Internal linking for online stores
Why most ecommerce stores only link top-down, category to product, and how to add the cross-links that turn a catalogue into a genuinely crawlable site.
Internal linking for online stores means adding links a pure catalogue structure never generates on its own: category to category, product to related product, and buying guide into the products it discusses. Navigation alone only links top-down, and a page with no other way in is the hardest page on the site to find.
This is one piece of the practical architecture work covered in the Salla SEO guide, which sets out where the rest of the leverage on a hosted store sits.
On this page
Why does a typical store's link structure fall short?
Because it is built for browsing, not for crawling or for connecting related ideas. The standard pattern — home page to category, category to subcategory, subcategory to product — is a tree, and a tree has exactly one path to every page. Remove that one path and the page becomes an orphan, reachable only through search or a direct link.
Google's guidance on crawlable links is specific about what counts: a real, followable <a href> pointing at a URL, not a button that requires JavaScript to reveal a destination or a search box a crawler cannot type into [1]. A store whose "related products" section is rendered entirely by a script that never resolves to real anchor tags has effectively none of the cross-links it appears to have, and neither a crawler nor most AI retrieval systems will find them.
| Link | Usually present | Usually missing |
|---|---|---|
| Home to category | Yes | — |
| Category to product | Yes | — |
| Product to its own category | Sometimes | A clear breadcrumb link back up |
| Product to related products | Rarely | Real, crawlable anchor links |
| Blog post to the products it discusses | Rarely | Almost always missing entirely |
| Category to a related category | Rarely | Cross-links between adjacent categories |
What actually counts as a useful internal link?
One that a real visitor would follow because it makes sense in context, not one added purely to satisfy a crawler. The clearest test is whether the anchor text describes the destination specifically enough that the link would still make sense as plain text — "the ceramic mugs in this collection" links somewhere useful; "click here" does not, even when it technically points at the same page.
Three categories of link do most of the work on a store. Product-to-product links, where a product page names and links to a genuine alternative or accessory, help both a shopper comparing options and a crawler discovering neighbouring pages. Category-to-category links, connecting adjacent or related categories, help a shopper who landed one step away from what they actually wanted. And content-to-product links, from a blog post or buying guide into the specific products it discusses, are usually the single biggest gap on a store, because content and catalogue are often built and maintained by different people who never think to connect the two.
Audit what currently links to what
Crawl the store, or check a handful of product pages by hand, and note what each one actually links to beyond its own category.
Find the true orphans first
Pages with no incoming link except from their own category page are the highest-priority fix, because they have the least chance of being found any other way.
Add product-to-product links where a real relationship exists
A genuine alternative, a complementary item, or the next size up — not a random cross-sell algorithm with no relevance.
Connect content to catalogue explicitly
Every buying guide and blog post should link into the two or three products or categories it most directly supports.
Should every filtered or sorted view get linked?
No, and this is where over-eager internal linking causes its own problem. A filtered view represents a real, distinct page worth linking to only when people actually search for that specific combination — a brand within a category, a common size, a specific material. A filtered view for an arbitrary combination nobody searches for should not receive links from elsewhere on the site, because doing so tells a crawler the page matters when it does not.
The same discipline applies to consolidating near-duplicate pages. Google's guidance on duplicate URLs asks you to pick the version you want treated as canonical and make every signal, including your internal links, agree with that choice [2]. A canonical tag pointing at the clean URL while your own internal links point at a filtered variant is a contradiction you are creating yourself, and it undermines the very consolidation you are trying to achieve.
How does this interact with crawl budget on a large catalogue?
Directly, and more so the larger the catalogue gets. On a store with tens of thousands of URLs, internal links function as a signal of priority: pages that receive more internal links, from more places on the site, tend to get crawled more often and treated as more important. A large sitemap alone does not solve this, because a sitemap lists what exists without expressing which pages matter most — Google's own documentation on large sitemaps covers how the file structure scales, but it is silent on priority, because priority is a signal your links create, not your sitemap [3].
Does this matter for platform migrations too?
Yes, and it is one of the most commonly dropped pieces of a move. Migrating Salla to Shopify covers URL mapping and redirects in detail, but redirects alone do not recreate the internal link graph a store built up over years. A new platform's default templates give you the same top-down navigation links every store starts with; the cross-links have to be rebuilt deliberately, the same way they had to be built the first time.
Does internal linking affect how AI assistants read a store?
