What an ecommerce SEO agency actually does
An ecommerce SEO agency makes an online store easier for search engines to crawl, understand and show, then turns that visibility into product and category traffic that buys. On a store that means work most general SEO never touches: thousands of filter URLs, product variants, out-of-stock items, paginated categories, product feeds, and structured data that has to match the price a shopper sees at checkout.
The job splits into four parts. Technical control decides which URLs Google should crawl and index. Category and product optimization maps search demand to the pages that can actually rank and sell. Product data keeps your structured data, Merchant Center feed and on-page prices consistent so you are eligible for rich results, free listings and AI shopping answers. Authority and content earns links and mentions that a catalog cannot earn alone.
What a credible agency will not do is promise positions. Google's own guidance on hiring an SEO is blunt: “No one can guarantee a #1 ranking on Google”. Expect a clear method, named deliverables and honest reporting instead.
Ecommerce SEO audit checklist, mapped to revenue impact
Many store audits arrive as a long list of “issues” sorted by count. That is backwards. The table below orders the checks by how directly each one touches pages that earn money, so the first fixes protect revenue rather than tidy reports. Priority here is a starting point; your own analytics decide the final order.
| Check | What it looks like | Why it costs revenue | Typical fix | Priority |
|---|---|---|---|---|
| Money categories not indexed or not ranking | Key category URLs missing from Search Console's indexed pages, or canonicalized to another URL | Category pages usually target the highest-intent commercial queries; if they are out of the index, nothing else matters | Fix canonicals, internal links and sitemap inclusion; resolve duplicates | Critical |
| Faceted URLs flooding the crawl | Large numbers of filter and sort URLs in crawl stats or “Discovered, currently not indexed” | Crawl time spent on filter combinations slows discovery of new and updated products | Apply the faceted-navigation decision table below | Critical on large catalogs |
| Price or availability mismatches | Structured data, feed and visible price disagree | Data that disagrees with the page may not be trusted or shown, and shoppers who see a different price at the click tend to leave | Generate markup and feed from the same source of truth | High |
| Product variants splitting signals | Each color or size indexed as a separate thin page | Links and reviews are spread across near-duplicates | Canonical base URL or ProductGroup markup, depending on setup | High |
| Pagination canonicalized to page 1 | Page 2+ declare page 1 as canonical | Products deep in the category can lose their only discovery path | Self-referencing canonicals and crawlable sequential links | High |
| Out-of-stock and discontinued handling | Soft 404s, mass redirects to the homepage, or empty pages left live | Wastes crawling and sends shoppers to dead ends | Keep pages for temporary stock-outs; 404/410 or a relevant redirect for permanent removals | Medium |
| Slow or unstable templates | Poor LCP, INP or CLS on product and category templates in field data | Template problems repeat across every product page at once | Fix at template level: images, scripts, layout shifts | Medium |
| Thin category copy and weak internal links | Category pages with only a product grid; no links between related categories | Harder for search engines to understand what the page is about and how categories relate | Short, useful buying guidance; related-category and breadcrumb links | Medium |
| No Merchant Center feed | Products absent from the Shopping tab and free listings | Misses free organic product surfaces entirely | Set up Merchant Center and a maintained feed | Medium to high |
If you want this run against your own store, our SEO audit service uses the same ordering, and the deeper crawl and indexing work sits with our technical SEO team.
Faceted navigation: a decision table
Filters for size, color, brand and price are good for shoppers and dangerous for crawling. Every combination can create a new URL, and a few dozen filters can produce more URLs than a crawler will ever need. Google's faceted navigation guidance (updated December 2025) gives two paths. If filtered URLs do not need to appear in Search, prevent crawling with robots.txt rules or by moving filter state into URL fragments, which Google generally does not crawl or index. Canonical tags and nofollow on filter links are described as secondary options that are generally less effective over time. If some filtered URLs should be indexed, use the standard & parameter separator, keep filter order consistent with no duplicates, and return a 404 when a combination has no results.
The practical question is which facets earn their own indexable page. We decide per facet type:
| Facet type | Example | Real search demand? | Typical decision |
|---|---|---|---|
| Brand within a category | running shoes + Brand X | Often yes | Indexable landing page with a clean URL, unique copy and internal links |
| Core product attribute | waterproof jackets, leather sofas | Often yes | Indexable if keyword research shows demand and the result set is stable |
| Size, fit or minor color | size 9, navy | Rarely on its own | Keep out of the crawl (robots.txt rule or fragment) |
| Price range | under $50 | Sometimes, for gift-style queries | Usually blocked; create a curated page only where demand is proven |
| Sort order and view options | sort=price_asc, grid/list | No | Block from crawling; never link as indexable URLs |
| Multi-filter stacks | brand + size + color + price | Almost never | Block from crawling |
| Empty combinations | filters with zero products | No | Return 404, as Google recommends |
One caution: robots.txt blocking is a crawl control, not a cleanup tool. If thousands of filter URLs are already indexed, the order of operations matters, and that is worth planning before anyone edits robots.txt on a live store.
