The Complete Guide to Ecommerce SEO
Ecommerce SEO fails more often from doing the right things in the wrong order than from doing the wrong things. Fix indexing before content, structure before pages, categories before products. This guide is that sequence.
What This Guide Covers
Six workstreams, in the order they pay. Technical foundation, structure, category pages, product pages, content beyond the catalogue, then off-site. Measurement runs alongside all of it.
Why the order is stated first. Because it is the argument of the whole guide.
Each stage depends on the one before it being sound. Work done out of sequence is not merely less effective. It is frequently wasted entirely, which block two demonstrates.
What this guide is not. A rebuild of every subject in the cluster.
Several of these have their own guide because they deserve two thousand words rather than three hundred. Where that is the case this page gives the position and points across rather than producing a worse version of a page that already exists.
If you are starting from the beginning. Our introduction to ecommerce SEO covers what the discipline is before this guide covers how it is sequenced.
How to use it if you are buying rather than doing. Read blocks two and three.
Those two contain the whole of what separates a supplier who understands shops from one who produces content. If a proposal does not address either, it is describing work on a website rather than work on a catalogue.
If your shop already exists and underperforms. That is a different starting point, covered in block ten.
Most shops reading this are in that position rather than building something new. The order of work is not the same.
Why Order Matters More Than Effort
A shop can spend a year doing genuinely good work and see almost nothing, because the work was done in an order that prevented it counting.
A concrete example. A shop commissions product descriptions for its eight hundred best selling items.
The writing is good. It takes four months and costs real money. Nothing happens.
What was actually wrong. The shop's filtered navigation was generating crawlable pages at a scale that consumed most of the attention the site received.
Search engines were spending their time working through colour and size combinations rather than reaching the products. The new descriptions sat on pages that were visited rarely, so months passed before most of them registered at all.
What the correct order would have produced. The same descriptions, counting immediately.
Two weeks of technical work first, then the same four months of writing, with every page seen promptly. Identical spend, entirely different outcome.
Why this is the most common failure we see. Because content is the visible part.
A store owner can read a description and judge it. Nobody can look at a crawl budget being wasted, so the invisible problem stays invisible while money goes into the visible one.
The same pattern in three other forms. All of them common.
Category content written for pages that duplicate each other, so the effort is split across several near identical versions. Product descriptions added to items nobody searches for. Links acquired to a shop whose structure prevents that authority reaching anything below the homepage.
What they share. Real work, correctly executed, placed where it cannot count.
The rule that follows. Nothing published matters until the shop can be worked through properly.
The Technical Foundation
Five things have to be right before anything else pays. None of them is visible to a customer, which is why they are so frequently skipped.
Indexing. Whether the pages you want found are actually eligible to be found.
Shops routinely block sections of themselves by accident. They carry directives left over from a staging site. They exclude pages a previous supplier judged unimportant.
Crawl. Whether attention is going to the pages that matter.
Per block two this is where the largest silent losses happen. On a big catalogue it is almost always faceted navigation causing it.
Duplication. Near identical pages competing with each other. Variants, filters and the same product reachable by several routes.
Redirects. What happened to everything that used to exist. Discontinued products, renamed categories and anything changed in a previous redesign.
Speed. Which on a shop is a crawl problem as much as a ranking one, covered in our page speed guide.
How to tell whether yours is sound. Two questions answer most of it.
How many pages does the shop have? How many are indexed? A large gap in either direction is the finding. Far fewer indexed than expected means something is blocking. Far more means filters or variants are generating pages nobody intended.
Where the detail sits. Our technical SEO guides cover the mechanics. Our mobile guide covers why the mobile version is the version that counts.
Structure
Structure sets the limit everything else works within. A shop can have excellent content on a structure that prevents it earning anything.
The single most valuable idea in it. Most shops inherited their category tree from a supplier's product data.
