How to use Sorting Optimization
The default sort on your category pages decides which products get seen — and therefore which sell. This guide covers how to pick it, segment it, and test it.
Sorting Optimization
Choosing and testing the default sort order on product listing pages so the right products get visibility and revenue.
Sorting optimization is the practice of deliberately choosing — and continually testing — the default sort order applied to a product listing page (PLP), and the sort options exposed to shoppers. The default matters disproportionately because the vast majority of visitors never change it: whatever order you ship is the order that converts.
In practice it covers four decisions: which products surface in the first two rows, what sort options appear in the dropdown, how sort changes by category or audience, and how you measure success — PLP click-through rate, revenue per PLP visit, and downstream conversion.
Most stores ship with whatever default their platform picked — "Featured" in Shopify, "Menu order" in WooCommerce, "Position" in Magento. None of those map to anything a shopper cares about, and yet they decide which products earn impressions on your highest-traffic pages.
Sorting is the lowest-effort, highest-leverage lever inside PLP optimization. You don't ship new product photography, you don't rewrite copy, you don't redesign the grid. You change a sort key and re-rank what shoppers see first — often worth a 3-8% lift in revenue per PLP visit.
Why the default sort decides what sells
Shoppers scan PLPs the same way they scan search results: the first row gets most of the attention, the second row gets a fraction of that, and anything below the fold gets a long-tail of clicks from determined buyers. Position is exposure, and exposure is sales.
Across typical Shopify and Magento stores, the first eight tiles on a PLP take 55-70% of all product clicks. Tiles 9-24 share most of the remainder. Anything past tile 40 might as well not exist for the average session — it gets <2% of clicks.
This is why the default sort matters more than the sort dropdown. Fewer than 15% of visitors ever change the sort. Whatever order you serve by default is, for practical purposes, the only order the catalog has.
The 80/15 rule
On a typical category page, ~80% of clicks land on the first two rows and ~15% of visitors ever touch the sort dropdown. Your default isn't a default — it's the experience.
Choosing the right default sort
There is no universally correct default. The right one depends on category breadth, margin spread across SKUs, and how confident a shopper is when they land. A 200-SKU apparel category and a 12-SKU electronics accessories category need different defaults.
Five candidate sorts cover almost every case: best-sellers, newest, relevance (for search-driven PLPs), price-low-to-high, and a custom merchandising score that blends conversion rate, margin, and inventory. The custom score wins most A/B tests because it lets you weight what you actually care about.
Share of PLP clicks by tile position
Best-sellers as a default is the safest starting point: it surfaces social proof, it tends to surface higher-converting SKUs by definition, and it self-corrects as buying behavior shifts. Its weakness is staleness — without a recency decay, the same hero SKUs sit on top for months and new launches starve.
Segmenting sort by category and audience
One default across the entire catalog is leaving money on the table. "New arrivals" benefits from sort-by-newest. "Sale" benefits from highest-discount-first. A core evergreen category benefits from best-sellers. Set the default at the collection level, not site-wide.
Returning visitors and first-time visitors also want different things. Returning shoppers respond well to newest-first ("what's changed since I was last here?"). First-timers respond to best-sellers or curated relevance. If your platform supports it, segment the default by visitor cohort.
Typical revenue-per-PLP-visit lift from changing the default sort (vs platform default)
| Default sort | Apparel | Beauty | Home & garden | Electronics |
|---|---|---|---|---|
| Best-sellers | +6 to +9% | +5 to +8% | +4 to +7% | +3 to +5% |
| Newest (with decay) | +3 to +5% | +4 to +6% | +1 to +3% | +1 to +2% |
| Custom merchandising score | +8 to +12% | +7 to +11% | +6 to +9% | +4 to +7% |
| Price low-to-high | -2 to +1% | -3 to 0% | -1 to +2% | 0 to +3% |
| Highest margin first | +2 to +5% | +3 to +6% | +2 to +4% | +1 to +3% |
Price low-to-high looks shopper-friendly and often is — for considered-purchase categories with wide price spreads. But on average-order-value-sensitive verticals it crushes AOV by promoting the cheapest SKU first. Reserve it as a user-selectable option, not a default.
Testing your sort changes
Sort changes are unusually clean to A/B test: split traffic at the PLP, hold all other elements constant, and measure revenue per PLP visit as your primary metric — not click-through rate. CTR can go up while revenue goes down if you've surfaced cheaper items first.
Run each test for at least two full weekly cycles. Sort behavior is sensitive to day-of-week (weekends skew toward browse, weekdays toward intent) and to the inventory mix that day. Cutting a test short on day 4 because it looks good is the most common way teams ship false positives.
Don't measure CTR alone
Click-through rate is a vanity metric here. A sort that pushes cheap impulse-buy items to the top will boost PLP CTR and tank AOV. Always pair CTR with revenue per PLP visit and downstream conversion rate.
Frequently asked questions
For most Shopify stores, best-sellers (with a 30-day rolling window so new products don't get buried forever) beats the platform's "Featured" default by 4-8% on revenue per PLP visit. Override it at the collection level for "New" and "Sale" collections, where newest and highest-discount-first respectively perform better.
PLP optimization covers everything on the listing page — filters, facets, grid density, image quality, lazy-load behavior, badges, and sort. Sorting optimization is the specific sub-discipline of choosing what order products appear in. It's usually the highest-ROI lever inside PLP optimization because it requires no design or dev work.
Keep the dropdown — removing it costs you the 10-15% of shoppers who do change sort, and those shoppers convert above average because they're filtering with intent. Just make sure your default is strong enough that the other 85% don't need to touch it.
A weighted formula that ranks each SKU on a blend of conversion rate, gross margin, inventory level, and recency. Typical weights: 50% conversion, 25% margin, 15% recency, 10% inventory cover. It's the highest-performing default for most catalogs because it optimizes for what you actually care about — profitable revenue, not just clicks.
Two full weekly cycles minimum, and ideally until you reach statistical significance at 95% confidence on revenue per PLP visit. For a category with 5,000+ weekly PLP sessions, that's usually 14-21 days. Sub-1,000-session categories may need 4-6 weeks or to be tested as part of a multi-category rollout.
Only indirectly. Search engines crawl the default-sorted view, so products on page one in the default sort get crawled and indexed faster. If you bury high-converting SKUs on page 5, you'll see slower indexation. Use canonical tags to point sorted variants back to the default URL.
Push them down — never up, never out. Hiding out-of-stocks loses you SEO equity and confuses returning shoppers; surfacing them at the top tanks PLP conversion. The standard pattern is to keep them indexed but rank them after all in-stock SKUs in every sort order.
Often yes. Paid traffic from a specific product ad benefits from a relevance-weighted sort that surfaces visually similar SKUs. Organic category traffic benefits from best-sellers. If your platform can read the referrer, segment the default — it's typically worth 2-4% on paid-traffic conversion.
Same default, but compress aggressively for mobile. The first-row advantage is even stronger on mobile because tiles are larger and stacked vertically — tile 1 alone can take 25%+ of mobile clicks. Make sure your top SKUs render above the fold with no scroll required.
Three signals: (1) PLP-to-PDP click-through below 35% on category pages, (2) more than 25% of PLP visitors actively change the sort, (3) the top eight tiles in your default sort don't include your top eight revenue SKUs from the last 30 days. Any one of these is a strong test candidate.
Get an AI expert review of your site
Paste your URL — Metricuno's AI runs the same heuristic checks a senior CRO consultant would, scoring your page and prioritising the fixes that'll move conversion fastest.