Your Filters Can Generate Infinite URLs. Ecommerce SEO Is Mostly Subtraction.

The main difference is that a store generates its own pages. On an ordinary website somebody decides each page exists. On a store, every filter, sort order, variant and page of a category listing creates another address automatically, and Google describes faceted navigation as capable of generating infinite URL spaces. So the first job is usually reduction rather than creation, which is the opposite of what the phrase SEO suggests to most people. The second difference is that commercial intent sits on category pages far more than people expect. Somebody searching for a product type lands on a category, not a product, so categories often deserve the attention that gets spent on individual items. The third is that your content changes without you: stock moves, prices change, products retire, and the search side has to keep up with all of it.

It depends which kind of duplication you have, and there are usually two. Variant duplication is six colours of the same shirt each generating a near-identical page. Google’s guidance is to make sure each variant can be identified by a separate URL, and then include the canonical product URL on all variant pages, so the variants remain linkable while one page carries the ranking. Where a variant is expressed as an optional query parameter, Google says to use the URL with the parameter omitted as the canonical. The second kind is filter and sort duplication, where the same products appear reshuffled at hundreds of addresses. That is not solved with canonicals so much as by preventing the crawling in the first place. Manufacturer descriptions copied across many retailers are a third, smaller issue, and the fix there is writing your own on the products that actually matter rather than all of them.

Expect very little in the first month, because the first month is structural and structural work is invisible while it happens. Once crawl control and pagination are fixed, there is often a step change somewhere in weeks six to ten, as Google reaches parts of the range it had not seen before. That can look dramatic, because it is correcting an accumulated deficit rather than making a gain. Content-driven improvement on categories and products is slower and steadier, usually across the second and third quarters. Competitive head terms, where marketplaces dominate, take much longer and sometimes never arrive, which we will tell you at the start. The honest framing is that the first quarter buys you a store Google can read properly and the quarters after that buy you position.

Categories first, for most stores, and the reasoning is about how people search. Somebody who knows exactly which model they want searches for the product name, often with the brand, and there are not many of those searches relative to the whole market. Somebody still deciding searches for the product type, and that is a category query, with far more volume and a buyer who has not yet chosen. Categories also tend to attract links and sit higher in the structure, so improving one helps the products beneath it. The exception is a store selling branded items people search for by name, where the product pages genuinely carry the demand. Before either, though, is the structural work, because a beautifully written category page that Google reaches on page one of four is not being judged on its writing.

Faceted navigation is the filter panel on a category page: colour, size, brand, price, rating. It matters because every combination can be a separate URL, and the combinations multiply rather than add. Five filter groups with four options each produce over three thousand variations of a single category, and that is before sort orders and pagination. Google is explicit about the consequence. Crawlers typically access a very large number of faceted URLs before determining they are not useful, and if crawling is spent on useless URLs there is less time for new, useful ones. There is also a direct cost, since Google notes that crawling faceted URLs tends to consume large amounts of a site’s computing resources. The fix is not subtle. Google’s own words are that there is no good reason to allow crawling of filtered items, with the exception of the few filter pages that match real demand.

A few, usually, and choosing them is one of the more valuable judgements in this work. The test is whether people search for that combination as a thing in itself. Something like running shoes in a specific size is a real query with real volume, and a page built for it can rank and convert. A combination of four filters producing eleven products that nobody searches for is noise. In practice we usually find a handful worth keeping per store, and those get treated as proper landing pages with their own title and some text, rather than left as a filtered list. Everything else is blocked from crawling. The mistake we see most often is at the extremes: either everything is crawlable, which is the default and causes the problem, or everything is blocked, which throws away a few genuinely valuable pages.

Often yes, and it is one of the easier things to check yourself. Google’s crawlers generally do not click buttons or trigger actions that need a user, so products loaded by scrolling or by pressing load more may never be found. Google’s guidance is that this content needs to be reachable through ordinary anchor links as well. The practical test is to disable JavaScript and look at a category page. If you see twelve products where a visitor sees sixty, then Google may be seeing twelve. The fix does not mean removing infinite scroll for shoppers. It means also providing real paginated URLs with proper links behind it, so the range is navigable both ways. This is one of the most common reasons a store has products that have simply never ranked for anything.

