Guide
How to Model Returns in Ecommerce
Returns should be treated as expected variable cost, not as an afterthought.
What it means
Returns modeling converts a noisy operational reality into a usable expected cost per order. The simplest approach is to multiply return rate by average return cost.
Why it matters
Categories with elevated return behavior can look profitable before returns and weak after returns. Ignoring that difference leads to bad discounting and overstated CAC tolerance.
Simple formula
Expected return cost per order = return rate x cost per returned order.
The cost per returned order can include return shipping, processing labor, repackaging losses, and write-offs.
Common mistakes
- Using refund rate without operational return cost.
- Applying one blended rate across categories with different behavior.
- Assuming returned inventory is always fully recoverable.
- Treating returns as fixed overhead instead of order-linked cost.
Practical example
An 8% return rate with a $10 cost per return creates an expected cost of $0.80 per order. That looks small until it is stacked on top of fees, fulfillment, and promo pressure across thousands of orders.
Related tool
Use the contribution margin calculator to test how returns change the economics of a single order.