Lead with the products you can sell this week: available in almost every size, in demand right now, and leaving you a decent margin. The order of a collection page is a business decision, and if you don't make it, the default sort makes it for you using yesterday's numbers.
The benchmark
For each product in the collection, three data points are enough: product page views, add-to-cart rate and stock coverage by size. The third is often forgotten, yet none of the automatic sorts Shopify offers looks at your stock.
We ran the exercise on a collection from a clothing brand. Over two weeks, the flagship piece accounted for 40% of the collection's revenue, with 106 units spread across 25 variants, or a little over four per size and color.
Another piece, which sold six times over the same period, had only 13 units left across 8 variants, fewer than two per combination.
That stock let the flagship piece hold the top spot all month, while the other, placed at the same rank, would soon have shown sold-out sizes to a good share of the people who came to see it.
The rule
The first two or three rows on mobile are your storefront. Put in them what you can ship this week, at several price levels, and use the automatic sort to prepare the list while keeping control of the result.
The steps
- The starting snapshot. Write down the first twelve products as they appear today on a phone, with each one's sales for the past month (Shopify's sales by product report is enough) and its stock coverage by size.
- Clicks by position. If your GA4 tracking passes along each product's position in the list, the "Item list position" dimension shows which rows are really working. If not, note that you don't know yet. In one audit of our own, the question "Does the default sort highlight the most-clicked products?" stayed marked "not verified," with only one note: "not enough data to determine."
- Incomplete sizes. Move down any product that's missing the middle sizes, even if it sells well.
- New arrivals. Slip two in among the safe bets in the top rows, so they get enough views before they're judged.
- Prices. Alternate price ranges, so the first screen doesn't show only the high end.
- The routine. Review the order every week with the stock status in front of you, then tell your leadership:
"Our top rows show what we can deliver in every size this week, and I check every Monday that this is still true."
The trap
The "Best selling" sort seems to answer the question, but it addresses a different one. Shopify calculates it from the total number of orders containing the product, over its entire history, with no time limit.
So the hit from two years ago, with only three sizes left, keeps its place ahead of the piece that's selling this season, and the ranking sustains itself, because what's at the top gets seen, and therefore ordered.
We once recommended this default sort to a brand, to reassure its cold traffic, and it was rereading Shopify's documentation that showed us what "best" actually meant.
The other settings have their own blind spots. Sorting by creation date puts products that haven't proven anything yet in the spotlight.
Manual order, for its part, freezes in place: a product added to a hand-sorted collection lands at the end of the list, and the case we run into most often is an order set at launch that nobody has reopened since.
