Applovin is the lead network in our mix, followed by Google Ads, Mintegral and Unity. It did not start that way. Like most studios, we began with small tests, and the work of moving from “this campaign looks fine” to “this is where most of our volume comes from” is different from the work of launching.

This post covers what changes when Applovin goes from a test channel to your main network, and how we approach scaling without giving up the returns that made it worth scaling in the first place.

Small tests and a main network are different jobs

A test campaign has one question to answer: can this game earn back its cost on this network? You run a few creatives, one or two countries, and a modest budget, and you look at early cohorts.

A main network has a harder question: can it keep earning back its cost as spend grows, for months, while creatives wear out and competition shifts? Scaling is mostly about protecting that answer.

Three things tend to change:

  • Volume becomes a data problem. The campaign needs enough conversions of your target event every day for the system to learn. Applovin’s own guidance suggests a daily budget that produces at least 15 to 20 conversions of the target event, and in our experience a campaign starved below that moves around too much to read.
  • Creatives become the main lever. At small spend, one decent video can carry a campaign. At larger spend, fatigue arrives faster and you need a steady supply of new assets.
  • Reporting needs to be trusted. A small error in revenue attribution is noise at test scale. At full scale it decides whether you keep spending.

Get the foundations right before you raise anything

Most scaling problems are foundation problems that were hidden by low spend.

Check tracking and postbacks. Applovin warns that if it does not receive conversions, future campaigns suffer. Confirm that install and in-app events arrive, and that ad revenue from your mediation layer is passed through your measurement partner, so each source is credited with what it actually earned.

Choose the optimisation goal that fits your revenue. Applovin lets you optimise towards different return goals depending on whether your game earns from ads, purchases or both. Our hybrid games earn roughly 80 percent from ads and 20 percent from in-app purchases, so a blended view is the honest one. If you optimise on purchases alone you ignore most of the value, and if you optimise on ads alone you ignore your payers. Pick the goal that matches how your game really earns, and see our post on monetisation design if that mix is still unsettled.

Be realistic with targets. A target that is too ambitious can stop a campaign from spending at all. Start from your break-even point, not from the number you wish you had, and tighten later.

How we raise spend

Once a campaign is clearing its goal, the temptation is to double the budget and walk away. We do not do that.

Raise in steps. Increase budget in measured steps and give each step enough days to produce readable data before the next one. Big jumps tend to disturb delivery and make it hard to tell whether a dip came from the change or from something else.

Use one global budget where you can. Applovin recommends a single budget across countries so spend can flow to the markets that perform best, rather than splitting it by hand. Target the countries you actually support, and let the system do the allocation.

Raise targets only when the data is stable. If a campaign is spending comfortably and returns are steady, you can tighten the goal to protect margin. If you tighten at the same moment you raise budget, you will not know which change did what.

Read cohorts, not just the first week. Early return numbers are useful but they are an estimate. We watch early cohorts for speed and later cohorts for truth, and we do not call a scaling step a success until the later numbers confirm it. Our UA strategy and process guide covers how KPIs like these fit into the wider setup.

Creatives carry the scale

Applovin inventory is mostly video, so video should be the core of what you send. The platform also supports playable and static formats, and a mix gives the system more combinations to try.

A practical routine looks like this:

  1. Keep a steady flow of new assets going in every week, not in occasional big batches.
  2. Test new hooks against your current best performer, rather than replacing it blindly.
  3. Retire creatives when their return falls below your break-even line for long enough to be sure it is not noise.
  4. Localise for non-English markets with translated ads or subtitles.

Playables deserve their own mention. They filter for players who have already tried the core loop, which often shows up as better quality downstream. We cover how to make them in our playable ads guide.

Keep testing budget separate

When Applovin becomes your largest network, there is pressure to put all of your budget behind what already works. Resist it. Reserve a slice for new creatives, new countries and new goals, and judge that slice by learning, not by its immediate return. If you do not, your winners age out and you have nothing ready to replace them.

Know when to hold

Scaling is not a straight line. Sometimes a campaign plateaus because the audience is saturated for that creative set, sometimes because the target is set too tight, and sometimes because the game itself has changed. When returns slip, change one thing at a time: first creatives, then targets, then budget. Pulling all three at once makes the cause impossible to find.

We also judge networks on longer windows than a single month. A strong month can hide a weak quarter and a weak month can hide a strong one, so we rely on rolling averages over time. Our blended ROAS has stayed above break-even across 40 months of aggregated results, and that steadiness comes from boring habits like these rather than from any one clever trick.

Where Applovin sits in the bigger plan

Applovin works best as one part of a portfolio, not the whole of it. The three-phase roadmap in our growth roadmap shows why: Google Ads and Mintegral validate the economics first, Unity widens the mix, and Applovin joins once the creative pipeline and attribution can support it. Diversifying also protects you when one network has a bad stretch.

If you are weighing whether to build this capability yourself or bring in help, our post on hiring a UA consultant versus in-house lays out the trade-offs.

The short version

  • Fix tracking and revenue attribution before you raise budget.
  • Choose the optimisation goal that matches how your game earns.
  • Raise spend in steps and change one variable at a time.
  • Feed the network a steady stream of new video and playable creatives.
  • Hold back a testing budget, and judge networks over quarters, not weeks.

Scaling Applovin is mostly discipline: protect the signal, keep the creative pipeline full, and be patient enough to let the later cohorts tell you the truth.