AppLovin ads reporting works best when each dashboard is built to answer one decision: whether to raise or cut budget, which creative to keep, which countries or placements to exclude, and whether the users you acquired pay back. The AppLovin dashboard shows delivery and cost; your mobile measurement partner shows what those users did afterwards. Useful reporting joins the two.
AppLovin’s user acquisition products, now centred on AppLovin’s AXON-driven campaigns, buy traffic across a large network of mobile apps, mostly games. Because the inventory is in-app and largely automated, reporting is the main way you understand what is happening.
What does each data source actually tell you?
Start by being clear about which number comes from where. Mixing sources without labels is the most common reason teams argue over dashboards.
- AppLovin dashboard: impressions, clicks, installs as AppLovin attributes them, spend, CPI, and the campaign’s ROAS or retention metrics where postbacks are configured
- Mobile measurement partner (AppsFlyer, Adjust, Singular and others): installs and in-app events attributed under your chosen rules, plus cohort revenue
- Your own analytics or data warehouse: in-app purchase and ad revenue at user level, subscriptions, retention by cohort
- SKAdNetwork and AdAttributionKit data for iOS: aggregated, delayed and limited, but the only signal for users who did not consent to tracking
Each source counts installs slightly differently. Do not expect them to match exactly. Decide which source is the “source of truth” for each metric and write that on the dashboard.
Which question should the first dashboard answer?
The first question is almost always “are we making money on this spend?”. For a game or subscription app, that means cohort ROAS: revenue generated by users acquired on a given day, divided by the spend that acquired them, measured at fixed ages such as day 0, day 7, day 30 and day 90.
A cohort ROAS chart with one line per campaign and fixed day markers is more useful than any total. A campaign at 20 percent D7 ROAS may be excellent if your apps usually reach payback by D90, and poor if most revenue normally arrives in the first week. Use your own historical payback curve to set targets.
How do you build a creative report that helps?
AppLovin campaigns rely heavily on creatives: playable ads, video and end cards. A creative report should show, for each creative set, spend share, install rate (IPM, installs per thousand impressions), and the downstream ROAS of users who installed from it.
The trap is that a creative can have a great IPM and poor ROAS. A misleading playable that shows a different game attracts the wrong users. Always pair top-of-funnel metrics with revenue metrics in the same row.
Columns for a creative report
- Creative name and format (video, playable, end card)
- Impressions and spend share in the period
- IPM and CPI
- D7 ROAS and D7 retention for users from that creative
- Days since launch, to spot creative fatigue
Which breakdowns are worth showing?
Country is the first breakdown, because both cost and user value differ widely by market. Placement or publisher app is the second; some apps send many installs with weak retention. Device and OS version can matter for performance-heavy games.
Avoid breakdowns that create tiny segments. A table with 500 publisher apps and a handful of installs each mostly shows noise. Set a minimum spend or install threshold before a row appears, for example 100 installs, so the dashboard highlights segments where a decision is possible.
How do you handle iOS privacy limits in reports?
On iOS, many users do not allow tracking. Their installs are attributed through SKAdNetwork or AdAttributionKit, which sends conversion values with a delay and limited detail. Your MMP decodes these values according to the schema you set, such as revenue buckets in the first 24 to 48 hours.
Show iOS and Android separately. Label iOS metrics that rely on modelled or aggregated data. When comparing campaigns on iOS, use the same postback window and conversion schema, or the comparison will not hold. Revisiting the conversion value schema every few months, as your monetisation changes, keeps the early signal useful.
What does a sensible dashboard set look like?
Three dashboards cover most needs for a growth team. More than that usually means no one checks them.
- Daily health: spend, installs, CPI, D0 ROAS by campaign and country, with alerts for spend spikes
- Weekly performance: cohort ROAS curves at D7 and D30, retention, creative table
- Monthly payback: D90 and longer cohort ROAS, LTV estimates, budget plan for next month
Tools vary. Many teams use their MMP’s built-in dashboards for daily checks and Looker Studio, Tableau or a warehouse-based tool for weekly and monthly views. The AppLovin reporting API lets you pull spend and campaign data automatically into the same place.
Common reporting mistakes
Comparing CPI across countries without revenue is the classic error: a low CPI market can produce users who never pay. A second is changing attribution windows mid-month, which makes before-and-after comparisons meaningless. A third is ignoring ad revenue for apps that monetise with ads; for many games it is a large part of the value of a user, and leaving it out understates ROAS.
Finally, watch the time zone. AppLovin, your MMP and your warehouse may each report in a different time zone. A cohort that is “Monday” in one source may be split across two days in another.
Setting alerts instead of watching charts
Nobody looks at dashboards every hour, so automate the checks that matter. Most BI tools and MMPs can send an alert when a metric crosses a threshold. Useful alerts for AppLovin spend include a daily spend more than 30 percent above plan, a D0 ROAS falling below a floor you set from history, and a campaign whose installs drop to near zero, which often signals a tracking break rather than a real collapse.
Keep alerts few and specific. If the team receives ten alerts a day, they stop reading them. Review the alert list each quarter and remove any that never led to an action.
- Spend spike alert per campaign
- D0 ROAS floor alert per country
- Install drop alert to catch tracking breaks
- Creative fatigue alert when IPM falls steadily for a top creative
Making reports readable for non-specialists
Finance and leadership rarely need creative tables. They need to know how much was spent, what came back and when the rest is expected. Build a one-page monthly summary with spend, cohort revenue to date, projected payback date, and a short note on changes. Use the same definitions every month, and add a small glossary for terms such as ROAS, IPM and LTV so the report reads the same way for everyone.
Checking data quality before trusting a dashboard
Before any dashboard goes live, compare one week of spend from the AppLovin reporting API with the invoice or billing view, and one week of installs from your MMP with the store consoles. Small gaps are normal; large ones point to missing postbacks, wrong currency settings or a filter applied in one source and not the other. Note the accepted gap on the dashboard so the team knows what “normal” looks like.
Where to start this week
Write down the three decisions your team makes most often about AppLovin spend. For each one, list the metric and the source that answers it. Then build the daily health dashboard with only those metrics, connecting the AppLovin reporting API and your MMP data in one view.
Check the time zones and attribution windows in every source and align them. Once the daily dashboard is trusted, add the weekly cohort view and the creative table.
Keep reading: Applovin Ads · Advertising