E-Commerce SEO Foundations

E-commerce SEO is not content SEO with products bolted on. It is a database-shape problem: tens of thousands of near-identical pages, faceted URLs that explode combinatorially, inventory that churns, and revenue that is measured per session, not per impression. The platform you ship on decides 60% of what’s possible before you write a single tag.

TL;DR

• E-commerce SEO is index-management first, content second. A 50,000-SKU catalog has more URLs than most publishers, and 90% of them fight for the same intent. Your job is to tell Google which 10% to rank.
• The platform sets the ceiling. Shopify ships with hardcoded /products/ and /collections/ paths and a duplicate-content tax via ?variant=. Magento gives you total control and demands you build everything. Pick deliberately.
• Revenue per indexed URL is the only metric that matters. GSC impressions on a category page that converts at 3.2% beats 50× the impressions on a tag page that converts at 0.04%.

The mental model

E-commerce SEO is like running a warehouse with infinite aisles where the floor staff (Googlebot) only has 30 minutes a day to walk the place. Every aisle they walk is one less they can spend on a different aisle. If you have 200 aisles full of the same shoe in seven colors, they will leave before they ever see the new arrivals.

The traditional SEO instinct — “more pages = more traffic” — inverts in commerce. Each additional URL competes with your existing URLs for crawl budget, link equity, and search query mapping. The goal is to publish the smallest URL set that captures the largest share of commercial intent.

Three layers stack: the catalog layer (PDPs and PLPs), the discovery layer (facets, search, breadcrumbs), and the trust layer (reviews, schema, returns content). Most teams optimize layer 1 in isolation. The compounding wins are in layer 2’s restraint and layer 3’s signals.

Deep dive: the 2026 reality

Google’s Helpful Content system, baked into the core algorithm since March 2024, treats commerce sites as a special category. The September 2023 update demoted thousands of affiliate review sites; the March 2024 core update extended the same scrutiny to retailers running thin manufacturer copy across 5,000-product catalogs. The signal Google appears to weigh: search-result quality per indexed URL, not raw page count.

AI Overviews appears for around 26% of commercial queries as of Q1 2026, and AI Mode (full-conversational SERP) ships defaults for 11% of US shoppers. AI Overviews preferentially cite pages with explicit Product schema, validated aggregateRating, and machine-readable price/availability — which means structured data is no longer cosmetic. Pages without schema simply do not appear in the AI panel.

Crawler reality: Googlebot (smartphone), Googlebot-Image, GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot, and Google-Extended all hit commerce sites differently. ChatGPT Search and Copilot pull from Bing’s index, so Bing Webmaster Tools coverage matters again for AI traffic. Perplexity blends its own crawl with partner feeds; getting your feed into Merchant Center indirectly helps you get cited there.

Platform impact is concrete:

Platform	Default URL pattern	Schema	Pain points	Best for
Shopify	/products/x, /collections/y	Theme-dependent JSON-LD	?variant= duplicate tax; collection product limit; no native server-side rendering for app blocks	DTC under 5K SKUs
WooCommerce	Configurable	Plugin-driven (Yoast, RankMath)	Plugin sprawl, slow admin, weak default permalinks	Existing WordPress sites
BigCommerce	Configurable, clean defaults	Native Product schema	Smaller theme ecosystem, fewer SEO apps	Mid-market 5K-50K SKUs
Magento (Adobe Commerce)	Fully customizable	Manual or via extension	Engineering-heavy, slow without Hyvä/PWA	Enterprise multi-store
Custom (Next.js, Astro, Remix)	Whatever you build	Whatever you build	You own every problem	Brands with eng teams

The choice is not “which platform is best for SEO.” It is “which platform’s defaults won’t fight me, and where am I willing to pay engineering tax to overcome them.”

