E-commerce SEO Audit: A Store Checklist, from Indexation to Schema, in the Order Problems Cost Sales

Three counts to write down before you open a crawler

Start with arithmetic, not software. From the store's admin, note how many products and categories are live. From the sitemap, note how many URLs you are submitting. From Search Console's Page indexing report, note how many pages Google has indexed. The gaps between those three numbers tell you which half of the audit deserves your afternoon.

• Indexed far below catalog size: Google is not finding or not keeping your products. Spend most of your time on the first check.
• Indexed far above catalog size: Google is indexing filters, sorts, parameters or duplicates. Spend it on the second and third checks.
• Sitemap count far from catalog count: the sitemap itself is stale or bloated, which a sitemap guide can sort out first.

This checklist uses Search Console because it shows Google's own view, free. A crawler from our roundup of SEO audit tools helps on a big catalog, and the free website audit gives a fast outside read on speed, SEO, mobile and security if you want a baseline before you start. The e-commerce SEO hub explains the concepts behind each check; this page is the inspection.

Check 1: Is Google indexing the products you sell?

Open the Page indexing report and use the filter above the chart to show only submitted pages, or one sitemap at a time if products, categories and pages each have their own. Submitted URLs are the ones you have asked Google to index, so problems there are the problems that matter. Then read the reasons Google gives.

1. 1

   Read "URL marked 'noindex'" and "URL blocked by robots.txt" first

   Any product or category in either list is hidden by your own settings. On stores these usually come from a theme option, an SEO app or a rule copied from a staging site.

2. 2

   Read "Discovered, currently not indexed"

   Google knows these URLs but has not crawled them, and its help says that is often because crawling then was expected to overload the site. On a store, a long list here usually means the crawler is busy elsewhere, often in filter URLs.

3. 3

   Read "Crawled, currently not indexed"

   Google fetched the page and chose not to index it yet. Open five examples. Thin products with a manufacturer's description and one photo are the usual pattern; our guide to product pages covers the rewrite.

4. 4

   Read "Soft 404"

   Google defines this as a page that says it was not found while returning a success code. A product page that says the item is gone, served with a 200, is the classic store example.

5. 5

   Test reachability by hand

   Pick three products from deep in the catalog and click to each from the home page. Google's e-commerce guidance strongly recommends linking to every product you want indexed, and warns that products reachable only through the search box may be missed, because Googlebot generally does not submit searches.

Navigation needs a quick look too. Google asks for standard links with an href, and says not to rely on JavaScript events attached to other page elements for navigation. If a category menu only works after scripts run, that is a JavaScript SEO problem to note. A store missing from Google altogether is a different job; start with the site not indexed fix, or Shopify store not showing in Google for Shopify.

Check 2: Find the duplicates that split one product into five

Duplicates are not a penalty. Google's SEO Starter Guide says content available under several URLs is fine, inefficient, and not something that causes a manual action. The cost is dilution: links, clicks and crawling spread over copies, and Google may pick a canonical you did not want. Two Page indexing reasons show it happening: "Duplicate without user-selected canonical" and "Duplicate, Google chose different canonical than user." For the second, URL Inspection shows which address Google picked.

Use the strongest signal that fits. Google ranks its canonicalization methods with redirects first, then rel="canonical" annotations, then sitemap inclusion as a weak signal. It also says not to use robots.txt or the URL removal tool for canonicalization: a blocked URL can still be indexed without its content, and the removal tool hides every version. Our guide to canonical tags covers the syntax and the common mistakes, and redirects covers the 301s.

Check 3: Click every filter, then look at the address bar

The hub explains why faceted navigation multiplies URLs. The audit question is narrower: what does your store do right now? Open your largest category, apply two filters and a sort, and copy the URL. Then run the same test on a filter combination that should return nothing.

• New parameters, crawlable, with no robots.txt rule: Google's faceted navigation guidance suggests disallowing filter parameters you do not need indexed, while keeping the unfiltered listing crawlable.
• Filters after a #: Google generally does not use fragments for crawling or indexing, so these create no new crawlable addresses. Fine, unless you wanted a filter to rank.
• An empty combination returns a 200 page saying no products: Google asks for a 404 for filter combinations with no results, and for duplicate or nonsensical ones, without redirecting them to a generic error page.
• Filters you want indexed: check that they use the standard & separator, appear in a consistent order and never repeat.
• Numbered pages: each should have its own URL and its own canonical, not page one's. Google no longer uses rel="next" and rel="prev", and recommends linking each page to the next with plain links.
• Infinite scroll or a load more button: Google's pagination guidance notes that its crawlers do not click buttons, so check that the products further down are also reachable through ordinary links.

Then compare with Search Console. If the indexed count from the first section was far above the catalog, the URL Inspection tool on a few indexed filter pages will usually confirm the leak. Deciding which filters deserve their own indexable page is a keyword question, answered in e-commerce keyword research.

Check 4: How the store treats products it cannot sell today

Out-of-stock handling is where well-meaning store owners delete the most search traffic. Google's guidance on pausing an online business is direct: it is better to keep the page and mark it out of stock, because if you remove the product from Search, people do not know why it is gone. It warns that 404, 410 and noindex all remove URLs from Google Search, and says not to use Search Console's temporary removal tool for out-of-stock products.

The successor redirect is our practice rather than a Google rule, and it only works when the new product genuinely replaces the old one; sending every dead product to the home page helps nobody. One more detail from Google's launch guidance: do not disable the add-to-cart button to stop orders, because Google may use add-to-cart to verify final prices. If old product URLs have piled up as broken links, fixing 404 errors is a contained project.

Check 5: Speed, measured on the templates shoppers actually use

A store has four templates that matter: home, category, product and cart. Search Console's Core Web Vitals report groups URLs with a similar user experience, so one slow product template shows up as one large group instead of a thousand separate complaints. The data comes from real Chrome users through the Chrome UX Report, and each group takes the status of its worst metric.

Low-traffic stores may see no groups at all, because the report leaves out URLs without enough data; test the four templates in PageSpeed Insights instead. On stores the usual culprits are oversized product photos, app scripts that load on every page and review widgets that shift the layout. Our guide to improving Core Web Vitals covers the fixes, e-commerce website design covers the layout choices behind them, and site speed work is there when the theme itself is the problem.

Check 6: Product markup and feeds that tell the same story as the page

Search Console has two reports for Product structured data: Merchant listings, for pages where shoppers can buy, and Product snippets, for other product pages such as reviews. Open whichever applies and read the errors before the warnings. Then run three best sellers through the Rich Results Test.

Leave FAQ markup off the checklist as a rich result play: Google stopped showing FAQ rich results on May 7, 2026. Our schema markup guide covers the types that still earn results. Clean product data also feeds AI shopping answers, which ChatGPT shopping covers.

Turning six checks into a fix list someone will actually work

An audit that ends as a 90-page PDF rarely changes anything. Rank what you found by how many sales it blocks per day. Anything that hides products you sell, such as a stray noindex, a robots.txt rule or out-of-stock pages returning 404, goes first. Duplicates and filter leaks come next, because they slow everything else down. Speed and markup follow, then content.

Most of this is template work: one fix to the product template repairs every product at once, which is why a store audit is usually shorter to fix than it looks. When the fixes need code, a platform change or a migration, DataCram's e-commerce development builds stores with these foundations in place, from $3,500 as a fixed price. Ongoing search work runs as a monthly program from $625 a month, with most local businesses between $625 and $2,250, per our SEO pricing. Our technical SEO guide covers the checks every site needs, store or not.