Deep-dive guide
Course Refund Tsunami Stress Test: What a Wave of Refunds Really Costs a Course Business
Executive industry analysis
A course launch is judged by its sales on day one. The dashboard shows a number of purchases, the bank balance rises and the launch feels like a success. The refund window then opens, and over the following weeks a share of that revenue goes back. Some buyers ask for their money, some dispute the charge with their bank, and the affiliates who brought them have already been paid. A month that looked like a record can end close to break-even. The stress test exists to show that effect before it arrives, not after.
The structure of the problem is a sale that is paid for up front and earned only after the window closes. Advertising is spent on every sale, refunded or not. Payment fees are usually kept by the processor even when you refund. Affiliate commissions are paid early and are only partly recovered. And the longer your guarantee, the more time there is for a buyer to change their mind, so the window itself raises the refund rate. These costs all happen on the same sales, so a rise in refunds compounds across several lines at once.
The domino model follows the cascade. The effective refund rate, which combines the base rate with the effect of the window, sets how much of each launch comes back. That produces the revenue leak, which adds chargebacks and fees kept by the processor to the refunds themselves. What remains per sale after ads, affiliates and delivery is the profit per sale. That feeds the margin on kept revenue, and the last box shows monthly profit. A small rise in refunds can move several boxes from green to red together because they share the same sales.
The iceberg separates the visible costs from those that arrive later. Above the waterline sit advertising, the affiliate commissions on sales that stayed, and the cost of delivering the course. Below it sit commissions that cannot be recovered on refunded sales, payment fees lost on refunds, chargeback losses with their dispute fees and the time spent handling refund requests. Together these hidden items can equal a large share of the visible costs, and they grow faster than sales when refunds rise.
Formula and methodology blueprint
The model works on one month. The effective refund rate is the entered rate scaled by how far the refund window is from thirty days. Refunded sales return the price, and a small share of sales also end in chargebacks that lose the price and a dispute fee. Processor fees are kept on all sales. Advertising is paid on every sale. Affiliate commission applies to a fixed share of sales, and only half of it is recovered on refunded affiliate sales. Delivery cost applies to customers who stay, and there is a fixed platform cost. The result is monthly profit.
- Effective refund rate
- r = refund rate × (1 + (window − 30) ÷ 150)
- A longer window gives buyers more time to ask, so the rate rises with it. It is capped at ninety percent.
- Refunded value
- R = sales × r × price
- Money handed back to buyers in the month.
- Chargeback loss
- C = sales × 0.6% × (price + dispute fee)
- Disputed sales lose the price and add a fee per dispute.
- Affiliate cost
- A = affiliate sales × price × commission × (1 − r × 50%)
- Commission on kept sales plus the part not recovered on refunded ones.
- Net profit
- Net = gross − R − C − fees − ads − A − delivery − handling − fixed
- The monthly result. The breaking refund rate is where this reaches zero.
- Break-even refund rate
- r* = where Net = 0
- The refund rate at which the month pays nothing, found by the breaking-point search.
Cascading elasticity
Refund rate is the most elastic input, because each point removes the whole price while ads, fees and part of the affiliate commission remain. Ad cost per sale and price come next, since they set the gap between revenue and cost. The refund window is a multiplier on the rate, so it matters most when the rate is already high. Sales volume is linear, which means growth helps only when each sale is profitable after refunds.
Three real-world case studies
Bootstrapped starter
A small launch with a short guarantee
A creator sells a few dozen copies of a modest course, with a two-week guarantee and cheap warm traffic. Refunds are low and there is little affiliate activity, so the month is profitable. Load it to see how a short guarantee protects a small launch.
Scaling mid-market
A large launch with affiliates
A course business sells fifteen hundred copies a month at a higher price, with an active affiliate programme and solid onboarding. Volume makes every refund point worth thousands of dollars. Load it to see how one point of refunds moves five figures.
