The acquisition proposal says a cohort pays back in three months. The proposed media budget doubles next month. Before approving it, ask two separate questions: what exactly crosses the recovery threshold, and how much cash stays committed while several cohorts are still earning it back?
The CAC payback period for a mobile app is the time a defined acquisition cohort takes to generate enough contribution, before acquisition cost, to cover its agreed acquisition cost. Calculate it from a cumulative cohort curve when value changes with age. Then build a separate calendar cash schedule. A contribution crossing does not establish settled cash, company profitability, or that the next cohort will behave the same way.
This guide is for app founders, growth leads and finance partners reviewing active paid acquisition or an agency's expansion proposal. The decision is how long the business can fund unrecovered acquisition costs, with which evidence and under whose authority.
Define the recovery claim before calculating it
Stripe describes the basic calculation as acquisition cost per customer divided by monthly profit per customer, after costs of serving that customer (Stripe's CAC payback explanation). That shortcut requires a meaningful monthly contribution estimate. It can mislead when trial conversion, renewal timing, churn or purchase concentration makes contribution uneven.
For those apps, use the first observed age at which cumulative contribution reaches acquisition cost. Label the cost boundary: media-only recovery, paid-acquisition recovery including agreed fees and production, or a broader fully allocated calculation. These are different claims. A media-only crossing should not be presented as recovering every cost of growth.
Write a short metric contract containing the cohort entry event, acquisition dates, included territories and products, cost allocation, contribution definition, reporting currency, age convention and extraction timestamp. Add the owner who can approve restatements. The same workbook should reproduce the claim next week, with a visible explanation for any changed result.
Match the population on both sides
An acquired install, activated user and new paying subscriber are different denominators. Choose the population that matches the decision and carry its acquisition costs with it.
If all paid installs form the cohort, include the contribution of that population and the cost of acquiring all those installs. If the cohort begins at first payment, document how acquisition costs are assigned to it, including spend associated with people who never paid. Dividing total acquisition spend by payers can be valid under an explicit policy; pairing install CAC with payer-only value inflates apparent recovery.
The CPI versus CPA guide covers denominator selection in more depth. Here, freeze the chosen population before looking for a crossing. Do not remove users who churned, refunded or generated no revenue simply because they make the payback curve less attractive.
Keep reactivation separate from new acquisition unless the contract deliberately combines them. Likewise, do not add platform-attributed populations together and assume they represent unique users. Privacy restrictions and incomplete identity can leave a material unmatched portion. Preserve that limitation instead of forcing a user-level join with unnecessary personal data.
Build a contribution curve, not a monthly average
For each cohort and fully observed age interval, calculate:
Cumulative contribution before acquisition = defined cohort value − agreed deductions − variable delivery costs
Then compare it with the same cohort's acquisition cost. When using per-user figures, divide both sides by the same original cohort count. The aggregate and per-user calculations should reach the same recovery conclusion.
Define what value already includes. A proceeds field may already deduct some charges; deducting them again understates contribution. A gross purchase field may require adjustments. Use the mobile app ROAS reconciliation guide to establish that value bridge, and the LTV calculation guide for later-value forecasts. The recovery calculation should consume those approved definitions rather than invent another revenue standard.
Keep each age interval's contribution as well as the cumulative total. That makes a late refund or serving-cost adjustment visible. A curve can cross the threshold and later fall below it; the first crossing is a historical observation, not an irreversible certificate of recovery.
A fictional cohort with a crossing and a reversal
Every number in this example is invented to demonstrate the calculation. These are not app benchmarks, fee rates, forecasts or Sharply Labs results.
Suppose an acquisition cohort contains 1,000 paid-acquired users. Media cost is $20,000; an explicitly allocated $4,000 of production and agency cost brings acquisition cost to $24,000. Its agreed contribution, after the relevant deductions and variable delivery costs but before acquisition, develops as follows:
Fully observed age Cumulative contribution Contribution less $24,000 acquisition cost Interpretation --- --- --- --- D30 $8,000 −$16,000 Cost not recovered D60 $17,000 −$7,000 Cost not recovered D90 $25,000 $1,000 Above the stated threshold D120 $23,000 −$1,000 Later net adjustment reverses the crossing D150 $27,000 $3,000 Above threshold again at this extraction
With only these snapshots, the first observed crossing is somewhere after D60 and by D90. Reporting “day 86” would invent precision. Daily records could narrow the interval; linear interpolation would be an estimate and should be labeled accordingly.
