Skip to content

Lesson 1 of 4

Site architecture and internal links

How to organise pages and links so authority flows to the pages that make money.

5 min readBeginnerUpdated 2026-08-22

A store selling trail shoes has a page for its best-selling waterproof model. It sits in the sitemap, but the only way to reach it is a filter on the category page that a crawler can't operate. Meanwhile a blog post about lacing techniques is linked from the footer of every page. Google's picture of which pages matter is the reverse of the owner's, and the owner built it that way without noticing.

Where a page sits in your site and what links to it decide how often Google visits it, how well Google understands it, and how much of the site's trust it inherits. Structure is the part of SEO you control completely, and usually the part nobody has designed.

Googlebot reaches a page by following a link to it. The same link does two more jobs: its clickable words (the anchor text) tell Google what the destination is about, and the link passes on a share of whatever trust the linking page has earned. Links from other sites do this too, but internal links are the ones you can add this afternoon.

So the pages with the most internal links pointing at them are the pages Google treats as the centre of your site. On most sites that centre is accidental: every template links to the homepage, the blog index and the privacy policy, while the service page that pays the bills gets one link from a dropdown menu.

Google's documentation on crawlable links is strict about what counts: a link is an <a> tag with an href pointing at a real URL. A button that runs a script, a card that responds only to a click handler, or a menu that renders after a tap may look like a link to you and be nothing to Googlebot.

Keep money pages close to the homepage

Two things fall off with every click away from the homepage: how often Google crawls a page, and how much link signal reaches it. The homepage usually holds most of a site's external links, and each hop inward passes along a smaller share.

You'll read that everything must sit within three clicks. That isn't a Google rule, and Google's own crawl budget guide says crawl budget only becomes a constraint for sites with over a million pages, or over ten thousand pages that change daily. For everyone else the problem is priority, not budget. A page five clicks deep is crawled less often and inherits less, so it ranks behind a shallower page saying the same thing.

The shape that works is a pyramid: the homepage links to a handful of hub pages (service categories, product categories, main topics), each hub links to the pages under it, and every page that earns money sits one click below a hub in the main navigation. A page reachable only through a filter, a search box or the sitemap isn't in the pyramid at all.

A hub is one page that owns a topic and links to everything underneath it: a category page for waterproof trail shoes linking to each model. Each of those pages links back to the hub and sideways to its closest siblings.

This concentrates link signal on the hub, so the hub rather than a random sub-page is what Google shows for the broad term. It also tells Google which page is the main one for the topic, so your own pages stop competing for the same search. Mapping keywords to pages covers which page owns which search; Topic clusters covers building out the spokes.

Breadcrumbs give you the upward links for free: a trail like Home > Trail shoes > Waterproof links every page to its hub and the homepage, and with Breadcrumb structured data, which Google's rich results gallery still supports, the path can show in the result itself.

Write anchor text that says what the page is

The anchor text is the clickable words. "Click here" and "read more" tell Google nothing. "Waterproof trail shoes for wide feet" says exactly what the destination is about, in the words a searcher uses.

Vary it: the same exact phrase on every link reads as mechanical, to readers and to Google. Position matters too. A link inside a paragraph about the destination is the clearest signal you can send, while a footer link repeated on every page says little about any one page.

Three habits waste links. Linking to a URL that redirects, so every click passes through a hop; point at the final URL. Adding nofollow to internal links; Google's 2019 announcement made nofollow a hint rather than a command, and on your own pages it only throws signal away. And leaving orphan pages, listed in the sitemap but linked from nowhere, which Google visits rarely.

What to take away

  • Internal links are how Google finds your pages, learns what they're about from the anchor text, and decides which pages on your site matter most.
  • Money pages belong one click below a hub that sits in the main navigation, reached through real links rather than filters, buttons or the footer.
  • Hubs that link down to their pages, and pages that link back up, concentrate signal on one page per topic instead of splitting it.
  • Descriptive anchor text inside real content is the strongest internal link you can build, and a page nothing links to is a page Google will mostly ignore.

Next

With the structure in place, the next job is the words Google shows for each page, in URLs, titles and meta descriptions.

Free tools this lesson uses

Saved in this browser only.

Chapter 3: Build a site Google can read

All 9 chapters