)
Demystifying Cohort KPIs, Part 3: Tracking event conversions and funnels
){kind=small}
Cohort analysis offers a clear view of how user behavior develops from the first app open through key milestones in the lifecycle. In Part 1 of this series, we explored retention and session KPIs to identify who stays and how often they engage. In Part 2, we examined revenue KPIs to determine which users generate value, when they do so, and how much they contribute.
In this final installment, we focus on the actions that lead to those outcomes. Here, we’ll break down the core event conversion metrics, show how they fit into funnel analysis, and share best practices for using them alongside retention and revenue KPIs to improve user experience, optimize campaigns, and drive long-term growth.
Why event conversion KPIs matter for app growth
Event conversion KPIs measure the number of users who complete a specific in-app action within a defined time frame. Examples include completing onboarding, starting a free trial, leveling up in a game, or making a purchase. These actions indicate how users progress through the app, highlight where they encounter friction, and reveal which steps are most likely to lead to monetization.
Tracking these metrics helps teams map the user journey from install to high-value behaviors, identify drop-off points, such as during onboarding, gameplay, or checkout flows, and address issues that limit progression. They also provide a reliable way to measure the impact of product updates, user acquisition (UA) campaigns, or lifecycle marketing by comparing conversion performance before and after changes.
Because many of these actions are strong predictors of downstream KPIs like lifetime value (LTV), average revenue per user (ARPU), and paying user lifetime value (PU-LTV), they offer an early signal of revenue potential. This allows teams to make faster, data-driven decisions about scaling campaigns, re-engagement strategies, and product investments.
Core event conversion metrics
Event conversion metrics quantify how many users take a specific in-app action and how those actions progress over time. For consistency, each metric in this section follows the same rules: a user is counted only once per day for a given action, but may be counted on multiple different days if they perform the action again.
In the formulas below, n refers to a specific day after install (DAI) — for example, n = 0 represents the install day, and n = 7 represents day seven after install.
Converted users
Converted users measure the number of unique users in a cohort who complete a specific in-app action on a given DAI.
For example, if the tracked event is “level completed,” the table below shows when users in a cohort completed the event:
When aggregated by day after install, the converted users metric for the cohort looks like this:
In this example, user 2 converted on the install day (DAI 0) and again on DAI 2, while user 1 converted on DAI 2 and DAI 4. Tracking these patterns helps identify daily conversion flow, highlight spikes or drops, and measure the immediate impact of onboarding, feature launches, or campaigns.
Converted users size
Converted users size is the cumulative number of unique users in a cohort who have completed the event at least once by a given DAI. Once a user converts, they remain counted for all subsequent eligible days, whether or not they perform the action again. For complete cohorts, this value can only stay the same or grow, and for incomplete cohorts, it may decline on later days if some users haven’t yet reached those days.
Let’s take the same example as above that shows when each user in a cohort completed the event:
When aggregated into converted users size, the same cohort looks like this:
By DAI 4, only user 1 remains eligible, causing the size to decrease. This metric reveals how quickly the converter pool expands, when it stabilizes, and how the timing of conversions varies across cohorts.
Conversion distribution
While converted users size shows the total number of unique converters, conversion distribution reveals when those conversions occur and how frequently converted users repeat the action over time. Conversion distribution is calculated by dividing the converted users on a given day by the total converted users size, then multiplying by 100.
Formula:
Conversion distribution (DAI n) = (Converted users on DAI n ÷ converted users size) × 100
Because the same user can complete the event on multiple different days, the percentages across DAIs can exceed 100%. This metric highlights when engagement is strongest and helps pinpoint where users may encounter friction in the journey.
Conversion per user
Conversion per user is the share of the entire cohort who convert on a given day, calculated by dividing converted users on that day by the cohort size for that day.
Formula:
Conversion per user (DAI n) = Converted users on DAI n ÷ cohort size on DAI n
In this example, on DAI 0, one out of three users completed a level, resulting in a conversion rate of 0.33 (33%). By DAI 2, every user who had the app installed for at least two days had completed a level, raising the rate to 1 (100%). By normalizing conversion rates for cohort size, this metric allows fair comparisons across campaigns, channels, and time periods, regardless of user volume.
Conversion per active user
Conversion per active user measures the share of active users who convert on a given day, using retained users on that day as the denominator.
Formula:
Conversion per active user (DAI n) = Converted users on DAI n ÷ retained users on DAI n
Here, on the install day (DAI 0), one-third of active users converted, while on DAI 2 and DAI 4 every active user completed the action. By focusing only on active users, this metric isolates how effectively engagement translates into key actions, making it a strong indicator of engagement quality and the impact of targeted product updates or marketing efforts.
Event volume metrics
Event volume metrics measure how often a specific in-app action is triggered, focusing on frequency rather than unique conversions. These KPIs provide a deeper view of engagement by showing total activity, average actions per user type, and differences between active and converted users.
Events
Events capture the total number of times a specific in-app action is triggered on a given DAI. Unlike conversion metrics, which record whether a user performed the action at least once in a day, this metric counts every instance of the action. The same user can trigger multiple events within a single day or session, and each is counted.
Formula:
Events (DAI n) = total number of times the event was triggered on DAI n
For example, if the tracked event is “level completed,” the table below shows how many times users in a cohort completed levels across different DAIs:
This metric reveals activity intensity. Spikes can signal high engagement periods, such as post-feature release or during promotions, while drops may indicate friction points. Tracking events over time helps assess feature stickiness, compare activity between cohorts, and identify opportunities to encourage repeat engagement.
