In Adjust, cost per install (CPI) is calculated by dividing your total ad spend for a campaign by the number of installs Adjust attributes to that campaign. The result gives you the actual cost for each verified install, based on Adjust’s attribution data rather than the raw numbers reported by ad networks. Because Adjust applies its own attribution logic, the CPI you see in Adjust will often differ from what your ad platforms report directly. The sections below explain exactly how each part of this calculation works and how to use it to make smarter campaign decisions.
What does Adjust count as an app install?
Adjust counts an app install when a user opens the app for the first time after downloading it. This is called a “first open” or “first session,” and it is the event that triggers Adjust’s attribution process. A download that never results in an app launch is not counted as an install in Adjust’s reporting.
This distinction matters because app store download counts include users who download an app but never open it. Adjust filters those out, which means your install numbers in Adjust will typically be lower than the raw download figures you see in App Store Connect or Google Play Console. The result is a cleaner, more accurate view of users who actually engaged with your app, even if only once.
Adjust also applies a reattribution window, so if a lapsed user reinstalls your app after a defined period, that can be counted as a new install depending on your configuration. It is worth reviewing your reattribution settings to make sure your install counts reflect what you actually want to measure.
How does Adjust attribute installs to ad campaigns?
Adjust attributes an install to a campaign by matching the install event to a click or impression that occurred within a defined attribution window. When a user clicks an ad, Adjust records a timestamped click fingerprint. When that user opens the app for the first time, Adjust checks whether a matching click or impression exists within the window, and if it does, the install is attributed to that campaign.
Adjust uses a priority-based attribution model. Click-based attribution takes precedence over impression-based (view-through) attribution, and the most recent qualifying click wins when multiple touchpoints exist. This last-click model is the default, though Adjust does support probabilistic matching for environments where device-level identifiers are unavailable, such as iOS after ATT opt-out.
The attribution window length is configurable. A common setup is a seven-day click window and a one-day impression window, but these can be adjusted per network. Shorter windows reduce the chance of attributing organic installs to paid campaigns; longer windows capture more installs but risk over-crediting ad spend. The window you choose directly affects your attributed install count and therefore your calculated CPI.
How is cost per install calculated in Adjust?
Cost per install in Adjust is calculated by dividing total ad spend by the number of installs Adjust attributes to a given campaign, channel, or time period. The formula is straightforward: CPI = Total Spend divided by Attributed Installs. The complexity lies in how spend data enters Adjust and how installs are counted, both of which affect the final figure.
Ad spend data reaches Adjust through cost integrations with ad networks. Platforms like Google, Meta, and Apple Search Ads push spend data directly into Adjust via API, allowing Adjust to combine spend and attribution data in one place. If a network does not have a cost integration, you can import spend manually, but any gap in cost data will make your CPI calculation incomplete or inaccurate.
Once both data points are present, Adjust calculates CPI at whatever granularity you need: by campaign, ad group, creative, country, or platform. This level of detail is what makes Adjust useful for optimization. Rather than seeing a blended average CPI across all channels, you can identify exactly which campaign or creative is driving installs at the most efficient cost.
Why does CPI differ between Adjust and ad network dashboards?
CPI differs between Adjust and ad network dashboards primarily because ad networks attribute installs using their own tracking logic, which almost always credits more installs to their platform than Adjust does. This discrepancy is normal and expected. It exists because ad networks have a commercial incentive to claim as many installs as possible, while Adjust applies neutral, cross-channel deduplication.
The most common cause of the gap is multi-touch attribution. If a user clicks an ad on Meta and then clicks an ad on Google before installing, both networks will claim the install. Adjust, using last-click attribution, will assign the install to Google only. Meta’s dashboard will still show that install as a conversion, inflating its reported install count and deflating its reported CPI.
Other contributing factors include differences in attribution window length, the timing of when data is processed, and how each platform handles view-through attribution. Some networks count impression-based installs aggressively, while Adjust may apply a stricter or shorter view-through window. When you are comparing CPI figures, always clarify which attribution source the number comes from before drawing conclusions about channel performance.
What factors cause CPI to fluctuate in Adjust reporting?
CPI fluctuates in Adjust reporting because both components of the calculation, spend and attributed installs, change independently based on campaign activity, audience conditions, and platform dynamics. A rise in CPI does not always mean your campaigns are performing worse; it can reflect changes in targeting, competition, or the quality of traffic you are buying.
Common factors that push CPI up include increased competition for ad inventory (which raises CPMs and CPCs), narrowing your audience targeting, moving into higher-cost markets or demographics, and creative fatigue where click-through rates decline over time. Seasonal periods like Q4 or major shopping events consistently drive up acquisition costs across most verticals.
CPI can drop when you introduce high-performing creatives that improve click-through and conversion rates, when you shift budget toward lower-cost channels or geographies, or when your app store conversion rate improves and more clicks result in installs. Because Adjust separates spend from attributed installs, you can diagnose which side of the equation is driving the change, something that is much harder to do when relying on individual network dashboards alone.
How can CPI data in Adjust be used to optimize campaigns?
CPI data in Adjust helps you optimize campaigns by giving you a single, deduplicated view of acquisition cost across all channels, so you can compare performance on equal terms and shift budget toward what actually converts. Rather than relying on each network’s self-reported numbers, you work from one consistent source of truth. This is the practical answer to the problem of having too many marketing dashboards showing conflicting results.
