7 technical SEO audit checks UK ecommerce sites often miss

What to take away

Faceted navigation and tracking parameters are the most common source of crawl waste on UK ecommerce sites.
Parameter handling, canonical signals and internal search controls give the quickest returns.
Site speed fixes only hold once analytics and consent scripts are crawlable.
Server logs and data retention rules rarely appear on generic checklists.

Why UK stores generate more crawlable URLs than they count

A category page with colour, size, price and sort filters can produce thousands of URL combinations. Each one is a candidate for crawling, and Googlebot will follow the links you publish.

How filters multiply crawlable URLs

Category page with colour, size, price, sort
Thousands of URL combinations generated
Googlebot follows every published filter link
Second storefront adds more parameters
Crawl waste delays new product discovery
Link equity splits across near-duplicates

Add a second storefront for Ireland or a .com alongside the .co.uk and the parameter set grows again. Crawl waste follows: new products wait longer to be discovered, and link equity splits across near-duplicate pages. A Screaming Frog crawl budget analysis shows how much of Googlebot's time goes to filters rather than products.

The seven checks

Work through these in order. The first four are usually quick wins.

Seven technical SEO checks

Facet link constructionreal anchors, one canonical
Sort and pagination parameters kept out of sitemaps
Internal search results noindexed and excluded
Multi-region parameter drift checked for hreflang
Site speed and script consent unblocked in robots.txt
Platform sitemap output reviewed for filtered URLs
Crawl before and after each change

Facet link construction. Check whether filter links are real anchors or JavaScript events, then confirm each indexable variant points to one canonical URL. Google's guidance on consolidating duplicate URLs explains how to choose that canonical.
Sort and pagination parameters. Sorting by price or popularity creates near-duplicates of the category page. Keep those URLs out of XML sitemaps and let canonical tags handle them.
Internal search results. Search result pages rarely deserve indexing. Check whether /search?q= or ?s= URLs appear in sitemaps or Search Console impressions, then exclude them.
Multi-region parameter drift. When one platform serves the UK and Ireland, the same filters should generate the same parameters on both storefronts. Inconsistent parameters break hreflang pairing and split signals.
Site speed and script consent. Blocking analytics or consent scripts in robots.txt distorts the field data that feeds Core Web Vitals. The ICO's guidance on storage and access technologies sets out when consent is needed for those cookies. A Core Web Vitals audit checklist lists the metrics to re-test once those files are crawlable.
Sitemap hygiene. Sitemaps should list products, categories and editorial pages only. Anything carrying a filter, sort or session parameter should be excluded and kept out of internal linking.
Server log review. Logs show which parameter patterns Googlebot actually requests, which often differs from what your crawler finds. Retention should follow data minimisation rules under the UK GDPR, because log lines contain IP addresses.

Sorting indexable from disposable URLs

The decision is rarely block or allow. Most stores need a middle path: keep the facets shoppers use and discard the rest.

URL pattern	Action	Why
/shoes?colour=black	Keep and link from the category page	Shoppers use it and it can rank as a landing page
/shoes?sort=price_asc	Canonical to /shoes	Duplicate content with no added value
/search?q=running+shoes	Noindex and drop from the sitemap	Thin pages that generate endless URLs
/shoes?sessionid=abc123	Block by pattern	Session parameters create a new URL per visit
/shoes?colour=black&size=9&sort=price	Canonical or block	Combined facets multiply fast

Test each change with a crawl before and after, and check the JavaScript fallback for filters that load without links. Record the decision for every pattern, so the next developer does not undo it.

Parameter handling beyond the crawl

Campaign parameters are the quiet version of the same problem. Paid and email traffic adds utm_source, gclid and fbclid strings to product URLs, and each variant has its own address. Canonical tags should point to the clean URL, and your analytics should strip the parameters before reporting. Where consent delays analytics, compare the reported sessions against server-side request counts to see the gap.

Platform defaults change the fix

Shopify filter links use query strings such as ?filter.v.option.size. Magento layered navigation generates path segments. WooCommerce extensions often append parameters after the product query. The remedies differ, but the audit question stays the same: is each variant reachable, indexable and canonicalised on purpose? Check the platform's own sitemap output early, because filtered URLs often appear there by default.

Common questions

Do I need to block faceted URLs in robots.txt?

Blocking stops crawling but not indexing, so a blocked facet URL can still appear in results with a poor snippet. Canonical tags and noindex usually work better, and they leave the signal visible.

How often should a UK store review parameter handling?

Quarterly suits most catalogues, plus a review after every platform migration, theme update or new filter release. Filter features change crawl behaviour quickly.

Which tools show facet crawl traps fastest?

A desktop crawler such as Screaming Frog or Sitebulb groups URLs by parameter pattern. Server logs then confirm what Googlebot really requests, which is where the two often disagree.

Does hreflang fix duplicate parameter URLs?

No. Hreflang only tells search engines which regional version to serve. Duplicate parameter URLs still need canonical or robots handling.