Syntha Lab RU

Returns: what they cost and how they hit margin

Calculates the loss from returns: reverse logistics, resale at a discount, write-off, lost profit. Shows the maximum return rate at which profit is zero.

Platforms and marketplaces For whom: Apparel brands and sellers online and on marketplaces, finance and operations directors.

Calculate on your data →

At a glance

What question does it answer?
What returns cost us and at what return rate a category becomes loss-making.
What data is needed?
Sales for the period, price, cost, variable costs per unit.
How to read the result?
Compare the break-even return rate with the actual one. Close to the limit means the category is on the edge of loss: the result turns amber when the return rate exceeds 70% of the break-even rate (a rule-of-thumb). Look for reserves in the share resold at full price and the reasons for returns: size, photos, description, quality. Example on the default values: 10,000 orders, returns 22% (2,200 units): losses of RUB 6.36m from a profit of RUB 15m before returns leave RUB 8.64m. The return rate at which profit is zero is 51.9%.

Why and how to reason

A return costs more than it seems. First, you lose the profit you expected. Second, you pay for return logistics and handling. Third, variable costs already spent on the order (packaging, delivery) do not come back. Fourth, part of what returned cannot be sold at full price: it is sold at a discount or written off.

So loss per return = unit profit + return costs + losses on resale. Profit after returns = sold x [(1 - return rate) x unit profit - rate x return costs].

The break-even return rate: the rate at which profit falls to zero = unit profit / (unit profit + return costs).

Example. Unit profit 3,500 - 1,400 - 600 = RUB 1,500. Returns 22% of 10,000 = 2,200 units. Per return 250 + 600 + 542 (discount and write-off) = RUB 1,392 plus the lost 1,500. Losses RUB 6.36 m, profit from 15 to RUB 8.64 m. The limit is 51.9%.

When to use it

It comes in handy for categories with a high return rate: footwear, dresses, online sales. It helps decide whether to limit discounts and size charts, or change the return policy. Assumption: the return rate is constant and part of what returned is resold. If goods spoil, increase the write-off share. It does not reflect the effect of returns on customer relations and rating. Take the return rate from actual season data.

How to read the result

Compare profit before and after returns: in the example returns take 42% of profit (6.36 of 15 m). The break-even rate of 52% shows headroom but is no reason to relax: already 22% eats almost half. Look at the scenarios of 5, 10, 20, 30 and 40%. Decision: work on the causes of returns (size, description, photos), negotiate paid returns or tighten conditions, raise the share resold at full price. Find the causes of returns by category: size, quality, mismatch with photos. Each cause is treated differently. Measure by how many points returns must fall to restore the target profit.

Common mistakes

Counting only return delivery; assuming returned goods always resell at full price; not including handling and packaging; mixing returns with cancellations before dispatch. Also: taking the return rate from a short period. Also: ignoring warranty returns, which cost more than ordinary ones.

What data is needed?

  • Sales for the period, price, cost, variable costs per unit.
  • Return rate, reverse logistics and handling per return.
  • What part of returns is resold at full price, the resale discount and write-offs.

Inputs

FieldUnitValue in the exampleMeaning
Units sold in the periodunits10,000How many units were sold in the period; the return rate is taken of them.
Average selling price₽3,500Average selling price per unit excl. VAT after discounts.
Unit cost₽1,400Unit cost excl. VAT.
Other variable costs per unit₽600Fees, logistics, packing, advertising: everything that grows with a sale and is lost on a return.
Return rate% of sold22The share of sold units that came back; take it from your own statistics by category and channel. For footwear and dresses online the return share is usually above the store average (trying several sizes); for children’s it depends on the child’s height, accessories have no sizes. We give no guide without your own data: take your own statistics by category and channel.
Reverse logistics and handling per return₽250The cost of reverse logistics and handling of one return.
Returns resold at full price% of returns50The rest are sold at a discount or written off. Resale only avoids the discount and the write-off: it does not restore the profit of the sale already reversed. Replace the value with your own (the example is illustrative): it depends heavily on category and season.
Discount on resale (optional)% of price30By how many percent the price is cut when a return is resold; empty counts as zero.
Share written off of those not resold at full price (optional)%10The share of returns not resold at full price that is written off entirely; empty counts as zero.

How it is calculated

StepFormula
Loss per returnReverse logistics + variable + (1 − full-price share) × [(1 − write-offs) × price × discount + write-offs × cost]
Profit after returnsSold × [(1 − rate) × unit profit − rate × loss per return]

Worked numbers

Default values from the app; in the interactive version you enter your own.

Units returned
2,200
Loss from returns
6.36 m ₽Lost profit + reverse logistics and variable costs + discounts + write-offs.
Profit before returns
15 m ₽
Profit after returns
8.64 m ₽
Break-even return rate
51.9 %At this rate profit is zero.

Profit at different return rates

Return rateProfit
5 %13.6 m ₽
10 %12.1 m ₽
20 %9.22 m ₽
30 %6.32 m ₽
40 %3.43 m ₽

Returns in apparel are high: watch the share by size and style. The share resold at full price is a separate indicator of warehouse processing quality.

Nuances and limits

  • Do not count a return only as logistics cost: profit and variable costs are lost and part of the goods goes to markdown.
  • The return rate varies greatly by style and size: calculate by group, not by the average.

Terms used

Calculate on your data →