Start by segmenting CPI by channel, campaign, and creative. This quickly surfaces which combinations are delivering installs efficiently and which are draining budget. Channels that look effective in their own dashboards sometimes look very different once Adjust removes duplicate attributions. Knowing your best ad channel for app installs based on attributed data, rather than claimed data, changes where you invest.
CPI alone does not tell the full story. Pair it with downstream metrics like cost per paying user, retention rates, and in-app conversion events to understand real app ROI. A channel with a low CPI but poor post-install engagement is less valuable than one with a higher CPI but strong conversion to paying users. Adjust lets you connect these dots by linking install attribution to in-app event tracking, giving you a complete picture of which spend is genuinely driving growth.
For teams managing multiple platforms and struggling to get a clear view across campaigns, this is exactly the kind of challenge we help solve. Our app growth stack services are built around making attribution, spend data, and performance reporting work together so you can make faster, better-informed decisions. If you want to talk through your current setup and where the gaps might be, request a free consultation and one of our specialists will take a look with you.
Frequently Asked Questions
What is a good CPI benchmark for mobile apps, and how do I know if my CPI in Adjust is too high?
CPI benchmarks vary significantly by vertical, platform, and geography, so there is no universal u0022goodu0022 number. A gaming app on Android in Southeast Asia will have a very different CPI than a fintech app on iOS in the US. Rather than chasing an industry average, benchmark your CPI against your own unit economics: if your CPI is lower than the lifetime value (LTV) of an acquired user, you are in a healthy range. Use Adjust’s cohort reporting to compare CPI against downstream revenue and retention metrics to set a target that is specific to your app and audience.
How do I make sure my cost data in Adjust is accurate and complete?
The most reliable way to ensure accurate cost data is to activate native cost integrations for every ad network you run campaigns on, since these pull spend data directly via API and sync automatically. For networks without a direct integration, use Adjust’s CSV cost upload feature to manually import spend data on a regular schedule. Audit your Adjust cost data against each network’s dashboard at least weekly, and flag any campaigns where spend appears as zero or is missing entirely, as these gaps will silently distort your CPI figures.
Can I track CPI separately for different countries or platforms within the same campaign?
Yes, Adjust allows you to break down CPI at a granular level including by country, platform (iOS vs. Android), campaign, ad group, and even individual creative. You can apply these dimensions directly in the Adjust dashboard using filters and groupings, or export the data to a BI tool for custom reporting. This level of segmentation is especially useful for apps running global campaigns, where CPI can vary dramatically between markets and a blended average can mask both your best and worst-performing regions.
What should I do if my attributed install count in Adjust drops suddenly?
A sudden drop in attributed installs usually points to one of a few causes: a broken or expired tracking link, a change in your attribution window settings, a lapse in ad spend, or a platform-level issue such as an iOS ATT prompt affecting match rates. Start by checking your Adjust tracker links to confirm they are active and correctly configured, then verify that your cost integrations are still syncing. If the drop coincides with an iOS update or a change in network settings, review your probabilistic matching configuration, as device-level identifier availability directly affects Adjust’s ability to attribute installs on iOS.
Is CPI the right metric to optimize for, or should I be focusing on something else?
CPI is a useful efficiency metric for the top of your funnel, but optimizing for it in isolation can be counterproductive. A campaign that drives low-CPI installs from low-intent users will underperform a higher-CPI campaign that brings in users who convert, subscribe, or make purchases. The better approach is to use CPI as a guardrail while optimizing toward cost per paying user, cost per registration, or another in-app event that correlates with real revenue. Adjust’s event tracking lets you define these downstream events and tie them back to the same campaign and creative data, so you can move from CPI optimization toward true ROI optimization.
How does Apple's App Tracking Transparency (ATT) affect CPI calculations in Adjust?
ATT limits Adjust’s ability to use device-level identifiers (IDFA) for attribution on iOS when users decline tracking, which means a portion of installs will be attributed using probabilistic matching or modeled data rather than deterministic ID matching. This can reduce the total number of installs Adjust is able to attribute with high confidence, which may inflate your apparent CPI since some installs go unattributed and are excluded from the calculation. To mitigate this, ensure you have SKAdNetwork (SKAN) configured correctly in Adjust, as Apple’s privacy-preserving framework provides aggregated conversion data that can fill some of the attribution gap left by ATT opt-outs.
How often should I review CPI data in Adjust, and what does a good optimization workflow look like?
For active campaigns, reviewing CPI data in Adjust at least three times per week gives you enough frequency to catch underperformance before significant budget is wasted, while still allowing enough data to accumulate for statistically meaningful decisions. A practical workflow looks like this: check CPI by channel and campaign early in the week, identify outliers (both high and low), cross-reference with post-install metrics like day-7 retention or purchase rate, then make budget or creative adjustments mid-week. Reserve weekly or bi-weekly reviews for deeper analysis across geos, platforms, and creative sets to inform longer-term strategy.
Related Articles
- Best MMP pricing compared for apps: AppsFlyer vs Adjust
- 6 ways to find your app's biggest drop-off point
- 5 signs your app attribution data is wrong
- 6 ways to tell your app attribution setup is broken
- AppsFlyer vs Adjust: which is more accurate for app attribution?
- What are the legal requirements for app advertising in Europe?
- How do you track the ROI of app store optimization?
- How does A/B testing work in the app store?
- How do you choose the right app store category?
- How often should you update your app store keywords?
EN 
NL 