Return rates on resale clothing tend to be highest in the middle tiers, not the lowest, because the gap between what a buyer expects and what arrives is widest there. The cleanest way to see it in your own shop is to sort past returns by tier and reason code, then compare.
A note on numbers before we start: we do not publish a universal return rate by tier, and you should distrust anyone who does. Every figure below is an illustrative example, labeled as one, meant to show the shape of the pattern and the method. Your category, price band, and platform will move the real numbers.
Why Condition Tier Changes How Buyers Return
A return is rarely about the garment alone. It is about the distance between the picture in the buyer's head and the item in their hands. Tier sets that picture before the buyer reads a single measurement.
Consider the scenario most of us have lived. You list a jacket as Excellent. It has light cuff wear you considered normal. The buyer expected near-new, files a not-as-described claim, and you eat the return shipping. The jacket was fine. The word was the problem.
Buyers behave differently by tier for three reasons:
- Expectation load. Excellent and NWOT carry a promise of almost no visible wear. Small flaws feel like broken promises.
- Price sensitivity. A buyer paying Good money for a Good item has already priced in wear. A buyer paying Excellent money has not.
- Scrutiny. Higher price means closer inspection on arrival, so more defects get noticed.
What the Pattern Looks Like Across Tiers
Here is an example of how a seller's history might look when sorted by tier. These numbers are invented to show the shape, not to benchmark you.
| Tier | Example return rate | Most common reason | Why it happens |
|---|---|---|---|
| NWOT | Low | Fit | Condition claims are rarely disputed; size is |
| Excellent | Moderate to high | Not as described | High expectation, small flaws noticed |
| Very Good | Moderate | Not as described, fit | Boundary with Excellent is fuzzy |
| Good | Low to moderate | Undisclosed defect | Wear expected; surprises are what trigger returns |
The point is the shape. Excellent can out-return Good, because Good buyers arrive expecting wear. A Good item with every defect disclosed often generates fewer disputes than an Excellent item with one unmentioned flaw.
What Our Data Shows and What It Doesn't
Dispute patterns we see point in a consistent direction: disputes cluster around tier boundaries, especially Excellent versus Very Good, and around defects the listing never named. That is a directional observation, not a guarantee.
Here is what the data cannot tell you:
- Causation. A lower return rate after adopting grades may also reflect better photos or a changed sourcing mix.
- Your category. Denim returns for fit. Knits return for pilling. A blended rate hides both.
- Silent dissatisfaction. Buyers who keep an item and leave a lukewarm rating never show up in a return count.
- Small samples. Twelve Excellent sales and two returns is a 17% rate that could easily be noise.
Treat any published rate, ours included, as a prompt to check your own history rather than a target.
How to Run This Analysis on Your Own History
You do not need special software. A spreadsheet export from eBay, Poshmark, or Mercari is enough.
- Export the last 6 to 12 months of sold orders, including return and cancellation status.
- Add a column for the condition tier you assigned at listing. If you never assigned one consistently, regrade a sample from your photos.
- Add a column for return reason: not as described, fit, changed mind, damaged in shipping, other.
- Group by tier and count sales, returns, and not-as-described claims separately.
- Compute return rate and not-as-described rate per tier. Keep fit returns out of the condition analysis.
- Flag any tier with fewer than 30 sales as low-confidence.
- For each not-as-described return, note which of the five factors the buyer cited: Fabric Condition, Structural Integrity, Cosmetic Appearance, Functional Elements, or Odor & Cleanliness.
Step 7 is where the value is. If most of your Excellent returns cite Cosmetic Appearance, your problem is probably a boundary call on light wear, not your whole process.
Reading the Results and Acting on Them
Once you have a table, three patterns are common.
Returns concentrated at Excellent
You are likely listing Very Good items as Excellent. Tighten the line: Excellent means minimal visible wear across all five factors. If one factor falls short, the tier drops.
Returns concentrated on one factor
If Odor & Cleanliness drives disputes, fix the pre-listing wash and sniff routine rather than the grading scale.
Returns low everywhere but price realization weak
You may be underselling. Grade conservatively, but not so conservatively that an Excellent item sells for Good money.
Whatever you find, pair each grade with a short defect note. A tier alone sets the expectation; the note closes the gap on specifics. Buyers cannot dispute a flaw you named.
Why a Standardized Grade Helps the Analysis Itself
Return analysis only works if your tiers mean the same thing in January and in June. If Excellent drifts depending on your mood or how many items are in the queue, your data is noise. A numerical grade on a 1.0 to 10.0 scale, with a defined tier mapping, gives you a consistent label to analyze against.
That consistency is also what lets a buyer verify your claim. A certificate tied to the grade turns your adjective into something checkable, which tends to reduce the expectation gap before the order ships. We would not promise it eliminates returns. Fit, shipping damage, and changed minds will always exist.
Pick one tier, one category, and run the seven steps above. The result is more useful than any average we could publish.
If you want a consistent starting point, grade one garment with GradeThread and read the five-factor report. Compare it to the tier you would have written yourself. The difference, if there is one, shows you where your own boundary calls drift.