The D120 reversal also matters. State both the first observed crossing and the latest recovery status. If the team uses an additional stability condition before releasing capital, document it as a business policy with its owner. Do not imply that a chosen number of days makes refunds or later costs impossible.
RevenueCat's Cohort Explorer illustrates why source treatment matters: it applies refunds in the period where they occur and defines customer counts according to the selected cohort (RevenueCat Cohort Explorer documentation). That is provider-specific behavior to reconcile, not a guarantee that every export uses the same timing.
When a cohort has not paid back, report the observation limit
If the latest complete observation is D90 and contribution is below cost, the defensible result is “not recovered through D90.” It is not a measured six-month payback. The curve may cross later, never cross, or change after adjustments.
Show forecast recovery separately, with assumptions for retention, payer conversion, renewals, ad yield and delivery costs. Record which mature cohorts support those assumptions and whether the model has been tested against withheld later observations. A forecast range is useful only if its uncertainty is carried into the funding decision.
Compare cohorts at equal maturity. Google Analytics lets analysts choose acquisition, event or transaction inclusion rules and standard, rolling or cumulative calculations. Its cohort exploration uses device data rather than User-ID (Google Analytics cohort documentation). A similarly named report therefore may not match a subscription ledger's people or time boundaries.
The cohort analysis guide explains that comparison contract. For payback, retain a visible observation cutoff and a coverage note beside each crossing. Missing or immature cells must not silently become zero contribution.
Translate recovery into a calendar cash schedule
A cohort curve follows user age. A cash schedule follows the business calendar. Several cohorts can leave cash committed simultaneously, even when each eventually recovers its acquisition cost.
Build a separate row for each acquisition cohort and calendar period. Record acquisition cash outflows when payable, attributable receipts when expected or settled, and delivery cash outflows when due. Use finance-approved timing and mark estimates. Do not assume the transaction date, analytics report date and bank receipt date coincide.
For a defined acquisition portfolio, a useful planning measure is:
Cumulative funding gap = cumulative included cash outflows − cumulative included cash inflows
The largest positive gap over the modeled calendar is the peak funding exposure under those assumptions. It is not the company's required cash balance: payroll, tax, debt, fixed commitments, reserves and other business activities may sit outside this scoped schedule. Finance must integrate them into the company plan.
Do not simply multiply today's media spend by a reported payback period. That can miss staged receipts, growing acquisition budgets, fees paid at different times and the overlap between cohorts. A calendar schedule makes the assumption behind every amount inspectable.
A second fictional example: overlapping cohorts
This separate example is hypothetical and does not reuse the contribution curve above. Assume a company launches one cohort per month at $10,000 of immediately paid acquisition cost. Each cohort generates net cash receipts, after the included delivery outflows, of $2,000 in its launch month, $3,000 the next month and $5,000 the following month. Assume no further amounts, no delay within a month and no opening portfolio. These simplifying assumptions are deliberate, not operating expectations.
Calendar month Acquisition outflow Net receipts from all launched cohorts Month funding gap Cumulative funding gap --- --- --- --- --- Month 1 $10,000 $2,000 $8,000 $8,000 Month 2 $10,000 $5,000 $5,000 $13,000 Month 3 $10,000 $10,000 $0 $13,000 Month 4 $10,000 $10,000 $0 $13,000
Every cohort reaches its own cash recovery by the end of its third month. Yet the ongoing portfolio still ties up $13,000 at these month-end checkpoints. Equal monthly receipts and acquisition outflows stop the gap growing; they do not repay that initial funding while acquisition continues unchanged.
Actual intramonth timing could produce a higher peak. The example also excludes overhead and reserves, so $13,000 is not a recommended cash buffer. Extending the schedule after new acquisition stops would show how later receipts unwind the remaining gap under the example's assumptions.
