Nobody outside a sorting app can see how hundreds of brands set their collection rules, so most fashion stores set theirs on gut feel and leave them. This post is the view from inside one: the store-level sorting settings of a sample of 45 fashion and footwear stores run by 26 brands using Bestsellers reSort, compared with a sample of 2,000+ stores in every industry, from a production snapshot taken on 10 September 2026. It describes what stores set. It does not measure what those settings earn – no A/B test sits behind these numbers, and we say where a figure might be a default rather than a choice.
Six findings, one chart each, a method note at the end, and three conditional suggestions for your own rules.
73% of fashion stores send sold-out products to the bottom, against 44% of all stores
The widest gap in the data. Three in four fashion stores have Push Down Out-of-Stock switched on; across every industry it is fewer than half. Fashion ranges sell through by size and colour, so a collection page without the rule fills its first screen with cards nobody can buy. The rule moves a product once its total inventory reaches zero – not when it is down to its last size; that rule lives in StockIQ, our out-of-stock app.
Share of stores with Push Down Out-of-Stock on. Bestsellers reSort snapshot, 10 Sep 2026; fashion sample of 45 stores, comparison sample of 2,000+ stores.
Fashion sorts by units twice as often: 38% use number of sales as the primary order, 51% revenue
Across all stores revenue dominates, 58% to 19%. In fashion the split is closer to even. Units favour volume styles at lower price points – tees, basics, socks – and keep a $30 tee that sells 200 a month above a $300 coat that sells 30. Revenue does the reverse. Neither is right in general; which one you choose decides what the first row of every collection looks like, so it is worth deciding on purpose.
Primary Sort Order at store level. “Other” is mostly inventory quantity (7% in both groups), then dates, price and manual.
82% of fashion stores look back 30 days or less; 13% look back more than 90 days
The window is the number of days of sales the order is built on. Fashion keeps it short: 47% sit on the 30-day default, 18% on 14 days or less. Across all stores only 54% stay within 30 days and 36% look back more than 90 days. One caveat: 32% of all stores – and 11% of fashion stores – use exactly 180 days, a value that appears as the preset in older screenshots of the app. We have not confirmed how many of those stores chose it and how many never changed it, so we read the long-window share as an upper bound, not a preference.
Lookback Period at store level. Median is 30 days in both groups.
Half of fashion stores sort by tags and two thirds tag automatically
Sort by Tags – a tag that fixes a group of products at the top, after new products, before sold-out, or at the bottom – is on in 49% of fashion stores and 18% of all stores. Automatic tagging (Bestseller, Trending, New arrivals, Aging) is on in 64% against 26%; the Bestseller tag alone in 47% against 22%. In fashion the tags do two jobs at once: they place products in the order, and a theme can read them to print the label on the product card. Trending, New arrivals and Aging are each on in 13% of fashion stores.
Blue: fashion stores. Grey: all stores. Store-level switches.
New products go on top in 40% of fashion stores – for a median of 7 days
Less than we expected. Push New Products Up is on in 40% of fashion stores and 30% of all stores, and where it is on, fashion keeps new products on top for a median of 7 days against 10 days overall. The data does not say why. A plausible reading is that brands with a dedicated New In collection do not want the drop on top of Bestsellers as well – but that is a reading, not a finding. Back-in-stock treated as new is on in 20% of fashion stores against 11%.
Push New Products Up on. Median “new” window where on: fashion 7 days, all 10 days.
The median fashion store runs 265 collections and 3,700 products – three to four times the average
All stores on the app: 73 collections and 921 products at the median. Fashion: 265 and 3,723. Category times gender times season times sale produces collections faster than anyone can sort them by hand, which is the practical reason the settings above are switches and not tasks. It is also why 20% of fashion stores deduct discounts from revenue before ranking (7% overall): with a permanent sale section, gross revenue would rank marked-down stock above full-price sellers.
Collections and products per store, Shopify counts synced to the app.
What to take from it for your own rules
If your range sells through by size and colour, sold-out to the bottom is what three in four of your peers do; the exception is a store that wants sold-out products visible as a waitlist. If your price points span a wide range, revenue keeps the pieces that carry the margin on top; if they sit close together, units is the more honest measure of what people want, and fashion stores choose it twice as often as everyone else. If you run drops every two weeks, a 7–14-day new-in window and a window of 30 days or less for the rest matches the stores in this set; if a style needs a season to prove itself, stretch it, knowing that most of your peers do not. Set the store default once, override Sale and New In, and check one collection after the first cycle – the setup is in how fashion stores order a collection page, and the rules themselves are on the Bestsellers reSort page.
Frequently Asked Questions
What sort order do most Shopify fashion stores use for collections?
In a sample of 45 fashion and footwear stores using Bestsellers reSort in September 2026, 51% order collections by revenue and 38% by number of units sold, over a window of 30 days or less in 82% of cases. 73% move sold-out products to the bottom and 49% use tags to fix groups of products in position. These are settings stores chose, not measured outcomes.
How long should new arrivals stay at the top of a collection?
There is no standard. Among fashion stores that push new products to the top – 40% of the 45 in this sample – the median is 7 days, against 10 days in the comparison sample. A window that matches your drop cadence is the usual choice: a brand dropping every two weeks keeps new-in on top for one to two weeks; a brand with two seasonal collections a year can hold it for a month. Check the collection after the first cycle and adjust.
Do fashion stores sort collections by revenue or by units sold?
Both, and more evenly than other industries: 51% revenue and 38% units in the fashion sample, against 58% and 19% in the comparison sample. Units keep high-volume, lower-priced styles on top; revenue favours the pieces with the highest sales value. 20% of fashion stores also deduct discounts before ranking, so marked-down stock does not outrank full-price sellers on gross revenue.
Method
Source: the Bestsellers reSort production database, snapshot of 10 September 2026, store-level settings only. Per-collection overrides and featured products are configured per collection and are not in this cut. Fashion stores were identified by brand from the app’s install base – 26 fashion and footwear brands, several of them on Shopify Plus – and matched to their stores by domain; 45 stores had collections enabled. The comparison sample is 2,000+ stores in every industry with collections enabled. Percentages are shares of stores. We do not name the brands in the sample. If you want the table behind the charts, ask us through the support chat and we will send it.
