Your dashboard books a sale the day it clears, but net revenue after refunds is the only figure your ad budget should trust
A sale is counted the instant it clears. The refund lands days later, the chargeback months later, and by then you have already spent against the money. Here is how refunds and chargebacks inflate your revenue and LTV, and why net revenue after refunds is the number to run the business on.

Key takeaways
- Net revenue after refunds is your booked revenue minus every refund and chargeback that later reverses. It is always smaller than the number your dashboard shows on sale day, because the reversals arrive after the sale is already counted.
- Typical mobile refund rates run 2 to 5 percent of paid transactions, reaching about 4.9 to 5.1 percent for education apps and 5.8 percent behind a hard paywall, per Adapty's analysis of 1.9 billion dollars across more than 11,000 apps.
- Failing to subtract a 3 to 5 percent refund rate overstates the return on your acquisition campaigns, so you bid to buy users against revenue that does not exist.
- A refund reverses your net, not the gross. Apple and Google both hand their commission back on a refund, so you return roughly what you were paid, not the full price the customer saw.
- A chargeback is the expensive reversal. Since August 3, 2026, Google Play bills the developer the purchase price less Play's service fee plus a bank chargeback fee, which payment-industry sources put around 15 to 25 dollars per dispute.
- The reversals land on different clocks. A refund can arrive days after the sale, and a card chargeback commonly has 120 days and up to 540 in certain cases, so a cohort you closed can keep shrinking for a year.
- The delivery cost never reverses with the sale. The compute, API calls, storage, and payouts you spent serving that customer are gone whether the money comes back or not.
Booked revenue and kept revenue are two different numbers. The gap between them is every refund and chargeback that reverses after the sale already looked final, and it is wider than most dashboards admit.
Your dashboard books a sale the moment it clears. The refund can arrive days later, the chargeback can arrive months later, and by then you have already reported the revenue, counted it toward a cohort, and in most cases spent against it to buy the next user. Net revenue after refunds is the figure that survives all of that, and it is the only one your acquisition budget should trust.
Here is where the gap comes from, what it does to the numbers you run the business on, and what it costs in real money once you stop counting the sales you had to give back.
What net revenue after refunds actually means
Net revenue after refunds is a plain idea that dashboards make easy to forget. It is the money you booked minus the money you had to give back, after the store settles its cut. A sale is recorded the instant it clears. A refund or a chargeback is recorded whenever it happens, which is later, sometimes much later. The two events land in different reporting periods, so the sale looks like clean revenue right up until the reversal quietly appears in a future statement.
A sale is counted once, and reversed on its own schedule
On Google Play a buyer can self-serve a refund for 48 hours after purchase, once per app. Apple decides refund requests case by case, usually within a day or two. A card chargeback runs on the bank's calendar, commonly 120 days from the transaction and up to 540 in certain fraud and future-delivery cases. That spread is the whole problem. The sale hits your dashboard now, and the reversal can hit any time from tomorrow to next year.
Gross bookings are not what you keep
Gross bookings are what customers paid. Net revenue after refunds is what stays in your account once the reversals clear and the store takes its cut. Both Apple and Google return their commission when a purchase is refunded, so a refund pulls back your net, not the gross. That is a relief on a plain refund and a trap in your metrics, because if your revenue number still carries the gross of a sale that was later refunded, you are reporting money that already left.
How refunds quietly inflate the numbers you run the business on
They overstate revenue and MRR
Every refund you have not subtracted is revenue you do not have, sitting inside a number you are treating as real. At a 3 to 5 percent refund rate, a few points of your reported revenue is money already promised back. On subscriptions it also props up MRR, because a subscriber who refunded and churned can still read as active revenue until the reversal is reconciled.
They inflate LTV, so you overpay for users
This is where the accounting mistake turns into a spending mistake. Lifetime value built on gross bookings counts refunded sales as lifetime value. Adapty's benchmarking is blunt about it. Failing to account for a 3 to 5 percent refund rate significantly overstates the return on acquisition campaigns. If your LTV runs a few points high, your allowable cost per install runs a few points high, and you bid up to buy users against revenue that reverses. You are paying real ad dollars to chase a number that was never yours to keep.
