SEO that fills carts, not dashboards.

Technical fixes and pages already sitting on page two can move within weeks; a new category cluster usually takes two to three quarters to earn its position, then keeps compounding. The useful framing is that search is a balance-sheet channel rather than a monthly one — organic search accounts for 35-55% of revenue-generating sessions at established retailers, and that share grows while click costs rise. We publish a monthly read on SEO orders from the first month so you can see the trend long before the totals look impressive.

Category pages, almost every time. They sit closest to the purchase, they carry the head terms, and they are the pages a shopper lands on with a card in hand. In a benchmark of 41 stores, category pages produced 60-70% of organic sessions, which is the number that settles the argument. Content still matters, but its job is to answer the questions a collection page cannot and to feed authority into it — a buying guide that links to three collections is worth ten news posts.

Ask three questions. What will you measure — an answer about traffic and rankings is a report, an answer about organic orders and revenue by page type is a channel. Which pages will you work on first, and why those? And what is the link building method, in detail? Then ask what they would stop doing on your store today. A useful answer arrives with reasoning attached, which tells you more about the next twelve months than any logo wall in a deck.

A fixed monthly fee scoped to the roadmap, quoted before anything starts and separate from any media spend. Scope tracks catalog size and how much page rewriting is needed rather than an arbitrary tier, and after a first review you get a plan, a ranked roadmap and a target. Where the honest engagement is an audit and a category map rather than a retainer, we scope it that way and say so. Content, tracking and accounts stay in your name throughout.

We treat filters as a demand question before a technical one. A filter combination people genuinely search for — a size, a colour, a material, a compatibility — deserves a real indexable page with its own copy and internal links. Everything else gets parameter and canonical rules so it stops eating crawl budget and stops competing with the collection it came from. Getting this wrong is the most common reason a large catalog has thousands of indexed URLs and very little organic traffic.

It is, and it is fixable in priority order. Identical copy means your page brings nothing to the SERP that twenty other listings do not, and it leaves the shopper's real questions — fit, materials, compatibility, what is in the box, how it ships — unanswered. We rewrite the products that carry search demand or margin first, give your team a description template and checklist for everything else, and keep the manufacturer specs as structured data underneath rather than as the whole page.

They share buyer research and should share reporting. Paid tells you within days which queries and product sets convert, which sharpens the organic roadmap; organic then takes over the terms where click costs are punishing and holds them for years. We run both with our ecommerce PPC team on one plan, and the monthly review looks at blended cost per order rather than letting each channel claim the same sale twice.

Same discipline, wider surface, and worth keeping in proportion. Across 94 ecommerce websites, ChatGPT referrals generated $474,000 against $32.1 million from non-branded organic search — real, growing, and still a fraction. Answer engines cite clearly written, well structured, corroborated pages, which is what good category and guide work produces anyway. We add the structured data and entity consistency those systems read and track citations through our GEO and AEO work.

It changes the sales more reliably than it changes the rankings. Speed is one modest signal among many for search engines, but on a phone in a shop queue a slow collection page is simply abandoned, and that shows up in revenue per session immediately. We measure field data rather than a lab score, fix the heaviest offenders — uncompressed imagery, third-party scripts, render-blocking apps — and check the result against add-to-cart rate, which is where the money is.

No, and we will decline the request politely. Bought links from networks are the one tactic that can cost a store its organic channel outright, and the sites selling them are the sites algorithms are best at spotting. What we do instead is make something worth citing — original data, a useful tool, a guide that does not exist yet — and put it in front of journalists, newsletter writers and partners. Slower to start, and it does not evaporate in an update.

They get a plan instead of a 404. A temporarily unavailable product keeps its page, states the situation honestly and offers a notify option, because deleting it throws away rankings you will want back next season. A discontinued line is redirected to the closest replacement or to its parent collection, never to the homepage, and the internal links pointing at it are updated. Seasonal ranges stay live year-round with dated content, which is how they rank on day one of the next season.

Often the best arrangement, and we scope for it. You get the category map, briefs, description templates, the review checklist and the technical fix list, while we handle editing, structured data and authority outreach. Store teams with a strong merchandiser usually save money that way and produce better pages, because nobody knows the range like the person buying it. Where writing capacity genuinely is not there, our writers cover it and still interview your team first.

By reporting source and influence separately instead of pretending one model settles it. We tag every commercial page, split organic orders and revenue by page type, compare year over year rather than month over month, and read the post-purchase survey next to the analytics. Where a competitor leaves a SERP or Google changes a layout, we say so rather than take credit. That habit is why most of our clients keep the same reporting view for years.

Weeks one and two are the audit: crawl, indexation, render, links and the pages already close enough to move. Week three delivers the category and keyword map with expected revenue per cluster. From week four the commercial page rewrites begin while developers work the fix list, and authority work starts with whichever asset your data supports. By day ninety you have measurable movement on existing pages, a rebuilt collection set and a reporting view that reads in orders.

We work in your Google Search Console and analytics accounts — those two are non-negotiable and stay yours — plus our own licences for keyword research, crawling and backlinks, so nobody has to buy a tool stack to begin. If your team already pays for one of the large platforms we will use it rather than duplicate it. Everything we produce, from the keyword sheet to the crawl exports, is delivered in files you keep, and no report sits behind a login only the agency holds.