Events per active user
Events per active user shows the average number of times retained users trigger a specific event on a given DAI. It focuses only on active users, those who opened the app that day, making it a strong indicator of behavioral depth among engaged users.
Formula:
Events per active user (DAI n) = total events on DAI n ÷ retained users on DAI n
In this example, active users completed an average of 4.67 levels on DAI 0 and peaked at 7.5 on DAI 2. Days with no active users are marked “—” since the metric cannot be calculated. This KPI helps evaluate engagement quality, spot peaks or declines, and measure the effect of product changes, campaigns, or gameplay adjustments.
Events per converted user
Events per converted user measures the average number of times users who have converted at least once trigger a specific event on a given DAI. It isolates repeat behavior within the converter group, offering insight into post-conversion engagement.
Formula:
Events per converted user (DAI n) = total events on DAI n ÷ converted users on DAI n
Here, a single converted user on DAI 0 completed the event 14 times, while on DAI 2, two converted users averaged 7.5 completions each. This metric helps assess the depth of engagement within converters and whether features, rewards, or challenges encourage repeat interaction after the first conversion.
Events per user
Events per user calculates the average number of times the entire cohort triggers a specific event on a given DAI. It includes all users, regardless of whether they were active or had converted.
Formula:
Events per user (DAI n) = total events on DAI n ÷ cohort size on DAI n
In this example, on DAI 0, the cohort averaged 4.67 event completions per user, rising to 7.5 on DAI 2 when only two users remained. This macro-level view helps identify overall behavioral trends, measure the effect of updates or promotions, and benchmark engagement across cohorts.
Using event KPIs in funnels
Event KPIs become most actionable when viewed together as a funnel. A funnel maps sequential in-app actions, such as install → onboarding completion → first purchase, to show how users progress toward high-value outcomes. Measuring the percentage of users advancing from one stage to the next pinpoints where drop-offs occur and how quickly users progress.
Time-based windows (e.g., D1–D7 or D0–D30) reveal not only whether users convert, but also how long it takes them. Faster progression can indicate smoother onboarding, a more intuitive experience, or a more compelling product offering.
Funnels also surface monetization drivers early in the journey. Mid-stage actions, like completing a level or initiating checkout, often predict revenue KPIs such as LTV, ARPU, and PU-LTV. Correlating funnel completion rates with these revenue metrics highlights which stages have the greatest influence on long-term value and where optimizations will deliver the highest return on investment (ROI).
Common pitfalls and best practices in event conversion analysis
Event conversion and funnel KPIs can yield strong insights, but without the right context they risk leading to inaccurate conclusions. Below are the most common mistakes, and how to avoid them:
Misreading incomplete cohorts
Some events, such as high-level completions, second purchases, or late-stage onboarding steps, take time to occur. Including users who haven’t yet had the opportunity to trigger the event will understate conversion rates. To avoid skewed comparisons, filter results for users eligible to complete the event or use maturity-aware metrics that account for how far along each user is in the journey.
Using totals without normalization
Raw event counts don’t reflect engagement efficiency. Larger cohorts naturally produce more events, even if per-user activity is lower. Use per-user, per-active-user, or per-converted-user metrics to normalize results and make fair comparisons across campaigns, platforms, or time periods.
Ignoring multi-dimensional segmentation
Aggregated data can hide critical differences in user behavior. Segmenting funnels by acquisition source, platform, or geography can reveal which groups move through key stages faster, convert at higher-value points, or require tailored onboarding and retention strategies.
Over-focusing on a single event
Looking at one event in isolation can obscure issues elsewhere in the funnel. For example, a strong checkout completion rate is less valuable if most users abandon their cart before checkout begins. Funnel views link related events, expose bottlenecks, and connect early actions to downstream KPIs.
Lacking business-goal alignment
Measuring events that don’t map to monetization, retention, or strategic adoption creates noise. For example, monitoring a cosmetic action like changing an app theme may not provide actionable insights unless it directly influences retention or monetization. Prioritize events tied to measurable business outcomes and regularly audit your event taxonomy to ensure continued relevance.
Closing the loop: Turning event data into measurable impact
With retention and session KPIs (Part 1), revenue metrics (Part 2), and now event conversion KPIs (Part 3), you have a complete framework for tracking how users progress from first interaction to long-term value. Event metrics close the loop by linking in-app behaviors to monetization outcomes, revealing exactly which moments have the greatest influence on the user journey.
Applied consistently, these insights help prioritize resources, refine experiences that accelerate progression, and connect early actions to downstream KPIs. The result is a direct, measurable link between product decisions, marketing strategies, and business impact.
To act on these insights, teams need tools that reveal patterns quickly, compare performance across cohorts, and drill into the journeys that drive value. Adjust Datascape, part of our Analyze pillar, consolidates cohort, funnel, and event data in one customizable view. With flexible filters, time-based comparisons, and links to essential KPIs, Datascape enables teams to move from insight to optimization—fast.
Ready to take your mobile app performance to the next level? Request a demo to see first-hand how Adjust can enhance your cohort analysis and accelerate your app growth.
Be the first to know. Subscribe for monthly app insights.
Keep reading
