A soft launch is not a smaller version of your global launch. It is a series of questions, answered with real players and real spend, before you commit the budget that cannot be taken back. This playbook walks through how to run a soft launch for a mobile game or app, what to measure, when to say go, and how to move from a few test markets to a global launch without breaking your unit economics.

We also cover the questions studios ask most often: how to choose experts or an agency for a soft launch, which networks suit global launches across several regions, what changes in Asia Pacific, and how Unity Ads and ad monetisation fit into the plan.

What a soft launch is actually for

A soft launch releases your game in a limited set of markets so that you can learn cheaply. The useful way to frame it is four questions, asked in order:

  1. Does the core loop hold players on day 1?
  2. Does retention keep holding over the following weeks?
  3. Does projected lifetime value cover what you would pay for installs at scale?
  4. Do those economics survive when you raise spend?

Agree a go, extend or kill threshold for each question before the data arrives. Without thresholds, every result gets explained away, and a weak game drifts into an expensive global launch. For a deeper look at the structure of a soft launch, see our earlier guide to soft launching mobile games.

A soft launch is also not the same as a marketing campaign. If the game is unstable, or the first session is confusing, buying more traffic just measures the problem more expensively. Check crash rate and load time on target devices before any paid install.

Choosing soft launch markets

Pick markets for reasons, not habit. A common approach is to use two kinds:

  • A cost-efficient market for retention reads and early creative learning, where installs are cheap enough to gather data quickly.
  • A market that resembles your eventual paying audience, even if installs cost more there, so that your monetisation reads are not misleading.

A single market can validate retention but mislead you on revenue, because spending habits and ad value differ by country. Keep the number of markets small so each one gets enough volume to read. Check current install costs for your genre and platform from a named, recent source, since regional prices move.

If your game will monetise through both ads and purchases, remember that the ad side behaves differently by region as well. A market that looks weak for in-app purchases can still be strong for ad revenue, and the reverse. Test both income lines in your soft launch markets, not just one.

What to measure and how to read it

Read results by cohort, by channel and by creative, not as one blended number.

  • Stability: crashes and start-up time before you buy traffic.
  • Day 1 retention: a quick read on whether the first session works. If it is weak, stop buying and fix onboarding.
  • Day 7 and later retention: whether players find enough depth to stay. Compare against peers in your genre.
  • Unit economics: projected lifetime value against the cost per install you expect at scale. The ratio matters more than any single rate. Our guide to LTV for mobile games explains how to build that projection from cohorts.
  • Scale response: how cost per install moves when you raise daily spend.

Keep test cells large enough to read. Splitting a small budget across many networks, creatives and countries gives you noise, not insight.

The go, extend or kill decision

When the four questions are answered, make the call and write it down.

Go when retention is in a healthy range for the genre, it holds over time, projected value supports the cost you would pay in target markets, and cost holds up as spend rises.

Extend when the picture is promising but unclear. Change one thing, such as onboarding or a reward curve, then re-test with fresh cohorts.

Kill or rework when signals stay below your thresholds after reasonable fixes. Delaying a kill decision costs runway, and runway is the one thing a studio cannot print.

Best UA networks for global launches across multiple geos

No single network wins every country, which is why a global launch needs a mix. Our own aggregated results come from running Applovin as the lead network, with Google Ads, Mintegral and Unity alongside it. That spread exists because each network reaches players in different contexts, and because relying on one source makes you fragile when its quality dips.

Some practical points when you plan the mix:

  • Start with the network that gave you the best cost per paying or engaged player in soft launch, and add others in sequence rather than all at once.
  • Give each network a fair test with enough volume, then compare them on the same metric. Our post on how to structure a multi-network test explains how to do this without fooling yourself.
  • Use our comparison of the main game networks as a starting point, and check Unity Ads against Google Ads if those two are your shortlist.
  • Expect network performance to differ by country. A network that is strong in one region may need a different creative or bid approach in another.

Launching in Asia Pacific markets

Asia Pacific is not one market. Languages, device mixes, store availability, payment habits and player tastes vary a great deal between countries, and several of them have their own store ecosystems and local publishers.

Treat each country as its own soft launch question. Localise the store page, the first session and the creatives, not just the text. Check which of your networks has real reach in the countries you target, since coverage and demand differ by region, and confirm it with a small test rather than assuming it.

Using Unity Ads and ad monetisation in the ramp

If your game earns from ads, soft launch is the time to test both sides of the ad business: acquisition and monetisation. Unity offers both UA campaigns and an ad monetisation product, so some studios run its monetisation alongside its UA. Whatever stack you choose, test it in soft launch.

A simple monetisation playbook for the move to global:

  • Start with a small set of ad placements that do not damage day 1 retention, and add more only if retention holds.
  • Watch how ad frequency affects retention and in-app purchase conversion together, not separately.
  • Compare ad revenue per player by country, since it will vary, and fold that into your cost limits for each market.

Then feed that ad income into your payback calculation, so that your cost ceiling for each country reflects what players are really worth there.

A staged ramp to global

Scaling is where soft launch discipline pays off. A staged approach looks like this:

  1. Confirm the gate. All four questions answered with yes.
  2. Raise spend in the best soft launch market first. Watch cost per install and early retention as volume grows. If cost climbs sharply, the channel may be saturating, so test a new one before you commit more.
  3. Add networks one at a time, each with its own creatives.
  4. Add countries in waves, with localised assets, and keep a baseline cohort to compare.
  5. Review weekly, with kill rules for any country or network that falls below your payback target.

Choosing experts or an agency for a soft launch

If you want outside help, look for people who have run soft launches to a go or kill decision, not just spent budget. Ask:

  • How do they set thresholds before data arrives?
  • Can they show how they read cohorts, rather than blended averages?
  • Will they tell you when the game is not ready, even if it costs them spend?
  • Which networks have they run in the regions you plan to launch in?
  • Who actually runs the account day to day?

Key takeaways

Treat soft launch as a set of questions with agreed thresholds, choose a small number of markets for clear reasons, and read everything by cohort. When the answer is go, ramp in stages: one market, one network, then more, with kill rules at every step. Test ad monetisation and in-app purchases together, and plan each region as its own case.

If you are heading toward launch and want a second opinion on your soft launch plan or your network mix, get in touch. We have run campaigns across Applovin, Google Ads, Mintegral and Unity, and we are happy to look at your numbers.