It affects what they can find at all. An assistant retrieving pages to answer a shopping question follows the same crawlable links a search engine does, and a product with no path into it from a page that actually discusses buying decisions is invisible to that retrieval the same way it is invisible to a search crawler. A buying guide that links into the products it recommends is doing double duty: it is a page a person can read, and it is the path that makes the products it names reachable at all.
If images are part of what makes those linked pages worth landing on, image SEO for Arabic stores covers the other half of what makes a category or product page genuinely useful once someone, or something, follows the link into it.
How do you keep this from decaying again?
Assign it an owner and a schedule, the same way a catalogue's pricing or stock levels have an owner. Internal linking decays quietly: a product gets discontinued and the pages that linked to it are never updated, a blog post gets edited and loses the links it used to carry, a new category launches and nothing points into it for months. None of these show up as errors anywhere, which is exactly why they accumulate.
A quarterly pass is usually enough for most stores. Pull a list of pages with the fewest incoming internal links, check whether any of them deserve more, and add two or three links each time a new product or article goes live rather than treating linking as a separate project to revisit later. The stores with the healthiest link graphs are not the ones that did one big linking project; they are the ones that made adding two or three relevant links part of publishing anything new.
It is worth writing the rule down somewhere your team can see it, because "link to related products where relevant" means something different to every person who has not been told what relevant means on your specific catalogue. A short, concrete guideline — link to the parent category, one genuine alternative, and any content that discusses the product — removes the guesswork and keeps new pages from launching as orphans in the first place.
The stores that struggle to rank a large catalogue almost never have a content problem. They have a hundred good pages with no way in except a category filter, and fixing that costs an afternoon, not a rewrite.
This is one of the cheapest structural fixes available on any store, and it compounds every time a new product or article is added, because a linking discipline applied once keeps paying off on everything published after it. Our ecommerce SEO service treats it as one of the first things worth auditing on any catalogue.
It is also one of the first things how to audit a Salla store's SEO checks for, since an orphaned page is invisible to a crawler no matter how good the page itself is.
Questions people also ask
What is wrong with a store that only has navigation links?
Nothing links a product to anything except its own category and the navigation above it. A crawler and a shopper both hit a dead end at every product page, and pages with no incoming links from content are the hardest pages on the site to discover and to rank.
How many internal links does a product page need?
There is no fixed number, but three to five relevant links — to the parent category, to a related product or two, and to a buying guide that discusses it — is a reasonable floor for most catalogues.
Should filters and sort options be linked internally?
Only the ones that represent something people actually search for. A filtered view nobody searches for should not be a page that other pages link into; a filtered view that is a genuine, searched-for combination deserves its own link and its own content.
Can internal linking fix a crawl budget problem?
It can help by directing crawlers toward the pages that matter, but it cannot substitute for cutting the number of low-value URLs a store generates in the first place. The two fixes work together, not as alternatives.
Is internal linking still worth doing if the store is small?
Yes, and it costs less time on a small catalogue. A store with a hundred products can add meaningful cross-links in an afternoon, and the smaller the catalogue, the more each individual link matters to the pages either side of it.
Sources
- Links: crawlable link requirements — Google Search Central (checked 22 September 2026)
- Consolidate duplicate URLs with canonicals — Google Search Central (checked 22 September 2026)
- Large sitemaps: managing multiple sitemap files — Google Search Central (checked 22 September 2026)
Find the orphaned pages in your own catalogue
Book a free 30-minute consultation with Ibrahem and Jeremy. We look at your site while we talk, and you leave with a prioritised plan.
Who wrote this
Ibrahem HelalCo-founder
Leads strategy and delivery.
Works on: SEO, GEO
Reviewed before publishing by Jeremy Zava, Co-founder.
- Salla SEO: the complete guideHow to rank a Salla store: what the platform controls, what you control, Arabic product pages, duplicate theme URLs, and the order to fix things in.
- Salla SEO apps worth installingWhich categories of Salla apps actually help SEO, which just add scripts that slow the store, and how to tell the difference before you install one.
- Migrating Salla to Shopify without losing SEOHow to move a Salla store to Shopify without losing rankings: URL mapping, redirects, what to keep unchanged for a month, and when the move isn't worth it.
- Arabic image SEO for online storesHow to write Arabic alt text and file names that actually get found, and why translating an English image SEO checklist into Arabic rarely works.