Before proposing expansion, model the new cohort sizes on the actual calendar. Compare an approved base case with clearly labeled slower receipts, weaker contribution and higher acquisition-cost scenarios. Their differences express funding sensitivity, not probabilities unless a defensible model supplies those probabilities.
Annual billing and mixed monetization need explicit timing
An annual payment can bring receipts forward while the obligation to serve the customer continues. Keep expected delivery obligations and refund exposure in the schedule rather than treating all upfront receipts as freely reusable acquisition capital. The accounting recognition and reserve treatment require the business's finance owner.
For ad-supported apps, separate estimated monetization from settled receipts and relevant adjustments. For mixed models, check that a total-value field is not being added to its own subscription or advertising components. Revenue concentration also deserves a sensitivity case: a cohort dependent on a small number of large purchases can have a fragile recovery forecast even when the average looks strong.
This guide does not prescribe store commissions, payout dates, accounting treatment or a legal reserve. Use the app's applicable contracts, ledgers and approved policy. A reporting tool's value estimate is evidence for a model, not proof of an available bank balance.
What is a good CAC payback period for a mobile app?
A suitable period fits the app's observed contribution pattern, uncertainty, funding capacity and approved risk limits. A universal month count cannot establish that fit across subscription, purchase and ad-supported models.
Ask whether the business can fund the peak exposure if the slower scenario occurs, whether recovery is observed or forecast, and how much future value depends on an untested retention or monetization assumption. The approved limit should also specify what happens when evidence deteriorates. A maximum age with no owner or decision rule is only a label.
LTV:CAC and payback answer different questions. A forecast can suggest attractive lifetime value while leaving capital unavailable for a long time. ROAS measures return at a chosen horizon, but its numerator may differ from contribution or cash. Keep these measures together without treating one as a substitute for the others.
The evidence to request from an agency
Before a payback claim supports a budget increase, request a compact, reproducible packet:
Definition sheet: population, cost boundary, value deductions, age convention and extraction date.
Observed curve: period and cumulative contribution, immature cells, unmatched amounts and later revisions.
Forecast bridge: assumptions and ranges clearly separated from observed values, with backtest limitations.
Calendar schedule: payment and receipt timing, overlapping cohorts, peak exposure and excluded company commitments.
Decision record: approved funding limit, bounded next step, review trigger and accountable owner.
Keep source-native campaign reports beside this business view. Google Ads documents conversion lag for eligible App campaigns and explains that delayed conversions can affect reported CPA and ROAS (Google conversion lag documentation). A campaign report's incomplete value window and a business cash schedule are different timing problems; one does not correct the other automatically.
The packet should explain what cannot be reconciled. It should not require a new measurement vendor merely to make a presentation cleaner. Nor does a descriptive payback curve prove incremental acquisition value. That causal question needs an appropriate experiment or other defensible identification strategy.
Make the next funding decision reviewable
Use the evidence to choose a bounded action. Repair definitions when the populations or costs do not match. Hold an expansion decision when a material value horizon remains unobserved. Test within an approved exposure limit when the uncertainty is measurable and the business can bear it. Escalate cash commitments to the finance owner when the schedule exceeds the authorized envelope.
The mobile app marketing budget guide places that decision within a wider acquisition plan. Do not automatically cut every long-payback cohort or scale every short one. The decision also depends on product quality, acquisition concentration, capacity and what the next cohort is expected to teach.
For established app teams reviewing an agency's recovery claim or funding proposal, Sharply Labs' mobile app growth practice and paid acquisition services are the relevant starting points.
Request a scoped payback evidence conversation with a redacted cohort curve, acquisition-cost definition and current receipt schedule. The initial discussion examines the recovery claim, observation limits and funding assumptions, and produces a prioritized decision agenda with evidence gaps and ownership questions. Deeper reconciliation, account access and test design are scoped separately.
The conversation does not promise a shorter payback, profitable acquisition, available funding or improved campaign performance. The practical aim is a funding decision whose definitions, uncertainty and authority can be checked.