Most store traffic is lost in the facets, not the blog.
Ecommerce SEO on a real catalog is mostly architecture. Which URLs exist, which of them a crawler is allowed to spend time on, and whether your category pages give a buyer any reason to choose. The content calendar matters far less than the tree it hangs off.
Request a store audit-
Architecture first
- Category tree, faceted navigation and crawl budget. The decisions that determine everything downstream.
-
Product page economics
- On a large catalog you cannot rewrite everything. We decide which products earn the effort and why.
-
The Arabic side
- Facet values, category names and attributes in Arabic, not just the theme strings. This is where regional stores lose buyers.
-
What we report
- Revenue and transactions from organic by category. Sessions on non-commercial queries are not a result.
How a store engagement runs
-
Architecture audit
Weeks 1 to 3
- Full crawl with facet permutations mapped and counted
- Crawl budget attribution: what the bot actually spends its time on
- Category tree against the queries your market uses, in both languages
- Cannibalisation between category, subcategory and product pages
-
Facet and index strategy
Weeks 3 to 5
- A decision per facet type: indexable, crawlable, or neither
- Implemented in routing and parameter handling rather than noindex alone
- Pagination and sort order handled so the crawler stops looping
- Internal linking rebuilt so priority categories get visited
-
Category and product work
Weeks 5 onward
- Category pages given content that helps somebody choose, not keyword filler
- Product pages prioritized by margin and search demand rather than alphabetically
- Product schema valid and matching the visible price and availability
- Arabic facets, attributes and category names researched rather than translated
-
Reporting
Monthly
- Organic revenue and transactions by category
- Index coverage and crawl budget movement
- Field Core Web Vitals by template, category against product
- In the portal beside paid media, so channel overlap is visible
How we report it
The clients behind the claims on this page
Published engagements with the numbers in full, and the citation-overlap evidence that decides how we report AI visibility.
Delta Medical Labs
5,570 pages cited across AI assistants, July 2026
Eduverse
171 pages cited across AI assistants, August 2026
- 89.1%of the sites ChatGPT cites, Perplexity never touches for the same questionWellows, 804,058 answers, Sept 2025 to May 2026
- 79.6%of sources appear on one engine only22.7M citations across 1,146,483 questions, 2026
- 46xgap in brand citation rate between ChatGPT at 0.59% and Perplexity at 13.05%Study of 34,234 AI responses, 2026
Delta’s AI Overview count is 27 times its ChatGPT count. Eduverse’s top and bottom engines sit 18 pages apart. Same agency, same method, opposite shapes. Any single score we quoted you would have described neither.
Request a store auditClient reviews
All reviewsThe architecture audit
What we map before touching a product page
Category architecture and facet handling change everything downstream, at a fraction of the cost of rewriting a catalog.
Facet permutations, counted
Every colour, size and price combination generating a URL. On most stores the count is far higher than anyone expects.
Crawl budget attribution
What the bot actually spends its time on, read from logs where the host provides them.
Category tree against real queries
In both languages. Category names translated literally rather than researched is the standard regional fault.
Cannibalisation between levels
Category, subcategory and product pages competing for the same term.
Product schema validity
Price and availability matching what the page visibly says. This is also what assistants read when asked to compare.
Arabic facet values
Filtering by size or colour in Arabic that returns English values or an empty set. A revenue bug before it is an SEO one.
We report organic revenue and transactions by category. Sessions on non-commercial queries are not a result.
Questions about ecommerce SEO
With the tree, not the products. Category architecture and facet handling change everything downstream, and they are a fraction of the work of rewriting a catalog. Product pages get prioritized by margin and demand afterwards.
Some of them. A filter combination people actually search for deserves a page. A four-way combination nobody has ever typed does not. The decision is made per facet type with the query evidence, and it is implemented in routing rather than with a noindex tag.
No, and on a large catalog it is not affordable. The question is which products justify it. For the rest, structured data, reviews and internal linking do more per hour spent.
It is usually the biggest untapped piece. The common faults are Arabic pages canonicalised to English ones, facet values left in English, and category names translated literally rather than researched.
WooCommerce and WordPress directly, plus Shopify, Magento and custom stacks where we specify and your developers ship. Regional platforms we handle case by case and will tell you honestly where our experience is thinner.
Organic revenue and transactions by category. If sessions rise and revenue does not, that goes in the report as a problem rather than a headline.
