Product and Category Page Optimisation

Commercial queries (“women’s waterproof running shoes”, “oak coffee table under £400”) resolve to category or product pages, not blog posts. At small scale, optimising each page individually is feasible. At scale, the challenge shifts to systems: how to maintain quality and relevance across thousands of URLs generated automatically by a CMS or ecommerce platform.

Why are category pages the primary ranking target?

Category pages sit at the intent level where most commercial queries resolve. “Running shoes”, “coffee tables”, “men’s winter jackets” are category-level searches: the user has not yet settled on a specific product. A well-optimised category page ranks for these terms and directs traffic toward product pages that serve transactional intent.

Category pages earn external links more readily than product pages. A journalist or blogger linking to “best coffee tables” will link to a category page; they would not link to a specific product SKU. This makes category pages the correct target for link building activity.

Internal linking should reinforce this. Link from editorial content and pillar pages to category pages, and from category pages to product pages, so that the pages you most want to rank receive the most internal links. Links from product pages back to supporting content (a size guide, a care guide, a buying guide) are fine and often useful to the buyer; the point is that editorial content should not be the main destination of your commercial pages’ internal links.

What does a well-optimised category page need?

A category page needs:

Unique introductory copy placed above the product grid. 100 to 200 words introducing the category, addressing common buyer questions, and giving search engines contextual signals. Google’s stated position is that content inside tabs, accordions and “show more” toggles is indexed and weighted normally under mobile-first indexing, provided it is present in the served HTML. Its mobile-first guidance goes further and recommends the pattern: rather than removing content on mobile, “consider moving content into accordions or tabs to save space”.¹ Content fetched on click is a different case. Google “does not interact with your page”, so a “show more” that requests content on interaction leaves that content uncrawled.² Asked directly whether such content would still be devalued, John Mueller answered: “No. Specifically when it comes to content on mobile pages we do take into account anything that’s in the HTML.”³ Hiding copy behind a toggle is not the ranking penalty it is often claimed to be, but the evidence is not all one way. In a split test on Iceland Groceries, SearchPilot brought product information such as ingredients and nutrition facts out of tabs and accordions so that it rendered on load, and measured a 12% uplift in organic sessions, statistically significant at 95% confidence and larger on mobile than on desktop.⁴ That is one retailer’s product pages, published in 2020 by the company whose platform ran the test, so it is a single controlled result rather than a general rule. The safe reading is unchanged: make the copy that carries your key context visible on load. A reader who never expands the toggle never sees it either.

Descriptive title tags and meta descriptions. “Women’s Waterproof Running Shoes | 120+ Styles | FreeRun” outperforms “Women’s Shoes | FreeRun”. The title tag should include the primary query term and, where it adds value, a differentiating qualifier: range size, a service promise, brand range.

Structured headings. H1 matching the category. H2s for subcategory sections or featured filter groups where they represent meaningful demand clusters.

Internal links to subcategories and products. Category pages should link through to their child subcategories and to the products themselves. Google is direct about the requirement: if category pages do not link to every product, “Googlebot might not find all of your products by crawling alone”, because it “generally doesn’t try to submit searches into a search box as part of crawling a site”, and it “strongly recommended” linking to every product you want indexed, falling back to a sitemap or a Merchant Center feed where that is not possible.⁵ These links also distribute link equity to deeper pages.

BreadcrumbList structured data. Google lists it first among the structured data types relevant to ecommerce sites, and a category page is where it does its work, stating the page’s position in the hierarchy.⁶

These are the same elements AI shopping agents read when selecting products, since they work from the served HTML rather than a rendered page. See AI shopping agents for the selection signals specifically.

Why won’t a category page win the shopping carousel?

Because that block is not fed by organic ranking at all. The product listings and shopping carousels that appear on commercial queries are populated from a Google Merchant Center product feed, including the free listings tier, and clicking one goes to the retailer’s product page. Google is explicit about which page type that is: free listings “help your product pages to appear in search results, not other pages like homepages, delivery and return policies, or product category pages”.⁷ No amount of category-page optimisation places you there, and no Merchant Center feed keeps you out of organic results.

So a commercial query is often two competitions rather than one, and they are won with different assets. The organic slot below the carousel is won by the work on this page. The carousel itself is won by feed quality: accurate titles, prices, availability and identifiers, kept in sync with the page. A product page and its feed entry disagreeing on price or stock is a feed problem, not an SEO one, and it is the more common failure.

Both still rest on the same underlying product schema, so the two are worth maintaining together rather than treating the feed as a separate system owned by someone else.

How do you handle product page content at scale?

