Ecommerce SEO audit checklist: indexing, faceted filters, categories, product pages, variants, Product markup, speed, stock, Merchant Center and AI shopping.

In short
An ecommerce SEO audit checks whether Google can crawl, understand and rank the pages that sell: categories and products. It covers indexing and faceted filters, site structure, category and product content, duplicate content from variants, Product structured data, images, speed, out-of-stock handling, the Merchant Center feed and visibility in AI shopping answers. The 13 steps below, each with a test and a fix, take a few days for a store of a few thousand products.
Want to go further? Ask
An ecommerce SEO audit differs from a generic site audit because stores share the same structural traps: thousands of near-identical URLs created by filters, variants and sorting, manufacturer descriptions copied across hundreds of sites, and products that come and go. Google's documentation on ecommerce site structure sums up the starting point: Google must be able to reach every product by following links from menus to categories and from categories to products. This checklist starts there.
| Area | Test | Typical fix |
|---|---|---|
| Indexing | Page indexing report, indexed vs submitted URLs | Remove noindex on key pages, clean sitemaps |
| Faceted filters | Count crawled filter URLs in a crawl or logs | Block useless combinations, index only searched ones |
| Structure | Click depth of categories and products | Links from menus, hubs and best sellers |
| Category pages | Unique H1, intro text, internal links | Short useful intro, buying guide links |
| Product pages | Share of copied manufacturer descriptions | Rewrite top products, add specs and FAQs |
| Duplicates and variants | Same product on several URLs | Canonical tags, variant markup |
| Structured data | Product rich results report | Complete Product and Offer markup |
| Images | Missing alt text, weight | Descriptive alt text, compression, modern formats |
| Speed | Core Web Vitals report | Lighter templates, fewer apps and scripts |
| Stock | Status codes of sold-out and discontinued products | Keep, redirect or remove by case |
| Merchant Center | Feed errors and disapprovals | Align feed, page and markup |
| Content and links | Blog links to categories | Guides that link to the pages that sell |
| AI visibility | Product questions asked to AI assistants | Comparison content, consistent product data |
Export from Search Console the clicks, impressions and positions of category and product pages over 16 months, and from analytics the revenue by organic landing page. You need this to prioritize (which pages bring sales) and to measure the effect of fixes. The ecommerce SEO ROI calculator turns position gains into estimated revenue.
In the Page indexing report, compare the number of indexed pages with the number of products and categories you sell. Far more indexed URLs than products signals duplicates or filters; far fewer signals crawling or quality problems. Check that key templates do not carry a stray noindex and that XML sitemaps list only canonical URLs returning a 200 status.
Filters for size, color, price and sorting are the first source of wasted crawling in stores. Google's guide to faceted navigation warns that parameter-based filters "can generate infinite URL spaces" and recommends preventing crawling of filter URLs that do not need to be indexed. Keep indexable only the combinations people search for ("women's black running shoes"), with a clean URL, a unique title and links from the parent category.
Crawl the store and export click depth. Categories should sit one or two clicks from the home page, products within three or four. Google states that Googlebot "generally doesn't try to submit searches into a search box", so a product reachable only through internal search may never be found. Our guide to site architecture for SEO gives a model tree for stores.
Category pages target the most valuable queries ("running shoes", "oak dining table"). Each needs a unique H1 and title, a short introduction that helps choose, filters that do not hide products from crawlers, and links to subcategories and buying guides. Avoid long blocks of generic text at the bottom of the page: a few useful sentences beat 800 words written for robots.
Measure how many products use the manufacturer's description word for word: the same text exists on every retailer selling the product. Rewrite at least the best sellers with what the buyer needs (dimensions, materials, use cases, compatibility, care), add reviews and answer frequent questions on the page. An on-page audit quickly flags missing titles, duplicate H1s and thin pages.
The same product often lives at several URLs: one per color, one per category path, one with tracking parameters. Each should point to one version with a canonical tag. Decide per catalog whether variants deserve their own page (a color people search for) or a single page with a selector. Our entry on duplicate content explains how Google picks the canonical version.
Google's Product structured data documentation distinguishes product snippets (for pages that review or describe products) from merchant listings (for pages where the product can be bought, with price, availability, shipping and returns). It also supports product variants and recommends adding return policy markup under Organization. Check the Product and Merchant listings reports in Search Console after each template change. Prices and availability in the markup must match what the shopper sees on the page.
Google's image SEO best practices call alt text "the most important attribute" for image metadata, alongside descriptive file names and images placed near relevant text. Compress product photos, serve modern formats and keep the main image crawlable. Our guide to image SEO lists the checks.
Test a category page, a product page and the cart, not only the home page. Stores slow down with apps, tracking scripts, sliders and unoptimized images. Use the Core Web Vitals report in Search Console for field data, and remove apps that are no longer used before optimizing code.
A Google Merchant Center feed makes products eligible for free listings and the Shopping tab. Google's guide on sharing product data recommends combining structured data on pages with a Merchant Center feed. Review disapproved products and make sure price and availability match the page exactly.
Ask ChatGPT, Gemini and Perplexity the questions your buyers ask ("best trail running shoes for wide feet", "oak or walnut dining table") and note which stores and brands are cited. AI assistants rely on clear product data, reviews and comparison content found across the web. AI mention tracking follows these questions over time, and our Amazon AI citation study shows how often the largest marketplace is cited. Read more on AI in ecommerce.
On Shopify, products are reachable both at /products/name and through collection paths; the platform adds canonical tags to the /products/ version, so the main checks are themes, apps, filters and duplicate collections. See our pages on Shopify SEO and GEO and the best Shopify SEO apps. On WooCommerce, check tag and attribute archives, which often create thin indexable pages, and the settings of your SEO plugin.
| Priority | Examples | When to fix |
|---|---|---|
| Critical | Noindex or robots block on categories or products, broken canonicals, products unreachable by links | Immediately |
| High | Uncontrolled filters, copied descriptions on best sellers, missing Product markup | Within the month |
| Medium | Slow templates, missing alt text, weak category intros | Within the quarter |
| Ongoing | Buying guides, internal links from the blog, AI visibility | Every month |
A suite such as those in our ranking of the best ecommerce SEO tools speeds up the crawl and the monitoring.
An audit is only useful if someone can act on it. Keep the deliverable short and operational:
An ecommerce SEO audit pays off when it fixes what blocks categories and products first, then builds content that sends buyers to them. Buying guides and comparisons that link to your categories are the lasting part of the work. Sorank, built for ecommerce sites, publishes these articles on autopilot with internal links to the pages that sell.
A full audit once or twice a year is enough for most stores, plus a check after every redesign, platform migration or large catalog change. Between audits, monitor the Page indexing, Core Web Vitals and Merchant listings reports in Search Console each month to catch problems early.
Not when the product is coming back. Keep the page live, show availability and alternatives, and update the structured data and the feed. Remove or redirect only discontinued products: a 301 to a close replacement or the category when one exists, a 404 or 410 when nothing equivalent is sold.
You can, but they rarely help you rank, because the same text appears on every store selling the product. Start by rewriting the descriptions of your best sellers and most searched products with original details, then extend the work to the rest of the catalog over time.
THEY TRIED SORANK BEFORE YOU
