The quick answer: most ecommerce SEO advice optimizes for traffic, and traffic is not the constraint. The pages that make money are category and collection pages, not blog posts, because that is where buying intent lands. Fix the four things that actually move revenue, in this order: make category pages rank for the terms people buy with, get product pages indexed and differentiated from the manufacturer description everyone else copied, stop faceted navigation from burning crawl budget on infinite filtered URLs, and make the site fast enough on mobile that the traffic you already have converts. Rankings for informational keywords are a nice byproduct, not the objective.
The reason so many ecommerce SEO programs produce charts that go up and revenue that does not is that they were built to grow sessions. Sessions were never the problem.
| Order | Work | Why it comes first |
|---|---|---|
| 1 | Indexation of product and category pages | A page search engines cannot see has no ranking problem to solve |
| 2 | Category pages built for buying terms | Highest commercial intent asset on the site, and usually the thinnest |
| 3 | Faceted navigation brought under control | Stops crawl attention draining into filter permutations |
| 4 | Mobile speed | Converts the traffic you have already paid to acquire |
| 5 | Blog and informational content | Real, but it compounds slowly and converts at a fraction of the rate |
Category Pages Are Where the Money Is
Someone searching “waterproof hiking boots” is shopping. Someone searching “how to break in hiking boots” already owns them. The first query should land on a category page. The second belongs on a blog post that will convert at a fraction of the rate.
Most stores get this backwards. They pour effort into a content calendar and leave category pages with a heading, a grid of products, and no text at all. Meanwhile the category page is the highest commercial intent asset on the entire site.
What a category page needs to compete is not much, and it is not a thousand word essay stuffed underneath the products where nobody reads it:
- A title and heading using the phrase buyers actually search, not the internal merchandising label. “Waterproof Hiking Boots” beats “Outdoor Footwear Collection 3”.
- A short block of genuinely useful text above or near the grid that helps someone choose. What separates the options, what to consider, who each suits.
- Internal links to the related categories a shopper would logically consider next.
- A URL that stays stable. Category URLs get links and rankings over years, and rewriting them for a redesign throws that away.
The test for the text is whether a shopper would find it useful. If it exists only for search engines, it reads that way to both.
Product Pages Nobody Can Find
Two failures dominate here, and they compound.
The first is duplicate description text. If you sell products you did not manufacture, the description in your catalog is almost certainly the same one on every other retailer’s site, because everyone imported the same supplier feed. There is no reason for a search engine to prefer your copy of it, and usually it does not.
Rewriting every product description is unrealistic for a large catalog, so prioritize. Rewrite the products that already get traffic, the ones with the best margin, and the ones you genuinely know more about than your competitors. Leave the long tail on the supplier text. A partial fix aimed at the right products beats a complete fix nobody finishes.
The second is that product pages are simply not indexed. Large catalogs routinely have thousands of products search engines have never crawled, usually because they sit too many clicks from the homepage, are absent from the sitemap, or are buried behind filtered navigation. Check coverage in Search Console before assuming a ranking problem. A page that is not indexed does not have a ranking problem, it has a discovery problem, and they need different fixes.
Reviews on product pages do real work here too, because they add text nobody else has and they answer the questions buyers are actually weighing.
Faceted Navigation Will Eat Your Crawl Budget
This is the most common serious technical problem in ecommerce SEO and the least visible from the outside.
Every filter combination a shopper can select generates a URL. Colour, size, brand, price band, rating, sort order, and any combination of them. A catalog with a handful of filters can generate hundreds of thousands of crawlable URLs, nearly all of them near duplicates of a page that matters.
Search engines allocate finite attention to any site. Spend it on filtered permutations and your actual category and product pages get crawled less often, which slows how quickly new products and price changes are picked up.
The approach that works is deliberate rather than blanket:
- Decide which filtered pages have genuine search demand. “Black waterproof hiking boots” plausibly does. “Boots sorted by price descending, page 4” does not.
- Let the valuable ones be indexable, with their own titles and descriptions.
- Canonicalize the rest to the parent category.
- Block the truly worthless parameters, sort order especially, in robots.txt.
- Keep your XML sitemap limited to canonical, indexable URLs. A sitemap full of filtered pages actively teaches crawlers the wrong lesson.
