A mobile game UA audit is a structured review of how you buy players, from tracking and campaign setup to creatives, bids and payback. Most accounts that struggle to scale do not have one big problem. They have four or five small ones that hide each other, and an audit is how you find them.

This checklist has 20 points, grouped into five areas. Work through it in order, because a problem early in the list, such as broken tracking, makes every later number unreliable. It also covers the second question we hear most: what should you be looking at in UA spend on a day to day basis?

How to run a mobile game UA audit

Set aside a few hours and pull the last 30 to 90 days of data. Open your attribution tool, each ad network dashboard and your revenue reporting side by side. Write down what you find rather than fixing as you go. If you change things mid-audit, you lose the baseline.

Then sort every finding into three piles: fix now, test next, and leave alone. Not every gap is worth closing. The aim is to remove the leaks that cost you the most, not to chase a perfect account.

Area 1: Tracking and data

Everything else depends on this. If the data is wrong, good decisions look bad and bad decisions look good.

  1. Install and event counts match across sources. Compare installs in the attribution tool with what each network reports. Small gaps are normal. Large or growing gaps point to a setup problem.
  2. Key events fire correctly. Check tutorial completion, level milestones, purchases and ad impressions. Test them on a real device, on both platforms.
  3. Ad revenue is sent back to the attribution tool. For hybrid games, ad revenue is often the larger share of LTV. If it is missing, your ROAS is understated and you may switch off campaigns that work.
  4. Purchase values and currencies are correct. A wrong currency or a missing refund adjustment quietly distorts ROAS.
  5. Privacy-limited data is understood. On iOS, much of your data is aggregated and delayed. Make sure the team knows which numbers are modelled and which are observed. Our guide to UA after IDFA explains how to read this.

Area 2: Campaign structure

A tidy structure makes it possible to learn from tests. A messy one makes every result ambiguous.

  1. Campaigns are separated by what actually differs. Platform, country group and optimisation goal should each have their own campaign. Mixing them hides which one is working.
  2. Budgets match the opportunity. Look for campaigns that hit their cap every day while performing well, and for campaigns spending freely with weak results. Reallocation is often the cheapest win. See how to plan a UA budget for a simple method.
  3. Optimisation goals fit the game. Install-based goals suit early testing. Event-based and return-based goals suit scaling, once you have enough signal. Check that each campaign uses the goal that matches its stage.
  4. Network mix has a reason. Each network should have a role. In our own mix, Applovin is the lead network, followed by Google Ads, Mintegral and Unity. If a network has spent for weeks without a clear job or a clear result, decide whether to test it properly or pause it.
  5. Learning phases are respected. Frequent edits to bids, budgets and creatives reset learning. Check the change history for campaigns that never settle.

Area 3: Creatives

Creative is usually the biggest lever and the most neglected part of an audit.

  1. There is a steady flow of new concepts. Count how many genuinely new ideas, not minor edits, you tested in the last month. If the answer is close to zero, fatigue is coming. Our post on creative fatigue shows the warning signs.
  2. Formats cover the placements you buy. Check you have video, playable and static versions where each network supports them, in the right sizes and lengths.
  3. Winners are identified with more than one metric. A creative with a high install rate and weak day 1 retention is not a winner. Compare install rate, retention and return together.
  4. Losers are retired. Old creatives that keep taking a small share of spend add noise and drag down averages. Set a kill rule and apply it.
  5. Ad promises match the game. If the ad shows something the first minute does not deliver, you pay for installs that churn.

Area 4: Bids, scaling and quality

This area checks whether you are buying the right players at the right price.

  1. Retention is compared by network, country and creative. A cheap source with poor day 1 and day 7 retention is expensive in practice.
  2. Payback is tracked against a target window. Pick a window, such as the period you can afford to wait for your money back, and compare each cohort with it. The metrics guide explains how CPI, retention, LTV and payback connect.
  3. Scaling is gradual. Look at how budgets changed. Large jumps often cause a drop in quality that is blamed on the network when the cause was the increase.

Area 5: Process and reporting

The last area is about habits. A good account with no routine drifts.

  1. A daily check exists and is short. See the next section.
  2. Decisions are written down. Keep a simple log of what changed, why, and what happened. It saves you from repeating tests and makes handovers easy.

UA spend on a day to day basis: what to check

You do not need to look at everything every day. A short routine catches most problems early.

Every day:

  • Spend against budget for each campaign, looking for sudden stops or runaway delivery.
  • Install volume and CPI against the recent trend, not against an arbitrary target.
  • Any network error, rejected creative or disapproved campaign.
  • Early signals such as day 0 or day 1 retention for newer cohorts.

Every week:

  • Retention, ROAS and payback by cohort, with enough maturity to mean something.
  • Creative results and the next batch to test.
  • Budget moves between campaigns and networks.

Every month:

  • A review of the full checklist above, plus a look at your assumptions about LTV. Our guide to LTV for mobile games shows how to check forecasts against real cohorts.

Daily checks are for spotting problems. Weekly and monthly reviews are for making decisions. Reacting to one day of noisy data is one of the most common ways teams damage a healthy account.

What to do with the findings

Start with tracking, because fixes there improve every other decision. Then handle anything that is clearly wasting spend, such as campaigns with weak quality that still receive budget. Next, set up the creative testing routine, and finally the reporting habits.

If you find more than you can fix in a month, pick the three items with the largest likely effect on payback and leave the rest on the list. A short, finished list beats a long one that nobody works through.

An audit is not a one-off. Repeat it every quarter, or whenever performance changes without an obvious reason. Accounts that stay healthy tend to be the ones that are checked on a schedule, not just when something breaks.