The most common product page problem is thin or duplicated content. Manufacturer-supplied copy is shared across every retailer stocking the same product. A product page using only manufacturer copy offers nothing unique for search engines to prefer over competitors using the same copy.

At scale, fully unique descriptions for every product are rarely feasible. A tiered approach works better:

High-priority products (bestsellers, high-margin, actively promoted): bespoke descriptions that answer buyer questions the manufacturer copy does not address: dimensions, compatibility, real-world use cases, what differentiates this product from similar options.

Mid-tier products: a template framework that forces meaningful differentiation on key fields (key feature, who it is for, key dimension, compatible with) while keeping structure consistent across the range.

Long-tail or low-traffic products: manufacturer copy with complete structured data is acceptable, provided variant pages that duplicate content without adding value are handled with canonicals.

The test for any product description: does it tell a buyer something they could not learn from the image and spec table alone?

How do you handle product variant URLs?

Colour, size, and configuration variants generate separate URLs on most platforms. “Running shoes in black” at /products/running-shoe-x?colour=black is not a distinct product: it is the same product in a different colour.

Unless a specific variant has its own demonstrable search demand, variants should point their canonical tag to the main product URL. This consolidates link equity and prevents duplicate content across potentially hundreds of near-identical pages.

When a variant carries its own demand, which is common in electronics where storage configurations (iPhone 16 Pro 256GB) carry distinct search volume, giving the variant its own indexed URL with unique, differentiated content is the better approach.

Whichever way the canonical goes, mark the relationship up. Google documents ProductGroup for variants specifically, alongside Product, and lists both among the structured data types relevant to ecommerce sites.⁶ The markup tells Google that a set of URLs are variants of one product rather than separate products that happen to look similar, which is the same judgement the canonical is making.

What is the right approach to pagination?

Google announced in March 2019 that it no longer uses rel="prev" and rel="next", and its documentation still notes that other search engines may use them.⁸ John Mueller said at the time: “We noticed that we weren’t using rel-next/prev in indexing for a number of years now, so we thought we might as well remove the docs”, adding that “for the most part, we just index the pages as we find them, so … it’s good to make sure that all pages can stand on their own”.⁹ That is the working principle: treat each paginated category page as an independently indexable URL carrying a self-referencing canonical.

The points that matter most on a large catalogue:

• Link from each page to the next with a plain <a href>, and from the deeper pages back to page one. Google names this first among its pagination best practices: the anchors are how Googlebot finds subsequent pages, and the links back signal which page is the start of the collection.⁸
• Titles and descriptions do not need to be unique across the sequence. Google differentiates paginated pages itself and states that you can reuse the same titles and descriptions on every page in the set.⁸
• Do not canonicalise all paginated pages to page one. Google’s instruction is direct: “Don’t use the first page of a paginated sequence as the canonical page. Instead, give each page its own canonical URL.”⁸ Because each page shows different products, the pages are not duplicates, and a canonical that claims they are is the kind Google is free to disregard, so the signal consolidation it is meant to deliver should not be relied on.
• Keep filtered and alternative sort orders out of the indexed set. Google lists this among its pagination best practices, handled with noindex or a robots.txt disallow, and it is where pagination meets faceted navigation.⁸
• Put the descriptive category copy on page one only. That is what makes page one the page matching the category query, rather than leaving Google to choose between near-identical listing pages on link signals alone.
• Treat noindex on deep pages as a last resort rather than a routine crawl-budget measure. Mueller has described what happens over time: Google keeps a noindexed page and follows its links at first, but “if we see the noindex there for longer … we will remove it completely. And then we won’t follow the links anyway”, so “a noindex and follow is essentially kind of the same as a noindex, nofollow”.¹⁰ Products reachable only from page four lose their discovery path.

Footnotes

1. Mobile site and mobile-first indexing best practices — Google Search Central ↩

2. Fix lazy-loaded content — Google Search Central ↩

3. Google’s Mueller on the myth of hidden tab content — Search Engine Journal ↩

4. SEO A/B test results: are tabs a problem for SEO? — SearchPilot (published July 2020) ↩

5. Help Google understand your ecommerce site structure — Google Search Central ↩

6. Include structured data relevant to ecommerce — Google Search Central ↩ ↩²

7. Free listings for products — Google Merchant Center Help ↩

8. Pagination and incremental page loading — Google Search Central ↩ ↩² ↩³ ↩⁴ ↩⁵

9. Google: rel=prev/next is not an indexing signal anymore — Search Engine Roundtable (John Mueller, 21 March 2019) ↩

10. Google: long term noindex will lead to nofollow on links — Search Engine Roundtable (John Mueller, December 2017) ↩