Most studios should plan for three to six new playables a month to start, then scale toward weekly releases as spend grows. AppAgent suggests three to six a month for Meta. At higher spend, practitioners recommend two to three new concepts plus six to ten variations a week per title. The top 20 grossing games ship a median of 64 new playables a month. Whatever the number, split it roughly 70% variations of winners and 30% new concepts, and give each playable enough data before you judge it: about 50 conversions in four days, or 100 in seven.
- Starting point: three to six new playables a month. Scale from there.
- Top-grossing games ship a median of 64 a month. That's the ceiling, not the target.
- Split output about 70% variations, 30% new concepts. Only about 5% of creatives become winners.
- Your budget sets the pace. Each test needs about 50 conversions in four days to read.
- Genie Playable makes each variation from the last version, so volume stops being a production problem.
How many playables do top studios run?
| Who | Volume | Source |
|---|---|---|
| Top 20 grossing games | Median of 64 new playables a month | Sett, Q4 2025 |
| Whiteout Survival | 42 playables on AppLovin in one month: 15 concepts, 27 variations | Sett, October 2025 |
| Leading games on AppLovin | 30 to 35 new creatives a week, 10 to 15 of them playables | M&C Saatchi Performance |
| Socialpoint | Five to eight playables a month across three games | AppAgent |
| Typical recommendation for Meta | Three to six new playables a month | AppAgent |
How many should I test at my stage?
| Stage | New playables | Mix |
|---|---|---|
| First campaign | One or two to launch, one new one by week two | Same hook as your best video |
| Scaling on one network | Three to six a month | One or two concepts, the rest variations |
| Multi-network, steady spend | Two to three a week | About 70% variations, 30% concepts |
| Top-grossing scale | Ten or more a week | Structured variation programmes around each winner |
A mid-size studio told us they run around 60 playables and 90 videos on one title and still feel they need far more creative volume. Volume grows faster than most teams plan for. Starting out? See Your first playable for your first AppLovin campaign.
Why the 70/30 split?
Because winners are rare and worth stretching. Playio estimates only about 5% of tested creatives produce meaningful results. Once you have one, variations on it are cheaper and more likely to work than a new idea. Sett found one studio putting 70% to 75% of its playables into deliberate variations of a winning concept and 25% to 30% into new directions. Whiteout Survival gave one winning concept 11 variations in a single month.
The 30% matters too. Without new concepts, every variation eventually fatigues and there's nothing to replace the winner.
How much data does each playable need?
Enough to separate signal from noise. A common rule of thumb is about 50 conversions per creative in four days, or 100 in seven (AppAgent). On AppLovin, practitioners report three to seven days of learning per creative, and advise against pausing by hand too early.
That rule is what links volume to budget. If each test needs 50 conversions and your cost per conversion is $20, each playable needs about $1,000 of spend to read. Ten tests a month means $10,000 of test budget before scaling anything. Run more playables than your budget can read, and you learn nothing from any of them.
How do I keep up with the volume?
- Keep a backlog per winner: three hooks, two difficulty levels, two end cards, the next event skin.
- Tie skins to your LiveOps calendar. Every new event is a ready-made variation.
- Change one thing per version, so each result teaches you something. See how to iterate a winning playable.
- Log results by version, so nobody retests what already failed.
- Fix the bottleneck, not the ideas. The constraint is how long each version takes to build.
"The idea is the easiest part. We have too many ideas."
Where does Genie fit?
Genie Playable builds playables from your game (your APK or HTML build plus the assets, narrative and live events in Genie), from a format that's already working, or from a brief. When you need the next variation, your team changes one thing on the last version and exports it again for Meta, Google, TikTok, AppLovin, Unity Ads, Mintegral, Liftoff, Moloco, Snapchat and Digital Turbine. Genie Core keeps every version on brand, so the fiftieth still looks like your game. Try it free on your game.
FAQ
How many playable ads should I test per month?+
Three to six to start is a common recommendation. At higher spend, two to three concepts plus six to ten variations a week per title.
How many playables do top mobile games run?+
The top 20 grossing games ship a median of 64 new playables a month.
What is the 70/30 rule for playable ads?+
About 70% of output goes to variations of winning concepts and about 30% to new concepts.
How long should I test a playable before judging it?+
Until it has about 50 conversions in four days or 100 in seven. On AppLovin, expect three to seven days of learning.
How much budget do I need to test playables?+
Multiply the conversions each test needs by your cost per conversion. At 50 conversions and $20 each, that's about $1,000 per playable.