They corrupt attribution and cohort data
Refunds also break the link between a sale and the campaign that produced it. AppsFlyer puts unhandled refunds among the pitfalls that reduce in-app purchase data accuracy, and estimates refunds alone can cut accuracy by 5 to 10 percent. The fix is tedious. Match each refund back to the original receipt and subtract it from the campaign and cohort that got credit for the sale. Skip that, and your best-performing channel on paper may just be the one with the highest refund rate.
| Reversal | Who decides | When it can land | What you give back |
|---|---|---|---|
| Google Play self-service refund | Within 48 hours of purchase | Your net; Google returns its service fee | |
| Apple refund request | Apple | Case by case, usually a day or two | Your net; Apple reverses its commission |
| Card chargeback, Google Play | The customer's bank | Commonly up to 120 days, 540 in some cases | Purchase price less service fee, plus a 15 to 25 dollar bank fee, since August 3, 2026 |
| Card chargeback, App Store | The customer's bank | Commonly up to 120 days, 540 in some cases | Apple resolves it as merchant of record; no bank fee passed to you |
What the gap costs you in money
Precision in the metrics matters because the gap is not academic. It is cash.
A refund costs you your net plus the service you already spent
On a plain refund the money side is bounded. Apple and Google reverse their commission, so on a refunded 100 dollar sale you give back roughly the 70 dollars you were paid, not 100. What does not come back is the cost of serving that customer. The compute you ran, the API calls you paid for, the storage you held, and any payout already booked are spent. A refund reverses the payment, not the work.
A chargeback costs more, and on Google Play it now costs you directly
A chargeback is the expensive reversal, and the bill moved. For orders placed on or after August 3, 2026, Google Play charges the developer the purchase price less Play's service fee, plus the bank's chargeback fee. Payment-industry sources put that bank fee around 15 to 25 dollars per dispute. On a 19.99 dollar sale, a lost chargeback reverses the roughly 17 dollars you netted and adds 15 to 25 dollars on top, so a single dispute can cost more than double the sale. On the App Store, Apple is the merchant of record and resolves card disputes itself, so it does not pass you the bank's fee.
The bill arrives after you have already spent the revenue
The timing is what makes it sting. You booked the sale, counted it in a cohort, and spent against it to buy the next user. The refund lands days later and the chargeback can land months later, on Google Play commonly up to 120 days and sometimes far longer. By the time the reversal hits, the revenue is already deployed. You are not trimming a forecast, you are covering a shortfall in cash you already committed.

How to run on net revenue after refunds
Subtract reversals at the cohort level, not just the top line
A company-wide refund percentage is not enough, because refunds are not spread evenly. Education and hard-paywall apps refund far more than travel or shopping. Subtract refunds and chargebacks inside each cohort and each campaign, so the channel you scale is the one with the best net, not the best gross.
Hold the sale open until the windows close
Do not treat a sale as final revenue the day it clears. Keep it provisional through the refund window and, for money you plan to spend against, through the chargeback window. A 48-hour Play refund and a 120-day card dispute are two different risk horizons on the same sale.
Answer the two windows where your evidence counts
You cannot stop most reversals, but two of them ask for your side, and answering protects real revenue. Apple sends your server a CONSUMPTION_REQUEST and gives you 12 hours to respond through Send Consumption Information V2. Google Play opens a 24-hour chargeback review through orders.reviewrefund. Everything else, the 48-hour Play refund, support refunds, voided purchases, is decided without you and can only be recorded. Miss the two windows that do take input, and you lose revenue you could have kept.
Reconcile late, because the number keeps moving
The revenue for a given month is not final at month end. A payout that looked clean can be reduced by a chargeback weeks or months later. Reconcile refunds and disputes back to the original orders on a rolling basis, so your net revenue after refunds settles to the truth instead of drifting.
This is the number RefundHalt is built to keep honest. Refunds and chargebacks are matched back to the sale and the cohort they reverse, the 24-hour Google review and the 12-hour Apple consumption request surface with their clocks so you can answer the ones that take input, and a dispute that arrives months later lands against the right sale instead of a mystery deduction in a later payout.
The short version
Your dashboard shows a sale the day it clears. Net revenue after refunds shows what you actually kept after every reversal, and it is always the smaller number. Refunds run 2 to 5 percent of paid transactions and more in some categories, they reverse your net rather than the gross, and a chargeback on Google Play now costs you the purchase price less the service fee plus a 15 to 25 dollar bank fee on top. All of it arrives after you booked the revenue and often after you spent it. Build your LTV and your ad budget on net, answer the two windows that take your evidence, and reconcile late, because a sale is not yours to keep until the customer can no longer take it back.