Product pages, variants and structured data
Product pages carry the transaction, but they are also where duplicate content multiplies. Google's ecommerce URL structure guidance says fragments are not used in indexing, that the same URL should be used in internal links, sitemaps and canonical tags, and that for variants exposed through paths or parameters the base product URL is a sensible canonical.
Product structured data
Google supports two flavors of Product structured data. Product snippets suit pages where you cannot buy directly, such as editorial reviews. Merchant listing experiences suit pages where shoppers can purchase, and support richer detail such as shipping and return information. Name, offer and price data, plus review or aggregate rating information where you genuinely have it, are the core. The same page notes that Google recommends providing both on-page structured data and a Merchant Center feed.
Variants
For products sold in several sizes or colors, Google documents product variant markup using ProductGroup, with productGroupID (the parent SKU), variesBy for the attributes that change, and hasVariant for each variant. It covers both single-page setups, where variants switch on one URL, and multi-page setups, where each variant has its own URL and the group definition is repeated on each page. Which one fits depends on your platform and how your theme renders variants, so we check rendered HTML rather than assuming.
On-page basics that still decide clicks
- Titles that name the product type, brand and the attribute people search for, not just the internal product name.
- Original descriptions for top sellers instead of manufacturer copy shared by every reseller.
- Image alt text that describes the product, and image file names that are not camera defaults.
- Visible shipping, returns and availability information that matches the markup.
Category pages, pagination and internal linking
For most stores, category (or collection) pages are the money pages. They match broad commercial queries like “men's trail running shoes” far better than any single product. Category work usually means mapping one primary query to each category, writing a short block of genuinely useful buying guidance, and linking related categories so both shoppers and crawlers can move sideways through the catalog.
Pagination is where good category work often breaks. Google's pagination guidance is specific: give each page its own URL (for example ?page=2), do not use the first page as the canonical for the whole series, and link pages sequentially with normal <a href> links. Google also states it no longer uses rel="next" and rel="prev", and that its crawlers do not click “load more” buttons, so infinite scroll needs a crawlable fallback or a sitemap and feed that list every product.
Supporting content earns its place when it answers a pre-purchase question the category page cannot, such as sizing guides or comparisons, and links back to the category. Our SEO content team plans those pieces against the category map so they support sales pages rather than compete with them.
Merchant Center, free listings and AI shopping surfaces
Ecommerce SEO is no longer only the ten blue links. Product data now feeds several free, organic surfaces, and an ecommerce SEO agency should manage them as one system.
| Surface | What feeds it | What to get right |
|---|---|---|
| Google Search results | Crawled pages plus Product structured data | Indexable product and category pages; markup that matches the page |
| Free listings (Shopping tab, Images, YouTube, Maps, Gemini, Lens) | Google Merchant Center product data | Complete feed: IDs, titles, prices, availability, identifiers, shipping settings |
| Google AI Overviews and AI Mode | Indexed pages; AI Mode shopping draws on Google's Shopping Graph | Snippet-eligible pages and accurate, current product data |
| ChatGPT product results | Web content and, for approved merchants, OpenAI's product feed | Crawlable pages, clear product facts, and a feed if you are accepted |
Google's Merchant Center help explains that with free listings, products can appear at no cost across Search, Maps, Gemini, YouTube, the Shopping tab, Google Images and Google Lens, provided you have a Merchant Center account and product data. Google's ecommerce documentation on sharing product data adds that a feed increases confidence that Google knows about all your products and is required for the Shopping tab.
On AI surfaces, Google's AI features documentation says there are no additional requirements to appear in AI Overviews or AI Mode beyond being indexed and eligible for a snippet. For shopping specifically, Google said in November 2025 that AI Mode shopping draws on its Shopping Graph of more than 50 billion product listings, about 2 billion of which are refreshed every hour. Outside Google, OpenAI states that ChatGPT product results are chosen independently. Since 2026 OpenAI has also been testing ads in ChatGPT, which it says are labelled as sponsored, kept separate from the answer and do not influence it. OpenAI also documents a product feed that approved merchants can use to share identifiers, pricing, inventory and media. Our GEO service and AI SEO service cover these surfaces in more depth.