Black-swan crisis
A viral launch and a refund tsunami
A promotion brings a wave of buyers who expected something different. Refunds reach four in ten, the guarantee runs sixty days, affiliates are already paid and disputes pile up. The launch loses money. Load it to see which lever, usually the refund rate or the window, restores the month.
Industry benchmark matrix
| Metric | Danger zone | Industry average | Top 10% elite |
|---|---|---|---|
| Refund rate (online courses) | Above 20% | 5–12% | Below 4% |
| Chargeback rate | Above 1% | 0.2–0.6% | Below 0.2% |
| Ad cost per sale as share of price | Above 60% | 20–40% | Below 15% |
| Affiliate commission | Above 50% | 20–35% | Under 20% |
| Margin on kept revenue | Below 8% | 20–35% | Above 45% |
Variables and levers
Course sales per month
Paid purchases in the month before any refunds or chargebacks.
Lever: Do not chase volume while refunds are high; each extra sale from the wrong audience adds a refund and a cost.
Course price
What a buyer pays for the course, as one payment or the programme total.
Lever: Raise the price only with a clearer promise and proof, since a higher ticket makes each refund larger.
Refund rate
Share of sales refunded at a 30-day window. The model raises it for longer windows and lowers it for shorter ones.
Lever: Fix the reasons for regret first: a clearer sales page, a fast-start module and a human check-in in the first week.
Refund window
Number of days a buyer has to ask for their money back under your guarantee.
Lever: Shorten the guarantee to what is fair, or pair it with simple conditions such as completing the first lessons.
Ad cost per sale
Average ad and promotion spend needed to produce one sale, which is paid whether or not the sale is later refunded.
Lever: Cut the campaigns that bring refunders, and judge ads by kept revenue after the window, not by sales.
Affiliate commission
Commission you pay affiliates on their sales. About three in ten sales are assumed to come through affiliates.
Lever: Hold affiliate payouts until the window has closed so refunded sales never cost you a commission.
How to use this simulator
- 1Start from the realistic preset or enter last launch's sales, price and refund rate.
- 2Read the five domino boxes to see which stage turns amber or red first.
- 3Press Find my breaking point to learn which refund rate or cost pushes profit to zero.
- 4Compare the current approach with the tighter guarantee plan over 24 months in the Parallel universe tab.
- 5Copy the share link so a launch partner can open exactly the same scenario.
Mistakes to avoid
- Judging a launch by day-one sales. Revenue is not safe until the refund window closes and disputes have settled.
- Paying affiliates before the window ends. Commission on a refunded sale is only partly recovered, so you fund the refund.
- Counting only the refunded price. Ads, processor fees, delivery and support time were all spent on a sale that no longer exists.
- Extending the guarantee to win a launch without modelling it. A longer window raises the refund rate and delays when cash is truly yours.
Frequently asked questions
What is a normal refund rate for an online course?
It varies with the niche, price and promise. Many courses see a mid single digit to low double digit refund rate. Compare yourself with your own history and with similar offers.
Does a longer guarantee increase sales enough to justify it?
Sometimes. A longer guarantee can lift conversion, but it also raises refunds and delays when revenue is safe. Test it, and compare profit after the window rather than sales on launch day.
How do I find my breaking refund rate?
Press Find my breaking point. The tool pushes each input towards its worse end in a thousand steps and reports the value at which monthly profit reaches zero, ranked by how close each one is.
Why hold affiliate payouts?
Because commission paid before the window closes is only partly recovered when a sale is refunded. Paying after the window protects you from funding refunds. Check your affiliate agreement and the rules where you sell.
Are chargebacks the same as refunds?
No. A chargeback is a dispute through the buyer's bank. You lose the sale and pay a fee, and too many can put your payment account at risk. A clear refund process helps prevent them.
How accurate is this?
It is an educational planning model with the assumptions stated above. It is not business, legal or financial advice. Replace the inputs with your own figures and check refund rules for your region.
This is an educational planning model with stated assumptions. It is not financial, tax or legal advice. Replace the inputs with your own figures and verify important decisions with a qualified professional.