eCommerce SEO Services

Your category tree is your keyword map whether you designed it that way or not. Most catalogues are structured around how the business thinks about stock — by supplier, by internal department, by warehouse location — rather than around how customers search.

We rebuild the tree around search demand: which attribute combinations have genuine volume, which deserve a dedicated indexable landing page, and which should stay as filter states that are crawlable but not indexed. That usually means creating new intermediate categories that did not exist, and consolidating ones that split demand across three near-identical pages.

Category pages also need content that is not just a product grid. A short useful introduction, buying guidance, and answers to the questions people ask before choosing — placed so it does not push products below the fold.

Faceted navigation is the single largest source of technical damage in eCommerce SEO. Every filter combination generates a URL, and a catalogue with six filters and forty categories can produce hundreds of thousands of crawlable permutations, almost all of them near-duplicates.

The result is crawl budget spent on filter states while genuine product pages go weeks without being crawled, and quality signals averaged across a vast set of thin pages.

Handling it means classifying every facet: high-demand combinations get indexable landing pages with unique content, medium-value combinations stay crawlable with canonicals pointing to the parent, and the long tail is blocked from crawling entirely. Implementation has to be consistent across robots.txt, canonicals, meta directives and internal link nofollow behaviour — inconsistency between those four is where most attempted fixes fail.

Product pages compete on long-tail, brand and model queries, and on rich result eligibility. The three things that matter most are unique descriptive content that is not the manufacturer's supplied copy, complete and valid Product schema, and genuine review content.

Product structured data needs price, currency in GBP, availability, condition, GTIN or MPN where they exist, and aggregate rating where you have genuine reviews. Getting shipping and returns detail into the markup improves how the result presents in Google Shopping surfaces. Invalid or incomplete markup is extremely common and silently costs rich result eligibility across the whole catalogue.

Out-of-stock and discontinued products need a policy rather than ad-hoc handling. Permanently gone products should 301 to the closest equivalent; temporarily unavailable ones should stay live with availability marked correctly. Deleting them and serving 404s across a catalogue throws away accumulated authority.

Each major platform has characteristic problems, and knowing them saves weeks of diagnosis.

• Shopify — forced /collections/ and /products/ URL structure, duplicate product URLs across collections, limited control over robots.txt, and app-injected JavaScript that degrades INP. Most of these have workarounds; some need accepting and designing around.
• WooCommerce — attribute and tag archives generating index bloat by default, plugin conflicts producing duplicate schema, and performance problems on large catalogues that need query and caching work rather than SEO work.
• Magento — layered navigation URL explosion, full page cache behaviour, and complex multi-store canonical configuration.
• Headless builds — rendering verification is essential. What Googlebot receives frequently differs from what the browser shows.

Search volume is a poor prioritisation signal on its own for retail. A term with 200 monthly searches on a £600 product with 40% margin is worth considerably more than one with 4,000 searches on a £12 accessory.

We weight the keyword map by margin, average order value and conversion rate from your existing analytics, so effort goes where it produces profit rather than where it produces sessions. This routinely reorders the priority list substantially against a volume-sorted one.