How do I reduce returns?

Sort by volume first. A product with a poor return rate that sells twice a month costs you almost nothing. A product with an average rate that sells every day is where your money is going.

Export a quarter of returns, group by product, and rank by units returned. In most catalogues a small number of products account for the majority of the problem, which means this is a short project rather than a catalogue wide one.

For each of the top few, do three things. Read every free text comment customers left, because they usually name the problem plainly and repeatedly. Look at the reason breakdown for that specific product rather than for the store. And buy your own product as a customer, unannounced, so you see the page, the parcel and the item exactly as they do.

That exercise usually produces an obvious list. The colour is not what the photo shows. The item runs small. The description omits that assembly is required. The box arrives crushed.

Fix those, then re-measure after a full cycle. Returns lag orders by weeks, so give the change time before judging it.

A measurement table in centimetres is necessary and it is not enough, because most people do not measure themselves. What reduces sizing returns is guidance in the language customers use.

Say how the item fits relative to expectations: true to size, runs small, generous through the shoulder, cut short in the body. Say what to do if someone is between sizes, and be specific about which way to go for that garment rather than giving one blanket rule.

Put the model's measurements and the size worn next to the photograph. It is one line and it answers the question people are actually asking.

Use review data. If several reviews mention the fit, surface those next to the size selector rather than burying them at the bottom of the page. A sizing note from a previous buyer at the point of choosing is worth more than anything you write yourself.

For non apparel, the equivalent is scale and compatibility. Show the product next to a common object. State the dimensions assembled and boxed. Say which models, sizes or standards it fits, and say plainly which it does not.

The pattern is the same: answer the question that causes the wrong choice.

Most not as described returns are honest misunderstandings created by a good looking page.

Colour is the usual offender. Screen colour varies, heavy editing makes it worse, and a saturated hero image produces a parcel that disappoints. Shoot in consistent, neutral light, avoid pushing saturation, and include at least one image in daylight without styling.

Texture and material need close ups. A shopper cannot feel the fabric, so a macro shot of the weave, the grain or the finish does the work.

Scale needs a reference. The single most common surprise in homeware and accessories is that the item is smaller than imagined. Show it in a hand, on a table, in a room.

In the description, write the things that stop the wrong person buying. What is in the box and what is not. Whether assembly is needed and how long it takes. Care requirements. Battery type. Whether it is suitable for outdoor use.

Saying who the product is not for feels like losing a sale. It loses the sale that would have come back, and it keeps the one that would not.

Two categories of return are entirely operational and both are fixable.

Damage in transit. If a product shows up repeatedly in damaged returns, the packaging is wrong for the carrier network it travels through, not for the shelf. Test by shipping a few to yourself across the routes you actually use. Fixing the pack size, the void fill or the box strength is usually cheap compared to the returns it prevents.

Late arrival. When something misses the occasion it was bought for, it comes back regardless of quality. So be conservative and specific about delivery dates on the product page, show a cut-off time for same day dispatch, and update it when the cut-off passes. A promise you comfortably beat generates fewer returns than one you occasionally miss.

Wrong item and missing parts belong here too. These are picking accuracy problems, and the fix is in the warehouse process rather than in marketing, but the cost lands on the same line.

And check the order confirmation. A clear summary of what was ordered, in which size and colour, gives the customer a chance to correct a mistake before the parcel is packed, which is the cheapest return there is.

Returns are going to happen, so use them. A return with no reason attached is a cost with no information.

Ask one required question with a short list of options at the moment the return is started, plus an optional comment box. Keep the list to five or six reasons so people choose honestly rather than picking the first item to get through the form.

Offer an exchange before a refund, with the alternative size or colour preselected where you can work it out. Many sizing returns are customers who still want the product, and an exchange keeps the revenue and the customer.

Make the process easy despite all of the above. A hard returns process suppresses the number and damages repeat purchase, which costs more than the refunds it avoids.

Then close the loop monthly. Take the top reasons by product, decide one change for each, make it, and note the date. Compare the following quarter against the same products.

That monthly habit, applied to a handful of products at a time, does more than any single policy change, and it compounds because every fix stays fixed.