Improving a large site’s rankings with the URLs it already has sounds simple. It isn’t. On an enterprise site, internal linking means stakeholders, a business case and template-level rules before a single link changes.
Key takeaways
- Map every URL by page type before you add a single link.
- Let GA4 and Search Console data choose which pages give and get links.
- Fix links at the template level: breadcrumbs, pagination and modules.
The payoff justifies the work. Internal links spread link equity, the ranking value links pass from page to page, across the site. They also clarify its structure and send visitors to the most relevant pages.
I use two retailers with huge catalogs as running examples: Wayfair and The Home Depot. Every section follows the same shape: what to do, the exact steps and what it looks like on their live sites.
Stakeholder Buy-In for an Enterprise Internal Linking Project
Pitch internal linking as the fix to a measured problem, with a cost attached. Pitched as SEO best practice, it gets a nod and a spot at the bottom of the backlog.
- Size the problem. Count orphan pages and pages with Search Console impressions but few clicks.
- Tie it to sales. Show which of those pages are categories or products that sell.
- Propose a template change. A related-links module on one template reaches thousands of pages at once.
- Scope the effort. Estimate engineering time and offer a test on one site section first.
- Agree on the metric. Settle what success looks like before any work starts.
Website Architecture and URL Structure by Page Type
Website architecture, URL structure and taxonomy all describe the same thing: how a site is organized and connected.
I explain it with a house. Pages are rooms and links are the hallways between them. The URL is each room’s address, the master bedroom, the office, the game room, so you always know where to find it.
Or think of the files on your computer. Files sit in folders by type or purpose, and the file path tells the computer where each one lives. A URL does the same job for a page. Either way, the test is the same: a visitor and a search engine should both find any page and understand where it sits.
URL Pattern Mapping From a Full Site Crawl
Identify every content type from the URL alone, whether the site is a blog, a publisher, an affiliate site or a store.
- Crawl the whole site. Use Screaming Frog or Sitebulb, with JavaScript rendering on for sites that need it.
- Export every URL. Include status code, crawl depth and the count of internal links pointing in.
- Split the path into columns. Department, category, subcategory and product each get a column.
- Label each page type. Write one rule per pattern, such as /cat/ means a category page.
A clear structure for a shoe store reads like this:
- /shoes/womens/boots/leather/
- /shoes/womens/high-heels/
- /shoes/womens/sneakers/
- /shoes/mens/loafers/
- /shoes/mens/sneakers/
Each URL tells visitors where they are and gives search engines a clear path. Once every page type is listed, related categories, and the links between them, are much easier to place.
Wayfair's Category URL Pattern as a Linking Blueprint
Wayfair, the home goods retailer, sells hundreds of thousands of products. Its main navigation lists the core departments: Furniture, Outdoor, Bedding & Bath, Mattresses, Rugs, Decor & Pillows, Lighting, Organization, Kitchen, Baby & Kids, Home Improvement, Appliances and Pet.
Every department link follows one pattern: the department, a /cat/ segment, then the category name with an internal ID, such as /furniture/cat/furniture-c45974.html. The homepage source carries 278 of these /cat/ links as plain HTML.
One predictable URL pattern lets a team write linking rules by page type instead of page by page.
Link Sources and Link Targets From Analytics and Search Console Data
Let traffic data decide which pages give links and which pages get them. I treat Google Analytics 4 as a map of the site: it shows which pages get traffic and which ones don’t.
Take the same shoe store. The men’s running shoes page gets heavy traffic, and the women’s hiking boots page gets very little. Those pages need opposite treatment. Running shoes has visibility to share, so I link from it to men’s athletic wear and running accessories. Hiking boots needs links pointing in, from women’s outdoor gear and hiking accessories.
| Data source | What it tells me | How I use it |
|---|---|---|
| Google Analytics 4 | Which pages earn sessions and conversions | Strong pages become link sources |
| Google Search Console | Pages ranking in positions 8 to 20 | Near-page-one pages become link targets |
| Site crawl | Internal links pointing to each URL | Under-linked pages move up the list |
- Export landing pages. GA4: Reports, Engagement, Landing page, with sessions and key events.
- Export striking-distance pages. Search Console: Performance, Pages tab, filtered to average position 8 to 20.
- Join both to the crawl. Match on URL and add the inlink count from the crawl export.
- Pair sources with targets. Link each strong page to related pages in the target list, inside the same department.
Built on that data, the plan helps visitors find what they came for. That shows up in engagement and conversions.
