Blog · Board and management reporting
App marketing KPIs connect channel spending to activation, first payments, and retained payers. Work through a channel cohort with explicit denominators, reconciled counts, and costs per outcome.
App marketing KPIs connect acquisition spending to activation, first payments, and paid retention within the same user cohort. Use a short chain of measures with explicit denominators, then compare channels on the cost of the outcome your company needs. A low installation cost alone does not show whether acquired users become durable paying customers.
Start with new users acquired during a fixed period. Give each user one acquisition channel and preserve that assignment as the user returns. Keep a separate unknown channel rather than spreading missing assignments across the channels you can identify.
The distinction matters when choosing an export. Google Analytics distinguishes traffic acquisition, which includes new and returning users, from user acquisition, which concerns new users. A returning session's source should not silently replace the acquisition channel in a cohort comparison.
Choose the observation windows before looking at the outcome. In the example below, activation means completing the first core task within seven days of the first installation. A payer makes a first successful payment within 30 days, and a retained payer has an active paid entitlement at day 60 after installation. All cohorts have reached that age before the report closes.
These are this company's definitions, not universal benchmarks. Keep the definitions with the workbook so a change in the activation event or billing rule does not look like a marketing improvement. The broader KPI examples page covers measures outside this app funnel.
| KPI | Formula | What it helps decide |
|---|---|---|
| Cost per install | Channel acquisition spend / New installations | Whether acquisition volume is becoming more expensive |
| Activation rate | Activated new users / New installations | Whether acquired users reach the initial value event |
| Activated-to-paid conversion | First-time payers / Activated users | Whether activated users become customers |
| Paid retention | Retained original payers / Original first-time payers | Whether the acquired paying cohort persists |
| Cost per retained payer | Channel acquisition spend / Retained original payers | Which channel produces the chosen durable outcome efficiently |
The conversion formula assumes payers are a subset of activated users. If your app allows payment before activation, report install-to-paid conversion separately instead of forcing those users into a sequence they did not follow.
Keep activity retention separate from payment retention. Google Analytics' retention documentation describes returning users grouped by acquisition cohort. That is useful activity evidence; the paid-retention measure here requires billing status as well. Covirage is not affiliated with Google.
Use a user export containing user ID, first-install date, fixed acquisition channel, and activation date. Match it to payment and entitlement exports containing the same user ID, first successful payment date, refund status, and entitlement status at the observation date. Keep failed payments separate from successful payments.
Bring a spending export with campaign, channel, acquisition period, currency, and included cost categories. Reconcile channel spending to the approved marketing expense report. If channel spend excludes salaries and shared costs, call the resulting measure campaign cost per payer; do not label it fully loaded customer acquisition cost.
The example is a fictional US subscription app's January 2026 acquisition cohort. The spending scope is attributable campaign expense in USD. An installation is counted once per new user ID, and all first-time payers in this example also activated.
| Acquisition channel | Spend (USD) | New installations | Activated users | First-time payers | Retained payers at day 60 |
|---|---|---|---|---|---|
| Paid search | 18,000 | 3,000 | 1,500 | 300 | 240 |
| Paid social | 12,000 | 4,000 | 1,200 | 240 | 144 |
| Total | 30,000 | 7,000 | 2,700 | 540 | 384 |
The following figures divide the counts above. Money is rounded to the nearest cent and percentages to one decimal place; calculations use unrounded values.
| KPI | Paid search | Paid social | Combined cohort |
|---|---|---|---|
| Activation rate | 50.0% | 30.0% | 38.6% |
| Activated-to-paid conversion | 20.0% | 20.0% | 20.0% |
| Paid retention at day 60 | 80.0% | 60.0% | 71.1% |
| Cost measure | Paid search (USD) | Paid social (USD) | Combined cohort (USD) |
|---|---|---|---|
| Cost per install | 6.00 | 3.00 | 4.29 |
| Campaign cost per first-time payer | 60.00 | 50.00 | 55.56 |
| Cost per retained payer | 75.00 | 83.33 | 78.13 |
For search, paid retention is 240 / 300 = 80.0%, and cost per retained payer is 18,000 / 240 = $75.00. For social, the corresponding cost is 12,000 / 144 = $83.33. Social buys cheaper installations and first payments, while search produces the lower cost per retained payer at the chosen age.
Compute the combined figures from combined counts. Total paid retention is 384 / 540 = 71.1%; it is not a simple average of channel percentages. The spending and retained-payer counts also prove the combined acquisition outcome: 30,000 / 384 = $78.13 per retained payer.
Each user must appear once in the acquisition table. For this nested funnel, the checks are:
| Reconciliation | Calculation | Result |
|---|---|---|
| Installations | Activated + Not activated | 2,700 + 4,300 = 7,000 |
| Activated users | First-time payers + Activated nonpayers | 540 + 2,160 = 2,700 |
| Original payers | Retained + No longer paid | 384 + 156 = 540 |
Match each count to distinct user IDs, not event rows. Repeated installations, payment attempts, and multiple entitlement records can otherwise multiply one customer. Investigate missing identity matches as a separate exception list; marketing operations KPIs cover that process health.
The acquisition lead owns spending and channel assignment. Product owns the activation definition, billing owns paid status, and finance approves the cost scope. Together they can decide whether the channel result justifies a small budget test, whether onboarding needs work, or whether retention needs investigation. This table alone does not establish lifetime profitability or prove that a channel caused the observed difference. For message-driven nurturing, use the separate marketing automation KPIs.
Bring your user, payment, entitlement, and channel-spending exports, with the cohort rules documented. Covirage maps the columns; its tools compute and check the funnel figures, and the external AI model explains the results with every figure traced to its rows. Create a free Covirage account and bring a sample of your own data; account creation takes about a minute.
See Covirage for sales insights for the supporting business use case.
Cost per install measures the expense of acquiring an installation, not the value of the customer who follows. Compare it with activation, payment, and retention measures from the same cohort. An inexpensive channel can cost more per retained payer if many users leave before renewing.
Choose a completed action that demonstrates the app's core value, such as completing a first business task. Record the event and the allowed time after acquisition. Opening the app can be useful activity evidence, but it should not automatically count as meaningful activation.
User retention asks whether acquired users return or remain active under a stated activity rule. Paid retention asks whether an original payer remains a paying customer at a stated age. A returning user can have canceled payment, so activity and billing exports answer different questions.
It depends on the measure. A campaign spending ratio can include only attributable media and campaign costs, provided it is labeled that way. A fully loaded customer acquisition cost needs the wider acquisition cost policy agreed with finance, including the relevant sales and marketing expenses.