It depends on whether it is coming back, and the decision should be deliberate rather than whatever your platform does by default. If the product will return, keep the page live and say clearly that it is out of stock, with an expected date if you have one and alternatives on the page. Deleting it throws away whatever ranking and links it accumulated, which you then have to earn again. If the product is permanently gone, redirect the URL to the closest equivalent product or to its category, so the value goes somewhere useful rather than to a dead end. What causes real damage is letting hundreds of retired products silently become errors, or keeping markup that claims availability you no longer have. We set a policy for this once and wire it in, because on an active store this situation recurs every week.

It is worth having and it is a smaller job than it is often priced as. Google requires only the product name plus one of review, aggregateRating or offers for the markup to be eligible. Most platforms output something reasonable already, so the work is frequently correction rather than creation. The part that genuinely needs ongoing attention is accuracy. Price and availability in your markup should match what a customer will actually encounter, which means it has to follow your inventory rather than being set once. Overstated availability is worse than no markup, because it brings someone to a page that cannot sell them anything. If a proposal prices schema implementation as a major line item, ask what specifically is being created rather than checked.

The principles are identical and the implementation differs, sometimes substantially. Shopify gives you less control over URL structure and robots directives, so some crawl control has to be achieved differently, and its collection and product URL conventions create their own duplication patterns to manage. WooCommerce gives you almost complete control, which is an advantage until it becomes an invitation for plugins to generate URLs nobody intended. Our Shopify and WooCommerce SEO services account for this rather than applying one template to both. What we will not do is tell you to migrate platforms for search reasons alone. If you are weighing the build itself, that is covered on our Shopify website design and WooCommerce development pages, or our ecommerce website development services page if you are starting from nothing.

Most likely because the descriptions were never the constraint. Suppose Google is spending its attention on a quarter of a million filtered URLs. Or your category pages canonicalise to page one. Or half your range only appears when a human scrolls. In any of those cases the writing on page 300 is irrelevant, because page 300 is not being considered. This is the single most common story we hear, and the money is usually not entirely wasted since the copy becomes useful once the structure is fixed. It is just that it was bought in the wrong order. Send us the store and we will tell you specifically which of those three problems you have, and you will probably recognise the answer.

Orders and revenue by landing page type, which on a store is the only measure that settles anything. We report sessions and orders split between category pages, product pages and everything else, against a baseline taken before we start. Alongside that we track how many of your URLs are indexed versus how many should be, because on an ecommerce site that ratio moving in the right direction is often the earliest real signal. We do not report average position across a keyword list as a headline, because on a store with thousands of pages that average can improve while revenue falls. If you want one number to watch, use organic revenue from category landing pages.

On broad head terms, usually not, and we would rather say that at the start than take a retainer against it. A marketplace has authority, inventory and a budget no independent store matches, and head queries in most retail categories are theirs. Where independent stores genuinely win is narrower: specific models, long-tail combinations, niches marketplaces serve badly, and categories where expertise or curation matters to the buyer. That is a real and defensible market and it is where we aim the work. The honest strategic answer is that your advantage is specificity, so the plan should be built around the queries where being a specialist beats being enormous, rather than around the terms you would most like to own.

Yes, though more for conversion than for ranking, and the distinction is worth keeping straight. Speed is a modest ranking consideration and a large commercial one, because a slow product page loses buyers who had already decided. There is also a crawling angle specific to stores: Google notes that crawling faceted URLs consumes significant server resources, so reducing the reachable URL space often improves speed as a side effect. We treat speed as a parallel workstream rather than part of this service, since it is mostly development work. Where it is the main problem we will say so and point you at our website speed optimization services instead of folding it into an SEO retainer.

Yes. We provide ecommerce SEO services in Delhi and throughout India, and we work with stores in the UAE, the UK and the USA. The work is crawling, analysis, configuration and writing, so it runs remotely without any loss, and most of it happens in your platform admin and your developer’s repository rather than in a room. Clients who prefer to meet are welcome at the studio in Vikaspuri, West Delhi. For ecommerce we recommend at least one session with whoever maintains the store. Decisions about crawl control and pagination settle faster in one conversation than over several weeks of email.