That tree reflects how an industry organises stock rather than how customers search for it. The two are frequently nothing alike. A shop organised by manufacturer when customers search by application is competing for terms that do not exist.
How to test whether yours came from a feed. Look at the top level.
If the first thing a visitor chooses between is a manufacturer name, the tree came from the supplier. If it is a type of thing or a use, somebody built it for customers.
What else it decides. How deep products sit, how filters behave, how pages link to each other and whether the catalogue can be discovered at all.
Our structure guide is the longest page in this cluster for that reason. It covers diagnosing a shop that already exists rather than designing one.
Category Pages
Category pages carry most of the commercial search on a shop. On most shops they are also the emptiest pages on the site. That gap is the largest single opportunity in this cluster.
Why they carry it. Because of what a category search means.
Somebody searching for a type of product has not chosen yet. Somebody searching for a specific item has. The first group is larger and still deciding, which means the page they land on gets to influence the decision.
What the typical category page is. A heading, a grid of products and nothing else.
What it should also do. Answer how to choose.
What the differences between these products are, what to look for, what suits which situation. Buying guidance rather than a paragraph of padding beneath the grid.
The most valuable version of this work. Creating a category that does not exist yet.
Where customers search for a grouping the shop does not offer as a page, building it is frequently the single highest return action available, because the demand is already there and nothing is competing for it internally.
Our category page guide covers it in full, including how to find categories customers search for that your shop does not have yet.
Product Pages
Most product pages will never rank for anything. That is entirely normal. Accepting it is the difference between a workable plan and an impossible one.
Why. Because there is no search behind most individual products.
A shop with five thousand items has perhaps a few hundred that anybody looks for by name. The rest exist to be bought once somebody is already on the site rather than to be found from outside it.
What follows. The work is choosing which ones could rank rather than treating five thousand pages equally.
Margin and demand together, not one or the other. Our product page guide covers how to identify them in your own catalogue.
The related problem. Supplier descriptions, identical across every shop selling the item.
The answer is not rewriting five thousand of them, which our guide to product descriptions explains. Reviews are frequently the cheapest realistic source of original text at that scale.
Content Beyond The Catalogue
A shop that publishes nothing but products can only be found by people already looking for products. That is a real limit. It is where most shops stop.
Three kinds of content that pass it. Each doing a different job.
Buying guides. For people who know roughly what they need without knowing which one.
These reach somebody earlier than a category page does. They earn the right to recommend rather than only to list.
Comparisons. For people choosing between two specific options.
Closer to purchase, lower in volume, frequently the highest converting content a shop publishes because the reader is one decision from buying.
Problem led content. For people who have a situation rather than a product in mind.
Somebody searching how to fix a thing has not yet worked out what to buy. A shop that answers the question gets to be the one that supplies the answer.
What all three share. A category page cannot carry them.
It has to remain browsable. The moment it becomes an article it stops doing its primary job, which is why this content lives separately.
How much of it a shop needs. Less than the volume of advice on this suggests.
A small number of pieces that genuinely answer something, kept current, outperforms a blog updated weekly with material nobody searched for. The test is whether somebody with the problem would be glad they found it.
The seasonal case. Content tied to a season needs to exist months before it, covered in our seasonal guide.
Off-Site
Links and mentions from elsewhere matter. They are also, on almost every shop we look at, not the thing holding it back.
Why it is rarely the first fix. Because links raise a ceiling that structure has already set lower.
A shop whose category pages are empty and whose crawl is consumed by filters does not have a link problem. It has the problems in blocks three to five. Links would make a badly organised site slightly more visible rather than fixing it.
When it genuinely is the constraint. A shop that has done the rest and sits behind competitors on otherwise comparable pages.
That is a real situation. It is simply much rarer than the industry selling links suggests.
What works for a shop specifically. Being worth mentioning.
The content in block seven is the mechanism, since nobody links to a product listing but people do link to an answer.
What to be careful of. Anything sold by volume.
