Setting a user acquisition budget is less about picking a number and more about deciding how fast you can afford to learn and how fast you can afford to scale. A budget that is too small never produces enough data to make decisions. A budget that is too large burns cash before you know what works.

This post walks through how we think about budget planning for mobile games and apps, from the first test to scaled spend.

Start from payback, not from a round number

The most useful budget question is not “how much should we spend?” but “how long can we wait to get it back?”

Your payback window sets how much cash is tied up at any moment. If you spend steadily and players take 60 days to pay back, you are always carrying roughly two months of spend that has not returned yet. Multiply your daily spend by your payback window and you get a rough picture of the working capital you need.

That is why a studio with a fast-paying game can scale harder on the same cash than a studio whose players pay back slowly. For how to set that window and the targets that go with it, see our post on what a good ROAS looks like.

Split the budget into test and scale

We treat budget as two pools with different jobs:

  • Test budget. Small, steady spend used to try new creatives, countries and optimisation goals. Its job is to produce learning, not profit, so judge it on whether the information is useful.
  • Scale budget. Spend behind campaigns that have already cleared your targets. Its job is to deliver volume at a return you can rely on.

Keeping the pools separate protects you from two common mistakes: cutting all experiments when you are scaling, and scaling something that was never properly proven. A standing test budget also feeds the pipeline described in our post on creative fatigue.

Size the test budget by what you need to learn

Test spend has to be large enough to reach a decision. Work backwards from the question:

  1. Pick the signal. For example, D7 ROAS or a retention milestone.
  2. Estimate how many installs you need for that signal to be stable enough to trust.
  3. Multiply by your expected cost per install in the countries you are testing.

If the budget you can afford is below that number, narrow the test instead of spreading it thin. Fewer countries, fewer creatives and a single optimisation goal will give you a cleaner answer than a broad test that never reaches significance. Our soft launch guide goes into what to measure before you commit to scale.

Raise spend in steps

When a campaign clears its targets, resist the urge to double it overnight. Large jumps push the network back into a learning phase and can change who sees your ads, so results often look different after the change than before it.

A safer pattern is to raise budgets in modest steps, wait a few days to see how cohorts behave, and only then step again. If ROAS holds across a couple of steps, continue. If it slips, pause at the last level that worked and look at creatives and targeting before pushing further. Our guide to scaling Applovin campaigns shows how this plays out on a single network.

Spread across networks carefully

Once one network is working, adding others can lift volume and reduce dependence on a single source. Applovin, Mintegral, Unity and Google Ads each reach different audiences and respond to different creatives and goals.

Add a new network the way you add anything else: with a test budget first, a clear target, and enough time to judge on cohort numbers rather than first-day results. Moving too much spend at once makes it impossible to tell which change caused which result.

Keep a buffer and a review rhythm

Two habits keep budgets healthy:

  • Hold a buffer. Leave room for months when ad prices rise or a strong campaign fades, so you are not forced to cut at the worst moment.
  • Review on a fixed rhythm. Check numbers weekly to make tactical moves, and revisit the overall budget every quarter against blended ROAS and payback, instead of reacting to each swing.

The three-phase approach in our growth roadmap shows how budgets typically grow as a game moves from validation to scale.

The short version

  • Work back from your payback window to see how much cash you need.
  • Keep separate test and scale budgets so learning never stops.
  • Size tests by the data you need, and narrow them if budget is tight.
  • Raise spend in steps and judge each step on cohort results.
  • Hold a buffer and review on a fixed weekly and quarterly rhythm.

If you want help shaping a budget for your next launch or for scaling an existing game, book a free call and we will plan it with you.