Orphan Page Detection and Fixes
A page with little traffic still belongs in the plan. Every page plays a role in the site’s structure. Leave a page out and it can become an orphan: a page with no internal links pointing to it. Search engines struggle to crawl and index orphans, and visitors can’t reach them.
- Collect three URL lists. The crawl, the XML sitemap and the GA4 landing-page report.
- Find the gaps. Any URL in the sitemap or analytics but missing from the crawl is orphaned.
- Decide per URL. Link it from its parent category, or redirect it if it no longer serves a purpose.
- Repeat after releases. Run the check after every major template or category change.
Pro tip: run the orphan check after every major release. New templates and retired categories create orphans faster than any other change.
Click Depth Reduction on Large Ecommerce Sites
Click depth, also called page depth, is the number of clicks it takes to reach a page from the homepage or main menu. Important pages should be easy to reach without a long chain of clicks.
Think of a grocery store. If the item you want is buried in a back aisle, you might give up before you find it. If it sits near the entrance, you find it right away. A website works the same way: pages buried deep get found less, crawled less and tend to rank lower.
Fix depth with template-level links that appear on every page of a type. The Home Depot uses all three that matter most.
Breadcrumb Trails on Product and Category Pages
Every Home Depot product page carries a breadcrumb trail back up the hierarchy. The Ninja BL610 blender links to Small Kitchen Appliances, Blenders, Countertop Blenders and NINJA Countertop Blenders.
The category page backs the trail with BreadcrumbList structured data in its server HTML, and Google shows the path in the result.
Crawlable Pagination Instead of Load More Buttons
The Countertop Blenders category holds 139 products, 24 per page. Pages 2 to 6 sit behind numbered links, and each one is a plain <a href> link with its own URL, such as ?Nao=24.
That matters because Google’s crawlers don’t click buttons. A load more button without links can hide every product after the first page.
Related-Product Modules as Template-Level Links
The same product page links sideways too. Customers Also Viewed and Other Shoppers Explored These put other blenders one click away. One module on the product template creates those links on every product at once.
- Add breadcrumbs to every template below the homepage. Mark them up with BreadcrumbList structured data.
- Use numbered pagination links. Give every page its own URL and a plain <a href> link.
- Add related modules by rule. Same subcategory, then same brand, then frequently bought together.
- Build hub pages. One page per department that links to every subcategory.
- Recrawl and compare. Check the click depth distribution before and after the change.
Crawl Budget Thresholds for Very Large Sites
Crawl budget is the number of URLs a search engine will crawl on a site in a given period. Google’s guide says it matters mainly for large sites with over 1 million pages that change weekly, or sites with over 10,000 pages that change daily. On those sites, internal links help decide which URLs get crawled first.
Read more: SEO site architecture services
Anchor Text Rules for Enterprise Internal Links
Write anchors that describe the destination, and use links Google can actually follow.
- Describe the destination: "women’s hiking boots", not "click here".
- Vary anchors naturally across templates instead of repeating one phrase sitewide.
- Keep one target page per topic, so two pages don’t compete for the same anchor.
- Use standard <a href> links.
The Home Depot breadcrumbs follow every rule. Each anchor names its category, Blenders, Countertop Blenders, NINJA Countertop Blenders, and each one is a plain link.
Server-Rendered Internal Links for AI Crawlers
Put key internal links in the server HTML, not only in JavaScript. AI crawlers find pages by following links, but many don’t run JavaScript.
Vercel’s analysis of AI crawler traffic found that the ChatGPT and Claude crawlers fetch JavaScript files but don’t execute them. Links that only appear after JavaScript runs are invisible to them.
Wayfair’s 278 category links and Home Depot’s pagination links are already in the server HTML, so any crawler can follow them. Google AI Overviews draw on Google’s regular index, so the links that help rankings also help a page qualify there.
- Compare raw and rendered HTML. Count internal links in view-source and in the rendered DOM for each template.
- Move key links server-side. Navigation, breadcrumbs, pagination and hub links first.
- Check bot access. Confirm robots.txt allows OAI-SearchBot, PerplexityBot and Googlebot.
Read more: Technical SEO services and AI search optimization services
Internal Linking Metrics and Measurement
Measure the project against the metric you agreed on at the start, before and after each template change.
- Click depth distribution before and after, from the crawl.
- Orphan page count.
- Internal links pointing to each target page.
- Search Console impressions, clicks and average position for the target pages.
- Indexed pages in the Search Console Page indexing report.
On a large site, internal links are the one ranking lever you fully control. Too many teams leave them to whatever the templates happen to output.
Planning a linking project on a large site? Get in touch.