Frequently asked questions
- What is net revenue after refunds?
- Net revenue after refunds is your booked revenue minus every refund and chargeback that later reverses, after the store's commission is settled. A sale is counted the day it clears, but refunds and chargebacks arrive later, so the net is always smaller than the gross your dashboard shows on sale day. It is the number to use for LTV and acquisition budgets, because it is the money you actually keep.
- How much do refunds reduce app revenue?
- Typical mobile refund rates run about 2 to 5 percent of paid transactions, and higher in some categories, around 4.9 to 5.1 percent for education apps and 5.8 percent behind a hard paywall, according to Adapty's analysis of 1.9 billion dollars across more than 11,000 apps. That is the share of booked revenue you should expect to give back, before chargebacks.
- Do I get the store commission back when I refund a customer?
- Yes. Both Apple and Google return their commission when a purchase is refunded, so you hand back roughly what you were paid, not the full price the customer saw. Apple's agreement reserves the right to keep its commission, but in normal practice it reverses it. The cost you cannot recover is the service you already delivered, not the platform fee.
- Why do refunds inflate LTV and ROAS?
- Because a sale is counted immediately and the refund is subtracted later, if at all. Lifetime value built on gross bookings treats refunded sales as value you kept, which overstates the return on acquisition. Failing to account for a 3 to 5 percent refund rate is enough to make unprofitable campaigns look profitable, so you overbid for users against revenue that reverses.
- How is a chargeback different from a refund in cost?
- A refund reverses your net and the store returns its commission. A chargeback is decided by the customer's bank, is final, and on Google Play since August 3, 2026 it bills the developer the purchase price less Play's service fee plus a bank chargeback fee, commonly 15 to 25 dollars. On the App Store, Apple handles card disputes as merchant of record and does not pass you the bank fee.
- When is a sale safe to count as kept revenue?
- Not until its reversal windows close. Google Play's self-service refund runs 48 hours, Apple decides requests case by case, and a card chargeback commonly has 120 days and up to 540 in certain cases. For revenue you plan to spend against, treat the sale as provisional through the chargeback window, because that is how long the customer's bank can still pull it back.
Sources and further reading
- Adapty: Refund rate metrics and benchmarking (typical mobile refund rate 2 to 5 percent, category rates including education and hard paywall, based on 1.9 billion dollars across 11,000+ apps)
- AppsFlyer: 7 pitfalls to watch out for when measuring your in-app purchases (unhandled refunds can reduce IAP data accuracy by 5 to 10 percent and misattribute revenue)
- RevenueCat: Does Apple keep its commission after you refund a purchase (Apple reverses its commission on a refund in normal practice, so the developer returns the net)
- Google Play Console Help: Updates to chargeback cost responsibility (for orders on or after August 3, 2026 the developer bears the purchase price less Play's service fee plus the bank's chargeback fee)
- Chargeflow: Visa chargeback dispute rules, fees and time limits (acquirers typically charge 15 to 25 dollars per chargeback; cardholders generally have 120 days, up to 540 for certain fraud cases)
- Google Play Help: Apps, games, and in-app purchases refund policies (48-hour self-service window, once per app, contact the developer afterward)
- Apple Support: Request a refund for apps or content that you bought from Apple (customers file at reportaproblem.apple.com and Apple decides case by case)
- Apple Developer: Send Consumption Information (App Store Server API, the 12-hour consumption request that feeds Apple's refund decision)
RefundHalt
The refund autopilot for the App Store and Google Play
Keep reading
Months after a sale, a chargeback can still claw the money back, and the chargeback time limit is longer than you think
A store refund window is short. A chargeback window is not. Under Visa and Mastercard rules a customer commonly has 120 days to dispute a charge, and some cases reach 540, long after any store refund window has closed. Here is the real chargeback time limit and what a late dispute costs you.
An annual subscription refund can claw a whole year's payment back at once, and each store handles it differently
A monthly plan gives back one month. An annual subscription refund is calculated on the last charge, so on a yearly plan that is the entire year. Here is how Google Play and the App Store each decide it, what you can control, and what a full year reversed really costs.