a product's D7 retention has dropped by 2 percentage points. But only a contextual understanding of the market makes it possible to assess whether this is a critical problem or a natural fluctuation, given that for the weakest 25% of projects D7 retention can drop as low as 1.5%.
The same goes for session metrics, when data records that on average a player has 4 sessions per day, while midcore projects have the highest number of sessions – 6-7 per day. Without understanding the specifics of the product's genre, a team may mistakenly interpret a high session frequency as an unconditionally positive signal, whereas for a specific type of product it may indicate excessively aggressive user return mechanics.
Conclusion
Data in iGaming is available today in unprecedented volume. But it is precisely this volume that creates the need for expertise that makes it possible to correctly segment, contextualize and interpret the numbers. Data without industry experience is just a large array of numbers. Intuition without data is unfounded assumptions. Product decisions that work are born at the intersection of these two approaches: numbers set the boundaries of reality, and an expert understanding of the market turns them into the right conclusion.
Why You Can Trust Us
We rely on open industry reports, including GameAnalytics research, and always indicate where the figures come from. Where sources differ, we show all the options rather than choosing the convenient one.
How We Verify Information
We check every indicator against its primary source and verify the publication date and sampling methodology. If a figure has no clear explanation, we do not use it as an argument.
Editorial Principles
We write concisely and without promises of quick earnings. We separate facts, expert assessments and our own assumptions so that the reader can see where data ends and opinion begins.
This material is analytical in nature and is not a recommendation to participate in gambling. Responsible gambling: only play with money you can afford to lose.