Visualizing it

flowchart TD
  A[Crawl budget] --> B{URL is canonical?}
  B -->|No| C[De-duplicate via canonical or block]
  B -->|Yes| D{Has commercial demand?}
  D -->|No| E[Noindex, follow]
  D -->|Yes| F{Has unique value?}
  F -->|No| G[Merge or strengthen]
  F -->|Yes| H[Index with full schema]
  H --> I[Track revenue per session]
  I --> J{ROI positive?}
  J -->|No| K[Demote or remove]
  J -->|Yes| L[Compound link equity in]

Bad vs. expert

The bad approach

The default Shopify install with no SEO discipline looks like this in robots.txt:

User-agent: *
Disallow: /admin
Sitemap: https://example.myshopify.com/sitemap.xml

Then every product is reachable via three URLs:

/products/red-leather-jacket
/collections/jackets/products/red-leather-jacket
/collections/all/products/red-leather-jacket?variant=42119283

All three index. None canonicalize correctly out of the box on legacy themes. The ?variant= URL is what users share, but Google treats it as duplicate. Crawl budget burns on permutations no one searches for, the strongest URL never accumulates link equity, and the brand wonders why their flagship product ranks page 3.

The expert approach

Self-canonicalize the bare PDP, redirect collection-scoped product URLs, and strip variant params from indexable URLs:

{%- comment -%} theme.liquid head {%- endcomment -%}
<link rel="canonical" href="{{ canonical_url | split: '?' | first }}">
{%- if template.name == 'product' and collection -%}
  <link rel="canonical" href="{{ shop.url }}{{ product.url }}">
{%- endif -%}

Pair with a Product schema that survives review:

{
  "@context": "https://schema.org",
  "@type": "Product",
  "name": "Red Leather Jacket",
  "sku": "RLJ-MED-RED",
  "gtin13": "0123456789012",
  "brand": { "@type": "Brand", "name": "Atelier Nord" },
  "image": [
    "https://cdn.example.com/rlj-front.jpg",
    "https://cdn.example.com/rlj-back.jpg"
  ],
  "description": "Lambskin moto jacket, asymmetric zip, satin lining.",
  "offers": {
    "@type": "Offer",
    "url": "https://example.com/products/red-leather-jacket",
    "priceCurrency": "USD",
    "price": "489.00",
    "priceValidUntil": "2026-12-31",
    "availability": "https://schema.org/InStock",
    "itemCondition": "https://schema.org/NewCondition"
  },
  "aggregateRating": {
    "@type": "AggregateRating",
    "ratingValue": "4.6",
    "reviewCount": "187"
  }
}

The canonical resolves the three-URL problem. The schema gets you into AI Overviews and rich results. The two together compound: link equity concentrates on one URL, that URL accrues clicks, clicks feed user behavior signals, and the page outranks duplicates within 6-10 weeks on most catalogs.

Do this today

1. Open Google Search Console → Indexing → Pages. Sort “Crawled - currently not indexed” descending. If more than 25% of your catalog sits there, you have a duplicate or thin-content problem before you have a ranking problem.
2. Run Screaming Frog with JavaScript rendering enabled, limited to your domain. Export the internal HTML report and pivot on Canonical Link Element 1. Any URL whose canonical is itself but is reachable from a non-canonical URL is a crawl-budget leak.
3. Audit your platform’s default duplicate sources: Shopify’s /collections/all/products/ paths, WooCommerce ?orderby= and ?per_page= params, Magento’s layered-nav URLs. Add Disallow: lines in robots.txt for unambiguous junk and <meta name="robots" content="noindex,follow"> for ambiguous-but-useful navigation states.
4. In Google Rich Results Test, paste your top 10 PDPs by revenue. Confirm Product, Offer, availability, priceValidUntil, and aggregateRating all parse. Fix any flagged warnings — the panel only renders if there are zero errors.
5. Connect Google Merchant Center if you haven’t. Free product listings have appeared in standard SERPs since 2020 and now feed AI Overviews citations. Module 63 covers the feed in depth.
6. Set up a revenue-per-indexed-URL report in GA4. Group sessions by landing_page matched against your XML sitemap. Anything in the bottom decile by revenue but indexed is a candidate for noindex or merge.
7. Decide your platform fit. If you’re on Shopify above 10K SKUs and fighting variant URL duplication, evaluate Hydrogen or a headless Astro/Next.js front end. If you’re on Magento under 2K SKUs, you are paying for capability you don’t use. Migration is the SEO decision most teams refuse to make and most regret not making.