Ecommerce SEO: The Page-Type Playbook for Category, Product and Filter Pages

What ecommerce SEO is (and why it is different from regular SEO)

Ecommerce SEO is SEO for a site where the inventory is the content. A service business might have 40 pages that matter. A store can have 400 products, 60 categories and, thanks to filters and sorting, tens of thousands of URLs that nobody wrote on purpose.

That changes the job. On a brochure site, the risk is too little content. On a store, the risk is too much of the wrong kind: near-identical pages that compete with each other, filters that trap crawlers in endless combinations, and product pages built from the same manufacturer copy as 200 other retailers.

So good ecommerce SEO is half addition and half subtraction. You add depth to the few page types that can win searches, and you remove or quiet the thousands of URLs that only dilute. The rest of this guide is organized by page type, because that is how fixes ship: one template change touches every page of that type.

If you run a local service business rather than a store, start with our guide to SEO for small business instead. The playbook below is for sites that sell products online.

The page-type priority map we start every store with

Rank your page types by the searches they can win and the revenue behind them, then spend effort in that order. On most stores we audit, the order looks like this.

The last row matters as much as the first. Every URL Google spends time on is one it is not spending on your best category. Google's own faceted navigation guidance says crawlers "will typically access a very large number of faceted navigation URLs" before they can tell those URLs are useless.

Score each page type out of 10 on two questions: how much search demand does it address, and how many pages share the template? A fix to a template used by 300 category pages outranks a perfect rewrite of one product page, every time.

Category pages: where ecommerce SEO is won

Category pages win because they match how people shop. Someone searching "mid-century walnut nightstand" wants to compare options, and a category page is the only page on your site built to show options. Most stores leave these pages as a bare product grid, which is why they lose to marketplaces and to competitors who did the work.

The anatomy of a category page that ranks

1. A title tag and H1 that name the category the way buyers say it. "Walnut Nightstands" beats "Bedroom Collection 3."
2. A short intro above the grid, 40 to 80 words, that answers the first question a buyer has: what makes these different, and how to choose.
3. The product grid, crawlable as plain links. If products load only after a script runs or a button is clicked, Google may not find them.
4. A buying-help block below the grid: sizing, materials, care, a comparison table or 3 to 5 real questions with direct answers.
5. Links to child categories and to the indexable filter pages that deserve their own rankings.
6. BreadcrumbList structured data that matches the visible breadcrumb trail.

Write for the buyer, not the keyword

The copy on a category page should help someone choose. Explain the difference between solid wood and veneer, what size fits a queen bed, which finish hides fingerprints. That copy earns rankings because it is useful, and it gives AI answers something specific to quote. Keyword-stuffed paragraphs at the bottom of the page do neither.

One test we use: if you deleted the product grid, would the remaining text still help a shopper? If not, the page is not finished.

Faceted navigation: the decision table for filters

Every filter on your store needs one of three fates: indexed landing page, crawlable but not indexed, or blocked from crawling. The mistake is not picking. Default platform settings usually either index everything or block everything, and both are wrong for most stores.

Google's faceted navigation documentation is direct: "Oftentimes there's no good reason to allow crawling of filtered items." It recommends blocking filter URLs in robots.txt when they do not need to appear in search, and it says to return a 404 when a filter combination has no results. Here is how we turn that into decisions.

How do you know which single attributes have demand? Check keyword data for "[attribute] + [category]" phrases, and check Search Console for queries that already land on the parent category with that attribute in them. If people search it, it gets a page. If they don't, it stays a filter.

One caution: rel=canonical and nofollow are weaker tools here. Google calls them secondary signals that are generally less effective over time than robots.txt or fragments. Use them as a backup, not the plan.

Product pages that earn the click

A product page ranks for specific searches: the model name, the brand plus product type, the part number. It wins those when it carries more and better information than every other store selling the same item.

The product page checklist

• Original description. Rewrite manufacturer copy, at least for your top sellers. Add what you know that the manufacturer does not say: fit notes, who it suits, what it pairs with.
• Complete specifications in a table: dimensions, materials, compatibility, what is in the box.
• Real reviews collected from buyers, shown as text on the page.
• Clear price, availability, shipping and returns above the fold. These are also the facts your structured data should carry.
• Descriptive image file names and alt text for every photo.
• A path back up: breadcrumb to the category, links to related products and the relevant buying guide.

Variants: one page or many?

Colors and sizes of the same product usually belong on one URL, with a selector. Give a variant its own indexable page only when people search for it separately and it differs in a way that matters, such as a different material or model. Otherwise you create a set of near-duplicates that split reviews, links and rankings. Google documents a ProductGroup pattern for marking up variants so it understands they belong together.

Product data: structured data and Merchant Center together

Send Google your product facts two ways: Product structured data on each page and a product feed in Google Merchant Center. Google's Product structured data documentation says you can use either, and that providing both "maximizes your eligibility" for shopping experiences in Search.

That documentation splits product markup into two experiences. Merchant listings are for pages where customers can buy from you, and support shipping details, return policies and apparel sizing. Product snippets are for pages where people cannot buy directly, such as review sites. A store wants merchant listing markup on every purchasable product page.