Core Web Vitals and catalog performance
Google's Core Web Vitals guidance sets good targets of Largest Contentful Paint within 2.5 seconds, Interaction to Next Paint under 200 milliseconds, and Cumulative Layout Shift under 0.1. Google presents these as part of good page experience, not as a shortcut to rankings. On stores, the usual culprits are oversized hero and product images, review and chat widgets, tag-manager bloat, and layout shifts from late-loading badges or price blocks. Because product and category pages share templates, one fix can improve thousands of URLs at once, which is why we measure by template, not page by page.
Crawl efficiency is the other half of catalog performance. Google's crawl budget guide is written for large sites (around a million pages that change weekly) and medium sites (10,000+ pages that change daily). If your store is in that range, duplicate URLs, soft 404s and faceted traps deserve more attention than any single page tweak.
Platform notes: Shopify, Adobe Commerce (Magento), WooCommerce and BigCommerce
The principles above apply on every platform; the controls differ. A few verified starting points:
| Platform | What to check first |
|---|---|
| Shopify | Shopify generates robots.txt by default; a robots.txt.liquid template lets you add or remove rules, which Shopify lists as an unsupported customization. Check that product links and canonicals consistently use one product URL and that theme apps do not inject duplicate or conflicting schema. |
| Adobe Commerce (Magento) | The admin has canonical meta tag settings for category and product pages under Catalog > Search Engine Optimization. Layered navigation is the faceted-navigation risk; review which attribute URLs are crawlable. |
| WooCommerce | Behavior depends heavily on the theme and SEO plugin. Check which plugin outputs Product markup (avoid two), how attribute filters create URLs, and whether tag and attribute archives are worth indexing. |
| BigCommerce | Check how faceted search URLs are exposed, how product URLs are formed, and that the theme's structured data matches visible price and availability. |
Running several country stores or languages on one platform adds hreflang and duplicate-catalog questions; our enterprise and international SEO page covers that work.
What a typical 90-day ecommerce SEO engagement looks like
This outline is typical, not guaranteed. Timing shifts with catalog size, platform access, developer availability and how much is already broken. Results depend on competition, budget and implementation, and we do not promise rankings or revenue.
| Weeks | Focus | Typical outputs |
|---|---|---|
| 1–2 | Access and baseline | Search Console, analytics and Merchant Center access; crawl of the store; revenue-weighted list of top categories and products |
| 2–4 | Technical audit and fixes plan | The revenue-mapped checklist above, a faceted-navigation decision per facet, canonical and pagination fixes, ticketed for your developers or platform |
| 4–8 | Category and product optimization | Keyword-to-category map, rewritten titles and category copy for priority pages, structured data and feed alignment |
| 6–10 | Product data and AI surfaces | Merchant Center feed review, free listings check, crawlability for AI search, internal linking between categories |
| 8–13 | Authority and reporting | Supporting content plan, link and mention targets, and a report tying indexing, visibility and organic revenue to what shipped |
Our process page explains how we run engagements in general. For a scoped quote, contact us with your store URL and platform.
How to choose an ecommerce SEO agency
Lists of the “best ecommerce SEO agencies” rank firms on reviews and size. Those signals matter, but they do not tell you whether an agency can fix your catalog. Ask every shortlisted agency these questions:
- How would you handle our filters? A good answer separates facets with search demand from facets that should be kept out of the crawl, and mentions Google's own guidance.
- Which of our pages earn the most, and what would you do to them first? Priorities should follow revenue, not issue counts.
- Who implements the fixes? Many recommendations need theme or platform changes. Know whether the agency writes tickets, edits templates, or both.
- How do you keep structured data, the feed and the page in sync?
- What will you report, and how often? Indexed pages, impressions, clicks and organic revenue by category are more useful than a single traffic line.
- Can we see real work? Ask for sample deliverables or verifiable references, and treat unverifiable growth percentages with care.
Red flags: guaranteed positions (Google says nobody can guarantee them), secret link networks, auto-generated pages for every filter combination, and contracts that hide what work is actually done each month.
How Mediardx works on ecommerce accounts
Mediardx is an SEO agency based in India that works remotely with stores in the US, UK, Canada and Australia. On ecommerce accounts we lead with the technical and product-data work above, because it decides whether the rest of SEO can pay off. Every engagement starts with an audit you keep, priorities ordered by revenue, and fixes written so your developers or platform partner can ship them. We report on what changed and what it did, including when something did not move.
We will not quote invented case-study numbers or promise rankings. If you want to see how we would approach your store, request an audit and we will explain its scope and what you will receive before you commit to anything larger.