Service · Ecommerce SEO
At scale, architecture beats content
On a ten-page site you can fix SEO page by page. On a store with twelve thousand SKUs and faceted navigation, the problems are structural — and so are the wins.
Category architecture, indexation control, internal linking and the handling of variants, filters and out-of-stock products decide most of what an ecommerce site is capable of ranking for.
In short
Category and product architecture, faceted navigation, indexation control and internal linking for stores where the catalogue is too big to manage by hand.
From [ $X ] · audit first
Almost always starts with a diagnostic, because at this scale the wrong fix is expensive to deploy and expensive to undo.
Symptoms
What usually brings ecommerce sites here
Google has indexed 200,000 URLs and we have 8,000 products.
Faceted navigation is generating combinations and they are all in the index.
Our category pages do not rank but our products do.
Backwards. Categories should carry the head terms.
We added filters and traffic fell.
A crawl budget and duplication problem arriving three months after the change that caused it.
Discontinued products are still ranking.
And returning 404s, or redirecting to the homepage, which is worse.
Every variant has its own URL.
Colour and size splitting authority across dozens of near-identical pages.
None of these are content problems. You cannot write your way out of an architecture that is working against you.
What ecommerce work covers
Four structural decisions that set your ceiling
Get these right and ordinary content starts ranking. Get them wrong and excellent content underperforms for reasons nobody can see from the page itself.
1
Category architecture
Which categories exist, how they nest, and which deserve to target head terms. Most stores mirror their internal merchandising, which rarely matches how people search.
2
Faceted navigation
Which filter combinations get indexed, which get crawled but not indexed, and which are blocked outright. Handled with canonicals, robots rules and parameter design rather than one blunt instrument.
3
Indexation control at scale
Out-of-stock and discontinued products, thin variants, paginated series, and internal search pages. Decided as policy and applied by template, not product by product.
4
Internal linking
How authority moves from your strongest pages to the categories that need it. On big catalogues this is the highest-leverage lever available and the least used.
What you get
Rules your developers can implement by template
Every recommendation is written to be applied across a catalogue, not to a single URL. If a fix cannot be templated it does not belong in an ecommerce audit.
- A category map. What should exist, what should merge, what should be removed, and which page owns which head term.
- A faceted navigation policy. Index, crawl, or block — per facet, with the reasoning and the implementation method.
- Indexation rules by page type. Including what happens to a product when it goes out of stock, and when it is gone for good.
- An internal linking plan. Where the links go, from which templates, and why.
- A migration-safe sequence. Ordered so you can deploy in stages and still tell what each stage did.
How it runs
How ecommerce work runs
1
Crawl and log analysis
What Google actually crawls versus what you think it crawls. Log files matter more here than anywhere else in SEO.
2
Architecture review
Categories, facets, variants and pagination assessed against the queries you need to own.
3
Specification
Templated rules, written for developers, with the edge cases named up front.
4
Staged rollout
Deploy in a sequence that keeps the effects readable, and watch indexation as it lands.
Fit
Who this suits
Worth it if
- You have more SKUs than you can manage page by page
- Your catalogue changes — products come and go, stock moves
- You have faceted navigation, or you are about to add it
- You have developer capacity, in-house or agency
- You are planning a replatform and want to not lose traffic this time
Not worth it if
- You have thirty products and no filters — ordinary SEO covers you
- Nobody can deploy template-level changes
- You want blog content rather than structural work
- You need the results before the next quarter closes
FAQ
Questions about ecommerce SEO
Should out-of-stock products be removed?
Usually not. If it is coming back, keep the URL live and say so on the page. If it is gone permanently, redirect to the closest equivalent or the parent category — not the homepage, which Google treats as a soft 404. Deleting a page that has links and rankings is throwing away an asset.
Should filtered pages be indexed?
Some should. A filter combination people genuinely search for — a brand within a category, say — can be a strong landing page. The combinations nobody searches for should not be indexed and often should not be crawled. It is a per-facet decision, which is why blanket advice about facets is usually wrong.
We are replatforming. When should you be involved?
Before the URL structure is decided, not after launch. Most of the traffic lost in a replatform is lost to decisions made in the planning phase, and by launch the expensive ones are already baked in.
Do product descriptions need to be unique?
Manufacturer descriptions used verbatim across every retailer rarely help you win. But rewriting twelve thousand of them is not the best use of your budget either. Prioritise the products that can realistically rank, and fix the architecture first — it usually returns more.
How do you handle variants?
Depends on whether people search for the variant. Colour rarely warrants its own indexed URL; size almost never does; capacity or model number often does. The default is to consolidate and make exceptions deliberately.
Let us look at the catalogue
Tell me the platform, roughly how many SKUs, and what is not ranking. I will tell you whether it is architecture or something else.
Or reach me directly — mehuldedhiaa@gmail.com · +91 93769 10109