What to mark up on a store

Google's ecommerce structured data guide lists Product, ProductGroup, BreadcrumbList, Organization and Review as the core types. The rule that trips stores up is consistency: the price in your markup, your feed and your visible page must match. When a sale price changes on the page but the feed lags a day, you get disapprovals and lost visibility.

Product data is also how AI shopping answers decide what to show. Clean, complete, consistent facts are the closest thing to optimizing for them. Our guide to AI SEO covers the content side of being cited.

Out of stock, discontinued, seasonal: what to do with each URL

Inventory changes are the most common way stores quietly lose rankings. A product sells out, someone unpublishes the page, and a URL with years of links and reviews turns into a 404. Use this table instead.

The availability values come from schema.org's ItemAvailability list. The principle: a URL that has earned links and rankings is an asset. Retire it on purpose, with a redirect to the most relevant page, never by accident.

Pagination, internal links and site structure

Keep every product within a few clicks of the home page, and make every path a plain HTML link. Google's crawler follows links; it does not click buttons or scroll.

Pagination rules from Google

Google's pagination guidance for ecommerce is specific. Give each page in a series its own URL, such as ?page=2. Do not use page 1 as the canonical for the rest of the series. Link pages in sequence with normal anchor links. Google no longer uses rel="next" and rel="prev", and it ignores URL fragments, so #page=2 will not work. If you use infinite scroll or a "load more" button, make sure every product is still reachable through crawlable links.

Internal links that move rankings

• Link from the home page to your top categories by name, not just through a mega menu.
• Link from buying guides to the categories and filter landing pages they discuss.
• Link from product pages back to the category and sideways to close alternatives.
• Use descriptive anchor text: "waterproof trail running shoes," not "shop now."

Google's guide to URL structure for ecommerce sites is worth a read before any replatform, because URL decisions made at launch are expensive to undo later.

Content that supports ecommerce SEO without becoming a blog for its own sake

Write content that helps someone choose a product you sell, then link it to the category where they can buy. That is the whole content strategy for most stores. A blog full of loosely related lifestyle posts rarely earns sales.

The formats that work hardest for ecommerce SEO:

• Buying guides per major category: how to choose, what matters, what to skip.
• Comparisons between two options shoppers genuinely weigh against each other.
• Sizing and fit guides, which cut returns as well as earning searches.
• Care and how-to pages for products that need maintenance.

Each piece should answer the question in its first lines, with a table or steps an AI engine can quote, then point to the products. Our content marketing strategy guide explains how to plan this so every article has a job.

Speed, mobile and conversion: where SEO meets revenue

Ranking a category page is only worth something if the page converts. Stores carry heavy scripts: review widgets, chat, personalization, several tracking tags. Each one slows the page and can push your Core Web Vitals out of the good range.

Audit third-party scripts once a quarter and remove anything nobody uses. Size product images for the space they fill. Read our Core Web Vitals guide for the targets and the common fixes, and our guide to improving website conversion rate for what to change once shoppers land.

Measure organic revenue, not just traffic. Tag campaigns consistently with UTM parameters so email and social visits are not mixed into organic, and judge ecommerce SEO on revenue from organic landing pages by page type.

Ecommerce SEO mistakes we see on almost every audit

1. Indexing every filter combination. Thousands of thin pages, crawl budget gone, rankings split.
2. Blocking every filter, including the handful that match real searches and should be landing pages.
3. Canonicalizing paginated pages to page 1, which hides products on later pages.
4. Manufacturer descriptions on every product, identical to every other retailer.
5. Deleting sold-out products and breaking links that took years to earn.
6. Feed, markup and page prices out of sync.
7. Replatforming without a redirect map, the single most expensive ecommerce SEO mistake there is.
8. Empty category pages with a heading, a grid and nothing to help someone choose.

Most of these are template or settings problems, which is good news. Fix the template once and the fix rolls out to every page. Our technical SEO checklist covers the crawl and index checks in the order to run them.

Your first 90 days of ecommerce SEO

Here is the plan we would run on a store with a few hundred products and flat organic revenue.

Expect early movement on category pages that were already close to page one. New landing pages and guides take longer. Report on organic revenue by page type every month so you can see which work is paying.

How Unified approaches ecommerce SEO

We have run marketing for more than 1,100 businesses since 2006, and the lesson that carries over to stores is simple: fix templates, not pages. A site at scale only improves when the fix lives in the code that every page shares. We learned that building Nusbaum Medical Centers, which now has 3,066 pages live and 137 page-one rankings.

Our SEO team and development team sit together, so the faceted navigation rules and the template changes ship as one piece of work rather than a ticket that waits in a queue. UnifiedIQ, our own platform, scores every page for SEO, so a template problem shows up as a drop across hundreds of URLs at once.

If you are comparing partners, our list of the best marketing agencies for e-commerce brands lays out who does what. If you want a page-type audit of your store, talk to us or start with a technical SEO audit.

Ecommerce SEO questions, answered