Speed Is a Revenue Problem, Not a Score
Ecommerce sites are heavy by nature. Large images, review widgets, chat, analytics, personalization, a tag manager containing years of accumulated scripts nobody has audited.
The reason to care is not the score in a testing tool. It is that mobile shoppers on imperfect connections abandon slow pages, and every one of those is traffic you already paid for. Speed work is conversion work that happens to also help rankings.
The highest yield fixes are boringly consistent across stores: serve images in modern formats at the size actually displayed, lazy load anything below the fold, audit the tag manager and remove what nothing uses, and defer third party scripts so the page is usable before the chat widget arrives. We covered the business case in the real cost of bad website speed, and it applies with more force to a store than to a brochure site because the loss is immediate and measurable.
Structured Data That Earns Its Place
Product schema is one of the few structured data implementations with a direct commercial payoff, because it drives the price, availability and review stars that appear in search results and make a listing more clickable than the ones around it.
Get the basics right and keep them accurate: product name, image, description, brand, price, currency and availability. Aggregate rating and reviews where you genuinely have them.
The failure that causes trouble is drift. Schema showing a price the page no longer charges, or in stock on a product that is not, is worse than no schema at all, and at scale it invites a manual action. If your catalog changes frequently, generate the markup from the same data the page renders from rather than maintaining it separately.
Measure Revenue, Not Rankings
A ranking report is not a performance report. The questions worth answering monthly are narrower and harder:
- Which landing pages produced revenue from organic search, and how did that change?
- What is organic revenue per session, and is it rising or falling as traffic grows?
- Which categories are gaining and losing, and does that match what you are merchandising?
- Where do organic sessions abandon, and is it a page problem or a product problem?
Revenue per session is the number that separates a program that is working from one that is only growing. Traffic climbing while revenue per session falls means you are ranking for progressively less commercial terms, which is the failure mode this whole article is about.
Where SEO and Paid Search Should Meet
Your shopping campaigns already know which products convert and at what value. That is search data you have paid for, and it should inform which categories get SEO attention rather than sitting in a separate account nobody cross references.
Getting that comparison right depends on both channels measuring revenue rather than events, which is why closing the loop on sales that finish offline matters for a store too: phone orders and long consideration purchases are invisible without it.
It works in both directions. Products that convert well in paid are worth ranking organically, and categories that rank organically can often have their paid bids reduced. Running the two channels as strangers is one of the more expensive habits in ecommerce, and our guide to Performance Max for ecommerce covers the paid side of that relationship.
Frequently Asked Questions
Should I write blog content for my ecommerce store?
Only after the category and product pages are working. Blog content targets people who are researching rather than buying, so it converts at a fraction of the rate and takes longer to pay back. It has a role in building authority and capturing early stage demand, but a store with thin category pages and unindexed products has more valuable work available. The common mistake is starting with content because it feels productive, while the pages that would actually make money stay untouched.
How long does ecommerce SEO take to show results?
Technical fixes such as indexation and speed can show up within weeks, because you are removing an obstruction rather than building authority. Ranking improvements on competitive category terms generally take three to six months of consistent work, longer in crowded categories. The sequence matters: fixing indexation first means every later improvement applies to pages search engines can actually see.
Do I need unique descriptions for every product?
Ideally yes, realistically no, and prioritizing is the whole skill. Rewrite the products that already receive traffic, carry the best margin, or where you genuinely know more than the supplier does. Leave the long tail on imported text. A partial rewrite aimed at the products that matter beats a complete rewrite that never finishes, and duplicate text mainly hurts on products competing directly against other retailers selling the identical item.
How do I stop filters creating thousands of duplicate pages?
Decide deliberately which filtered combinations have real search demand and let those be indexable with their own titles and descriptions. Canonicalize the remainder to the parent category, block worthless parameters such as sort order in robots.txt, and keep the XML sitemap restricted to canonical indexable URLs. The goal is not to eliminate filtered pages, which shoppers need, but to stop search engines treating every permutation as a page worth crawling.
Where to Start
Open Search Console and compare how many product URLs you have against how many are indexed. If the gap is large, that is your project, and no amount of content or link building matters until it closes.
If the gap is small, look at revenue per organic session over the last year. If traffic rose while that number fell, you have been winning the wrong keywords.
We work with stores on exactly this through our ecommerce marketing practice, and where the constraint turns out to be the platform rather than the optimization, our guide on what to look for in a website redesign is the honest place to start instead.