Most store SEO advice is written for blogs with a shop attached. This is written for stores where the catalogue is the site.
Start from revenue, not from traffic
Pull your organic landing pages and sort them by revenue, not by sessions. Then sort your collections by revenue regardless of channel.
Those two lists tell you where to work. A five per cent gain on the collection that makes a third of your money beats doubling traffic to a page nobody buys from.
Everything below should be applied to the top of those lists first.
1. One keyword group, one page
The most common structural problem in a store audit is two or three collections competing for the same search.
Build a sheet: every commercial keyword group in one column, the single page that owns it in the next.
- Where two pages claim one group, merge and redirect the loser.
- Where a group has no page, that is your build list.
- Where a page owns a group with no search demand, leave it alone and stop optimising it.
This is unglamorous and it is usually the highest value hour in the project.
2. Make category pages real pages
Collection pages target the terms with volume and they outlive individual products. On most stores they are close to empty.
What belongs on one:
- An H1 containing the search term, not the store name
- Two or three sentences above the grid, then the products
- Genuinely useful content below the grid: how to choose, sizing, the questions support answers every week
- Internal links to related collections, the relevant buying guide, and the parent collection
The detail, including the Shopify-specific traps, is in Shopify collection page SEO.
3. Control what your filters create
Five filters with six options each generates more URLs than most stores have products. Add sort order and pagination and the number climbs again.
Google then spends its crawl budget on near-duplicate variations and never reaches the products that make you money. This is the single biggest technical problem in large catalogues.
The three ways to control it, and how to tell whether it is happening to you, are in faceted navigation is eating your crawl budget.
4. Product pages that earn rich results
For your best sellers, in this order:
- Unique descriptions. Manufacturer copy appears on every competitor's site. Rewriting all 5,000 is rarely worth it. Rewriting your top 200 usually is.
Productschema withOffer,AggregateRatingandReview. Validate it. Rich results noticeably lift click-through.- Real review content on the page. It is unique text, it answers buying questions, and it converts.
- Images with alt text describing the product, not the filename.
- Variants handled properly. One page with variant selection, not twelve near-identical pages.
5. Handle out of stock properly
Deleting a product page throws away everything it earned. The decision tree depends on whether it is coming back, and it is covered in what to do with out of stock product pages.
Short version: coming back means keep the page indexed with a notification option. Gone for good means a 301 to the closest equivalent product, not to the home page.
6. Speed
Stores are slow for three reasons, usually all three at once:
- Apps. Every app adds scripts. Audit them. Most stores carry apps nobody has opened in a year.
- Images. Serve AVIF or WebP, sized for the slot they occupy, with explicit width and height.
- Themes. Page builders and sliders cost seconds. A slider above the fold is the most expensive design decision in e-commerce.
Measure with field data from Search Console, not lab scores.
7. Content that connects to a product
Blog content on a store pays for itself only when it feeds a collection.
Worth writing:
- Comparison pages: "X versus Y for Z"
- How to choose guides, linked directly into the collection
- Sizing and compatibility pages, which capture high-intent searches
- Care and maintenance content, which earns links
Not worth writing: general lifestyle posts with no path to a product.
Every one of these should link into the relevant collection with descriptive anchor text, in the body, near the relevant sentence.
8. Links to commercial pages
This is harder than link building for a blog, because editors will not link to a page that only sells.
Two approaches that work:
- Build links to genuinely useful content, then pass authority internally to the collection with contextual links.
- Create something linkable: original data, a calculator, a sizing tool.
The vetting process for suppliers is in how to vet a link building supplier.
9. Do not break it during a replatform
Migrations are the most dangerous thing a store does to its organic traffic, and the failures are predictable. The full checklist is in the e-commerce migration checklist.
The one line version: build the complete redirect map before launch, not after.
What to report
Organic revenue, then organic sessions to commercial pages, then rankings on the mapped keyword groups.
If your report leads with total sessions, it can show a good month during a bad quarter. That is not a reporting style, it is a hiding place.
We run this as e-commerce SEO, usually alongside technical SEO on larger catalogues. The short version of what we would change first is on SEO for e-commerce stores.