The Repeat-Sale Index: Which Brands and Categories Convert Best and Why
Category beats brand almost every time. A pair of straight-leg denim in Excellent condition will usually outsell a logo hoodie from a hyped streetwear label, because buyers search categories and sizes first and brand names second — unless that brand is a top-20 name with strong resale demand. If you want to know which clothing brands and categories sell fastest, stop asking "is this brand good?" and start asking "what's the days-to-sell and repeat-buy rate for this category at this condition tier, in my store?" That number is your repeat-sale index, and you can build one in an afternoon with data you already have.
Most resellers make sourcing decisions off memory and vibes: "denim always sells," "I never touch dresses," "that brand is dead now." Memory is biased toward your best and worst outliers, not your averages. A repeat-sale index replaces the vibe with a number pulled from your own sold history — the only dataset that actually reflects your store, your pricing, and your buyer pool.
What a Repeat-Sale Index Actually Measures
A repeat-sale index is a simple table that tracks, per category and per brand, how often an item type sells again and again at acceptable margin and speed. It has three core metrics:
- Days-to-sell (DTS): median days from listing live to sold, by category/brand/condition combination.
- Sell-through rate (STR): percentage of listed units in that segment that sold within a fixed window, usually 60 or 90 days.
- Repeat-buy rate: how often you've resourced and relisted the same brand/category combination because it kept converting — a proxy for demand durability, not a one-off fluke sale.
None of these numbers matter in isolation. A brand with a 9-day DTS but only three units sold in six months is a sample-size trap, not a trend. You want segments with both a fast DTS and a sample size north of 15–20 units before you trust the pattern enough to change your sourcing.
How to Build Your Own Repeat-Sale Index
You don't need new software to start. A spreadsheet and your last 90–180 days of sold history is enough to get a usable first pass.
- Export your sold listings from eBay (Seller Hub > Orders > Download report) and any other platforms you sell on.
- Add columns for category, brand, size, condition tier (NWT, NWOT, Excellent, Very Good, Good, Fair, Poor), list date, and sold date.
- Calculate days-to-sell for each row by subtracting list date from sold date.
- Group rows by category + brand + condition tier using a pivot table, and calculate median DTS and count of units per group.
- Pull your active/expired listing counts for the same period to calculate sell-through rate per group (units sold ÷ units listed).
- Flag any group with 15+ units and a DTS below your store-wide median as a "repeat" segment — these are your reliable converters.
- Flag any group with 15+ units and a DTS above your store-wide median, or an STR under 40%, as a drag segment — candidates to cut from future sourcing or reprice aggressively.
[Screenshot placeholder: pivot table showing category/brand groups sorted by median days-to-sell, with a highlighted "repeat" tier above the store median line]
Rerun this every quarter. Demand shifts with season, trend cycles, and your own store's growing feedback score — a repeat-sale index from a year ago is a history lesson, not a sourcing plan.
Category Velocity: A Working Comparison
Every store's numbers will differ, but the pattern below is representative of what shows up across mid-volume resale operations once you run the pivot. Use it as a sanity check against your own output, not as a substitute for it.
| Category | Typical median DTS | Typical STR (90-day) | Why it moves this way |
|---|---|---|---|
| Straight/bootcut denim | 10–18 days | 65–80% | High search volume, forgiving fit tolerance, brand-agnostic buyers |
| Outerwear (coats, jackets) | 14–25 days | 55–70% | Seasonal spikes, higher price point slows impulse buys but raises margin per sale |
| Basic knitwear/sweaters | 12–20 days | 60–75% | Repeat-buy category — buyers restock every fall regardless of brand |
| Graphic/vintage tees | 7–15 days | 70–85% | Fast at low price point; margin depends entirely on sourcing cost, not brand |
| Dresses (occasion/formal) | 25–45 days | 35–50% | Size- and fit-specific, narrow buyer window, condition sensitivity is high |
| Activewear/athleisure | 10–16 days | 65–80% | Consistent year-round demand, low return-to-condition friction |
| Formal suiting/blazers | 30–60 days | 25–40% | Narrow size bands, niche buyer, but high per-unit margin when it hits |
Notice what's missing from this table: brand name. That's deliberate. Category and condition tier explain most of the variance in velocity. Brand mostly acts as a price multiplier within a category, not a velocity driver — with the exception of a small handful of brands that have genuine cross-platform search demand (think the names that show up unprompted in eBay's autocomplete for a category).
Why Condition Tier Moves the Needle More Than Logo
Here's the part most sourcing guides skip: within any category, condition tier does more to predict days-to-sell than brand does. A pair of no-name straight-leg jeans graded Excellent will often out-convert a recognizable brand name graded Good, because buyers filter on condition language in the description before they commit — and a vague or inflated condition claim is the single biggest driver of hesitation, or worse, a return.
When you're scoring items for your own index, grade consistently using the same five factors every time: Fabric Condition, Structural Integrity, Cosmetic Appearance, Functional Elements, and Odor & Cleanliness. If your condition tagging is inconsistent — one lister calls a pilled cardigan "Very Good," another calls the same defect level "Excellent" — your repeat-sale index will show noise instead of signal, because DTS differences will reflect grading inconsistency, not real buyer behavior. This is the same problem that drives "not as described" returns, and it's why standardizing condition language (NWT, NWOT, Excellent, Very Good, Good, Fair, Poor) before you build your index matters as much as the sourcing decision itself. GradeThread's AI grading exists specifically to remove this variance — a consistent 1.0–10.0 grade and condition report means your DTS data reflects true demand, not grading drift between whoever listed the item that week.
Turning the Index Into a Sourcing Plan
Once you have real numbers, the index should change three things:
- What you pick up at the source. If your denim segment shows 14-day DTS and 75% STR at 20+ units, that's a green light to buy more of it even at a slightly higher acquisition cost — the turnover carries the margin.
- What you deprioritize. Formal suiting at 45+ day DTS and 30% STR isn't dead, but it's a bin-space liability unless your per-unit margin justifies the hold. Cap how much shelf space you give it.
- What condition tiers to accept. If Fair-and-below items in a category consistently sit past 45 days regardless of price cuts, stop sourcing that condition tier in that category — the labor to list it costs more than the eventual sale.
Run this loop quarterly: index, adjust sourcing weight, reindex next quarter, compare. Three cycles in, you'll have a sourcing playbook built entirely from your own sold data instead of secondhand advice about which brands are supposedly hot.
When a Spreadsheet Stops Being Enough
A pivot table works fine under a few hundred SKUs a quarter. Past that, most resellers hit the same wall: the export lags, the categories don't map cleanly across platforms, and by the time the report is built, the quarter it describes is already over. If you're crosslisting across eBay, Poshmark, and Mercari, you're also reconciling three different category taxonomies by hand just to get one clean pivot.
That's the threshold where inventory-ops tooling earns its keep — automatically tagging category, brand, and condition tier at intake, then rolling up days-to-sell and sell-through by segment without a manual export every time you want to check your numbers.
Try It on Your Own Inventory
Pull your last 90 days of sold data this week and run the seven-step pivot above on just two categories you source heavily. You'll likely find at least one segment you've been overbuying and one you've been underbuying. If you want that analysis running continuously instead of quarterly — with condition grading standardized at intake so your DTS numbers reflect real demand, not grading drift — that's exactly what FlipDesk's inventory ops module is built to track. Grade one batch, list it, and watch where it lands in your own index.