Search Engine Optimisation
Ecommerce SEO
Most ecommerce SEO problems are architectural. A store with 400 products and 40,000 indexable URLs has a crawling problem, not a content problem.
Ecommerce SEO makes category, product and filter pages rank without generating thousands of near-duplicate URLs. The Nexclick fixes faceted navigation, thin product descriptions and out-of-stock handling, then builds category content strong enough to compete with retailers far larger than you.
Is this you?
What usually prompts the call
- Search Console reports far more URLs than you have products, and most are not indexed.
- Your product pages rank and your category pages — the ones with commercial volume — do not.
- Product descriptions came from the supplier and appear identically on twenty other retailers.
- Discontinued products return 404s and you have been losing rankings and links slowly for years.
What we do
The actual deliverables
Things that appear on an invoice, not adjectives.
- Faceted navigation control
- Filter and sort parameters generate combinatorial URLs. We decide which combinations deserve indexing, which get canonicalised and which get blocked — usually the single biggest win on a large store.
- Category page content and structure
- Categories carry the commercial search volume and are usually a grid of products with no content. We add genuinely useful buying guidance without pushing the products below the fold.
- Product page differentiation
- Supplier descriptions rewritten, specification tables added, and customer questions answered on the page. Identical manufacturer copy across twenty retailers cannot rank for any of them.
- Out-of-stock and discontinued handling
- A decision framework: keep and mark unavailable, redirect to the successor, or return 410. Blanket 404s on discontinued lines throw away links and rankings quietly over years.
- Product and offer schema
- Price, availability, reviews and shipping marked up correctly so listings can earn rich results — and kept accurate, since incorrect price markup draws manual actions.
- Internal linking between related products
- Cross-sell and related-product modules built to pass link equity deliberately rather than at random, so deep catalogue pages are reachable.
- Speed on category templates
- Product grids are usually the slowest templates on a store and the ones with the highest commercial value. Image handling and lazy loading addressed at template level.
Comparison
What to do with every kind of product URL
The decisions that create or prevent most ecommerce SEO problems. Get this table right at architecture level and the majority of large-store issues never occur.
| URL type | Index it? | Canonical | Why |
|---|---|---|---|
| Main category page | Yes | Self | Carries the commercial search volume |
| Paginated page 2, 3, 4… | Yes | Self | Products must be discoverable; canonicalising to page 1 hides them |
| Single filter with real demand (e.g. "waterproof") | Yes | Self | People search it; treat it as a category |
| Two or more filters combined | No | To parent category | Combinatorial explosion, near-zero demand |
| Sort order (price, popularity) | No | To unsorted parent | Same products, different order — pure duplication |
| Product page | Yes | Self | The destination that converts |
| Product variant (size, colour) | No | To main product | Unless variants are genuinely searched separately |
| Out of stock, returning soon | Yes | Self | Keep rankings; mark availability in schema |
| Discontinued with a successor | Redirect | 301 to successor | Preserves links and sends the customer somewhere useful |
| Discontinued with no successor | No | 410 Gone | 410 removes it faster and more cleanly than 404 |
| Internal search results | No | noindex | Infinite low-quality URLs; a classic crawl-budget sink |
| Basket, checkout, account | No | noindex | No search value, and a privacy risk if indexed |
How it works
Step by step, with timeframes
Timeframes are typical rather than guaranteed, and they assume we get account access and approvals when we ask.
- 01Week 1–2
Crawl and quantify
Full crawl plus index coverage analysis. On large stores the crawl alone can run for days, and the URL-count-versus-product-count comparison usually tells the story immediately.
- 02Week 2–6
Fix architecture first
Faceted navigation, canonicals, pagination and out-of-stock rules. Nothing else pays off until crawl effort is going to pages that matter.
- 03Month 2–5
Build category content
Priority categories first, ordered by search volume against your margin rather than by volume alone.
- 04Month 3 onwards
Product and schema rollout
Rewrites prioritised by revenue and impression data. Rewriting 4,000 products at once is neither affordable nor necessary.
What you get
Reporting and ownership
- A URL architecture document setting out what is indexable, what is canonicalised and what is blocked.
- A category priority list ordered by search volume against your actual product margin.
- A written out-of-stock and discontinued-product policy your team can apply without asking.
- Monthly reporting on organic revenue and category-level position, not just sessions.
- Validated Product and Offer schema across every template.
Tools and platforms
- Screaming Frog
- Google Search Console
- Google Merchant Center
- Ahrefs or Semrush
- Server log analysis
- Schema.org validator
- Looker Studio
Timeline
How long this actually takes
Architectural fixes show in crawl and index data within two to six weeks, and that is often visible before any ranking change. Category content takes three to six months to mature. Product-level work compounds slowly and is judged over quarters rather than months. Seasonality complicates all of it: a store whose revenue triples in December cannot judge results against the previous month, so we benchmark year-on-year from the start rather than being surprised by it in January.
Pricing model
Monthly retainer
Monthly retainer, sized against catalogue depth. Very large catalogues usually begin with a fixed-price architecture project before ongoing work makes sense.
Questions
Ecommerce SEO questions
Should category pages have text on them?
Yes, but not a wall of it beneath the products where nobody reads it. A short, genuinely useful introduction above or beside the grid — what to consider when buying, how the range differs — serves both the customer and the search engine. Hidden text dumped in the footer serves neither.
What do we do with products that go out of stock seasonally?
Keep the page live, mark availability correctly in schema, and tell the customer when it returns. Removing and recreating it every season means starting from nothing each time. Seasonal products that keep their URL accumulate authority year after year.
Do we need unique descriptions for every product?
Ideally, but prioritise by revenue and impressions rather than attempting all of them. Your top 200 products by revenue justify bespoke copy. A long tail of low-margin lines can share a templated structure with genuinely varying specification data, and that is a reasonable trade.
How do we handle products sold in several categories?
Pick one canonical URL for the product and keep it consistent regardless of the browsing path. Category-nested product URLs that change depending on how you arrived are one of the most common duplication problems on custom-built stores.
Is Shopping feed optimisation part of this?
It overlaps significantly. The same product data problems — poor titles, missing attributes, wrong categories — hurt both organic listings and Shopping performance. We will flag feed issues found during the work, though managing the feed for paid campaigns sits under our Google Shopping service.
Our platform will not let us change URL structure. What then?
Common, particularly on Shopify. The answer is to work the levers the platform does give you — canonicals, internal linking, content, schema — and be honest about the ceiling. Sometimes the constraint is genuinely material, and that is a replatforming conversation rather than an SEO one.
How do we measure ecommerce SEO properly?
Organic revenue and margin, not sessions. Traffic to a discontinued line is worthless and traffic to a high-margin category is not, so a report showing sessions alone can look excellent while revenue falls. We report at category level against margin from month one.
Last reviewed 28 July 2026.
Tell us what you are trying to fix
A 20-minute call, no pitch deck. The Nexclick will tell you what we would do, roughly what it costs, and whether we are the right people for it.