Customers Order Two Sizes and Return One: Diagnose Clothing Bracketing

When a customer buys the same garment in two sizes and intends to return one, they are using delivery as a fitting room. This is commonly called bracketing. The useful question is which uncertainty caused that order, and whether the product page could have resolved it before checkout.

A size recommendation may help someone choosing between M and L. A visual preview may help someone unsure about a silhouette or colour. Neither fixes a late delivery, a defective seam or a misleading fabric description. Start with the reason, then choose the intervention.

I am the founder of aisthetix, which provides virtual try-on and size recommendations. This guide includes reasons to improve product information before installing an app. It is a diagnostic method, not a claim that our product has reduced a particular merchant's returns.

Identify candidates without labelling customers

Review order lines for the same product and colour purchased in different sizes within one order. Flag these as possible bracketing, not proven intent: the order might be for two people, a gift or a genuine repeat purchase.

Link those order lines to eventual returns and their recorded reasons. Keep customer names and contact details out of the analysis sheet. Product, size, delivery cohort and return outcome are enough for an initial aggregate investigation.

Separate colour testing from size testing. A customer comparing black and navy has a different question from someone comparing a medium and a large.

Use a consistent denominator

Define the measurement before changing the page:

• Possible bracketing order share: orders containing multiple sizes of the same product and colour divided by all orders containing that product. This does not prove every flagged order was intended to produce a return.
• Size-related unit return rate: units returned as too small, too large or fit/shape divided by delivered units in the same cohort. This does not prove the size chart caused the returns.
• Net retained units: delivered units minus returned units after the observation window. This does not establish incremental demand caused by your change.
• Contribution after returns: retained sales less product, fulfilment, return, payment and intervention costs. Results are not comparable if cost definitions change between periods.

Choose one cohort definition, such as delivery week, and keep it fixed. Give each cohort the same time to return. Comparing last week's deliveries with last month's settled returns makes the recent group look artificially good.

Download the blank product audit worksheet. It contains no customer information or example results. Fill it with aggregate counts from your store and document missing return reasons. The worksheet is a collection template; calculate the measures above in your spreadsheet.

Match the question to the page change

“Which size should I choose?”

Check whether the chart contains body measurements or finished garment measurements, whether units are explicit, and whether it belongs to that product. Explain the intended fit and how to measure. An oversized cut needs different guidance from a close-fitting one.

If accurate charts still leave shoppers uncertain, test a size recommendation tool against those charts. Record the required shopper inputs and how it handles missing measurements. A recommendation should not present an estimate as a guarantee. See our guide to garment measurements and size charts.

“Will that shape or colour suit me?”

Start with representative product photography and descriptions of fabric and cut. A shopper photo preview can add context. Inspect whether the result preserves the garment's neckline, sleeves, hem, pattern and colour before showing it to customers.

A convincing image is not a physical fitting test. The distinction between appearance, fit and trust matters when deciding what you promise.

“Will it feel or behave as expected?”

Describe stretch, lining, opacity and fabric weight where you can verify them. Show movement with product video. Neither a size label nor a generated photo reliably communicates every material property. If the return reason is “not as described,” improve the description before attributing the problem to size selection.

Test one change on a defined set of products

Pick a few products with enough settled orders to investigate. Document the current page and your selection criteria. Choose one intervention and a primary outcome before launch. Keep the return window, cost definitions and product grouping consistent.

If you can run a properly designed randomized experiment, decide allocation and analysis in advance. Otherwise treat a before/after comparison as observational. Promotions, stock availability, customer mix and seasonal demand can change alongside your page.

Watch conversion, contribution after returns, support contacts and technical performance. Fewer returns accompanied by fewer profitable orders may not be an improvement. A lower percentage based on a small sample may simply be noise.

Decide whether a fitting tool belongs in the test

Evaluate the job you need done, not just the phrase “virtual fitting room.” Some products emphasize size advice; others emphasize photo previews; some combine them. Ask every supplier the same questions:

1. What shopper and product inputs are required?
2. What happens when a chart, photo or product is unsupported?
3. What does the dashboard count, and over which attribution window?
4. Are returns included, or only purchases?
5. What will your expected usage cost, including overages?

aisthetix is one option for a Shopify or WooCommerce merchant who wants a photo preview, size guidance and an effectiveness dashboard. Our purchase attribution is observational and is not a returns report. Match those events with your own settled return data before drawing conclusions about bracketing.

Use the comparison hub to build a shortlist and the measurement guide to evaluate it. Choose an intervention that addresses the uncertainty you found, at a cost justified by your own results.

Sources and scope

Shopify documents return reasons and the returns workflow and how to use returns information. These are operational references, not evidence that a particular app prevents returns. The worksheet and diagnostic sequence above are our proposed method; no merchant outcome or controlled experiment is being reported.