Problem
High ecommerce return rate: how to fix it
Returns are the quietest drain on ecommerce profit because they arrive weeks after the revenue was celebrated. The instinct is to tighten the policy, which reduces returns and reduces orders with them. The better sequence is to record why each item comes back, find the products and the reasons that dominate, and remove the causes. Most of what drives returns is a mismatch between what the shopper expected and what arrived.
By CartKernel · Last reviewed 2026-09-07
Does this look familiar?
- Contribution after returns is far below what the revenue reports suggest
- A small number of products account for most of the returned units
- Returns cluster in one size, one colour or one variant
- Return rates rose after a supplier change, a photography update or a campaign
- Customers order several sizes of the same item intending to keep one
- Return reasons are recorded as other, or not recorded at all
Causes, ranked
Why it happens, most common first
Check them in this order. The first two account for most cases we open.
- most common
The product page sets the wrong expectation
Colour that photographs differently from reality, missing dimensions, no scale reference, or a description that overstates a feature all produce a product that is not what the customer pictured. This is the largest cause across most categories and the most fixable.
- most common
Sizing information is thin or inconsistent
A generic size chart, no measurements taken from the actual garment or item, and no guidance on fit relative to other brands leaves the shopper guessing. Guessing means ordering two sizes, which turns one order into one order and one return by design.
- common
Advertising promises more than the product delivers
Creative that shows a use case the product does not really serve, or copy that implies a result it cannot produce, brings buyers with the wrong expectation. Return rate by acquisition channel usually makes this visible immediately.
- common
Quality or consistency varies between batches
A supplier change, a new production run or a substituted component can shift fit, finish or performance without anything on the site changing. Returns rise for that product only, starting from a datable point.
- common
The item is damaged in transit
Underspecified packaging, items shipped without protection, or a carrier handling issue produces returns recorded as damaged or faulty. This concentrates in specific products and sometimes in specific routes, which is how you find it.
- occasional
Delivery took longer than the customer accepted
Late arrivals lead to returns for items bought for an occasion or bought impulsively. The product was fine; the timing was not. Return reasons and delivery times correlated together will show this clearly.
- occasional
The policy encourages ordering to try
Free returns with a long window is a legitimate commercial choice that increases orders and increases returns together. It becomes a problem only when the return cost is not modelled into the margin on the products where it happens most.
The fix
In this order
Each step is something you can do today. Do them in sequence; skipping ahead is how a review fails twice.
Prevent it next time
- Require a structured return reason on every return and review the mix monthly
- Set a photography and measurement standard every new product must meet before launch
- Watch return rate by product for a datable change after any supplier or batch change
- Report contribution after returns rather than revenue when judging product and channel performance
Record a real reason on every return
Replace a free text box with a short list of specific reasons, including too small, too large, not as pictured, damaged, arrived late and changed mind. Make it required. Without this data every other step is guesswork.
Report returns by product, variant and channel
Build a view of return rate and return cost by product, by variant and by acquisition channel over a period long enough to include the return window. The concentration is usually severe, which makes the work list short.
Fix the pages behind the worst offenders
For the products driving most returns, rephotograph with accurate colour and a scale reference, add measured dimensions, state materials and weight, and describe fit relative to something the customer already knows.
Build sizing guidance from your own returns data
Use the too small and too large split per product to publish honest guidance, such as advising a size up on a specific item. Add measurements taken from the product itself rather than a generic chart, and show what the model or reference is wearing.
Align advertising with what arrives
Review the creative and copy for the channels with the highest return rates and remove anything the product cannot support. Check landing pages match the ad, since a mismatch between the two produces buyers with a different expectation entirely.
Investigate quality and packaging where reasons point there
For products with damage or fault reasons, review packaging specification, test a shipment through the actual carrier route, and take the batch data to the supplier. Track whether the rate changes after each intervention.
Offer exchange before refund
Make exchange or store credit the easiest path in the returns flow, with the correct size or an alternative suggested. This retains revenue on returns caused by fit rather than by dissatisfaction, without making the process harder.
Model return cost into product decisions
Calculate contribution after returns per product, including shipping both ways and handling, and use that figure in merchandising, pricing and advertising bids. Products that look profitable on revenue and lose money after returns should not be scaled.
When to get help
It is worth bringing help in when returns are eating a meaningful share of margin and nobody can say which products cause it, because the first job is building the reporting rather than changing the policy. Once the data exists, the work spans product content, advertising creative, packaging and supplier conversations, so it sits across teams. Help is also useful for modelling how a policy change would affect orders and returns together, since tightening one without forecasting the other usually costs more revenue than it saves.
Free. We reply within one business day.
Should I make my returns policy stricter to reduce returns?
A shorter window or paid returns will reduce returns and will also reduce orders, because a generous policy is part of why people buy. Model both effects before changing anything. Reducing the causes of returns keeps the orders and removes the cost, which is almost always the better trade.
How should return rate be calculated?
Measure returned units against units sold in the period those units were sold, not against units sold in the period the returns arrived. Returns lag orders by weeks, so comparing the two within one month understates the rate and makes recent performance look better than it is.
Do product videos and detailed images reduce returns?
They tend to, because most returns come from a gap between expectation and reality, and richer media closes that gap. The effect is strongest where fit, scale, texture or colour matter. Measure it per product by comparing return rate before and after the media is improved.
Should returns be deducted before judging advertising performance?
Yes, if the channel mix affects the return rate, which it usually does. Judging paid performance on gross revenue rewards channels that bring buyers who send items back. Report return adjusted revenue and contribution per channel so budget decisions reflect what the business keeps.
Services that fix it for good
Related problems, answers and terms
Stop the leak.
A free Growth Analysis finds what is broken and ranks it by the revenue it costs you.