— E-COMMERCE
Most stores pour effort into product pages and leave the category and filter architecture to the theme. That architecture is what ranks, what wastes crawl budget, and what decides whether an assistant can tell a shopper you stock the thing they asked for.
None of them are content problems, which is why publishing more blog posts rarely moves the revenue line.
01
Filter combinations generating thousands of near-duplicate URLs, consuming crawl budget and splitting authority across pages nobody should land on.
02
A heading, a grid and no reason to rank. These are the pages with the commercial search volume, and usually the least attention.
03
The same manufacturer copy as forty other retailers, which gives a search engine no reason to prefer you over any of them.
04
Hundreds of dead product URLs returning 404s or redirecting to the homepage, throwing away links and traffic they still attract.
AI SHOPPING
Shoppers now ask an assistant to compare options, check whether something is in stock, or find an alternative to a product that sold out. Answering those questions requires structured product data, not a well-designed grid.
Which means the unglamorous work of getting schema, availability and pricing correct is now a merchandising decision rather than a technical footnote.
Price, availability, GTIN, condition and review data marked up correctly and kept accurate as stock moves.
Titles, attributes and categories that match how people search rather than how your ERP names things.
Dimensions, materials, compatibility and delivery terms in crawlable text, because that is what a comparison answer needs.
Stated plainly on the page. Frequently the deciding factor, and frequently buried in a policy PDF.
ORDER OF WORK
FIRST
Decide which filter combinations deserve to be indexed, and stop the rest from existing as far as a crawler is concerned.
SECOND
Rewrite the ten categories carrying the most commercial search volume, with buying guidance rather than filler paragraphs.
THIRD
Image weight, third-party scripts and app bloat, measured on a mid-range phone rather than a desktop lab score.
FOURTH
A standing rule for discontinued lines: redirect to the nearest live equivalent, or keep the page and say what replaced it.
COMMON QUESTIONS
Anything not covered here, ask us directly and you will get a straight answer.
More than most agencies suggest, and less than you would like. Collection structure, metafields, schema, templates and internal linking are all workable. Some URL patterns are fixed, so we design around them rather than fighting them.
No. Write it for the products and categories that carry the search volume and the margin. For a catalogue of thousands, prioritising is the whole skill.
Usually yes. It wastes the crawl budget you need spent on real category pages. A small number of high-intent filter combinations are worth indexing; the rest should not be crawlable.
If a close replacement exists, redirect to it. If not, keep the page, mark it unavailable in schema and link to the category. Sending everything to the homepage throws away links and confuses shoppers.
Rarely, in our experience. Both want a fast page that answers the question. Where they do conflict, we test rather than argue.
Yes, and it usually helps. Shared keyword and feed data makes both channels cheaper to run.
The consumer search programme most stores run.
Shopify, WooCommerce and headless builds.
Where a store engagement normally starts.
FREE SEO & AEO AUDIT
Send your domain and we will send back a written audit: rankings, click-through gaps, technical blockers, and how often the answer engines cite you. No cost, no obligation, no automated PDF.
✓ Written by a senior strategist within two working days
✓ Covers Google, Bing and AI answer engines
✓ One follow-up email, and only if you want one