A shop is an obvious target for link selling because the commercial intent is visible. Packages offered by the hundred are the clearest signal that whoever is offering them has not looked at the shop, since nothing about a specific catalogue could be diagnosed from the outside at that price.
Our guides on backlinks cover the subject properly, including the same caveat that links are usually not the first fix.
Measurement
Reporting is not fixing. In ecommerce there is one distinction that matters more than every other number combined.
Brand against non-brand. People searching for your name against people who found you without it.
Demand for your own name grows for reasons that have nothing to do with this work: advertising, word of mouth, returning customers. A report combining the two can show healthy growth while the work has achieved nothing at all.
What to measure instead. Revenue from non-brand organic traffic, ideally at margin rather than at revenue.
And what to report before revenue moves. The leading indicators.
Indexed pages, non-brand impressions and category level visibility all move before money does. Reporting those in month two is how a client sees work happening without anybody having to overstate what it has produced yet.
And what distorts it. Seasonality, which makes month on month comparison in this sector actively misleading.
Our measurement guide covers the whole subject. Our guide to timescales explains what to expect to see and when.
What Improving An Existing Shop Involves
Everything above assumes a clean sequence. Most shops are not starting clean. They have an existing site, existing rankings and existing decisions somebody made years ago. The order of work changes accordingly.
The sequence for an existing shop. Audit, fix, then build. In that order and not overlapping.
Audit first, resisting the urge to act during it. The strongest discipline available here.
A shop that starts fixing things while still diagnosing ends up unable to tell which change caused which effect. Finish the picture before touching anything.
What the audit has to establish. Four things.
What is indexed against what should be. Where crawl attention is going. What is duplicating what. And which pages currently earn anything, since that decides what must not be broken.
Fix before build. Repair what is losing before adding what might gain.
Fixing is usually cheaper, faster and lower risk than creating. It frequently releases visibility the shop had already earned rather than requiring new visibility to be built.
The specific danger on an existing shop. Protecting what already works.
A restructure can improve a shop's future and destroy its present in the same afternoon. Anything that changes addresses needs redirects planned before the change rather than discovered afterwards, which our migration guides deal with.
What frequently comes out of an audit on an established shop. That the problem is smaller than expected.
A shop underperforming for years is rarely broken in twenty ways. It is usually broken in two, with eighteen symptoms. Finding the two is the entire value of the exercise. It is why a proper audit is worth more than a list of two hundred issues sorted by severity.
If the shop used to do better than it does now. That is a diagnosis question rather than an improvement one.
Our guide to why shops fail separates never ranked from used to rank and stopped. Our page on penalties covers the case most people wrongly assume they are in.
Where The Rest Is Explained
This guide covers the sequence. Several subjects inside it, plus several decisions around it, have their own treatment.
Anything decided while a shop is being built. Structure, category design, product page layout, platform and speed as build decisions all sit in our ecommerce web design guides.
That is the boundary running through this whole cluster. That silo answers what to decide while building. This one answers how to improve what already exists.
Buying decisions. What this costs is in our pricing guide, whether it is worth it in our worth it guide, then how it compares with advertising in our comparison with paid.
Who does the work. What a service should include and how to compare two proposals is in our guide to choosing an agency, with the case for doing it yourself in our comparison of the two.
Specific subjects. Platform capability in our platform guide, images in our image guide, reviews in our reviews guide and the marketplaces in our guide on competing with them.
Generic pricing and strategy, meaning anything not specific to a shop, sits in our local SEO guides.
Right work.
Wrong order.
Eight hundred product descriptions, four months, real money, nothing happens. The filters were consuming the crawl the whole time. Two weeks of technical work first would have made the identical spend count immediately.
What is included every month:
£350 per month, one target area. No setup fee, nothing billed separately.
Twenty-two guides.
One subject.
This guide is the sequence. The rest of the series covers structure, category and product pages, technical health, measurement, the marketplaces and everything a store owner has to decide.
