Short answer: Internal links matter far more than breadcrumbs. Links are how crawlers discover pages and how both people and machines learn which pages relate, so every site with more than a handful of pages should link its pages together using real anchor tags and descriptive text.
Breadcrumbs are a small extra that makes sense only for pages deep in a genuine hierarchy, and the markup makes a page eligible for a breadcrumb display, not guaranteed to get one.
What do breadcrumbs and internal links each do?
Two things tell a crawler how a site fits together. The first is the links between pages. The second, optional one is a breadcrumb trail: a visible path such as Shop, then Shoes, then Runner, with matching structured data. They are not equals. Links carry most of the weight.
Google can only follow a link if it is an anchor element with an href attribute. That sentence is the whole technical foundation of internal linking. Everything else is about choosing which pages to link, and what the link text says. Descriptive anchor text such as "pricing for small teams" tells a reader and a crawler what is on the other side. "Click here" tells them nothing.
| Internal links | Breadcrumbs | |
|---|---|---|
| Needed on | Every site with more than a handful of pages | Pages deep in a real hierarchy only |
| Main job | Discovery, and signalling how pages relate | Showing the path to a page, plus optional markup |
| Can be skipped when | The site is a five-page brochure | It is the homepage, or the site is flat |
| What you can promise | Pages become easier to find | Eligibility only, never display |
When should you skip breadcrumbs entirely?
Skip them on the homepage. It is the root, and a one-item trail pointing at itself is noise. Skip them on a flat site where every page sits at the top level, because there is no hierarchy to show. Do not invent a hierarchy just to satisfy a checklist item.
Breadcrumbs earn their place on pages like a product inside a category inside a shop, an article inside a section, or a documentation page inside a guide. There the trail tells a visitor where they are and gives them a way up a level.
Do not promise a display
Google documents the breadcrumb feature as available on desktop. Marking a trail up makes the page eligible, it does not guarantee that any result will show it. If someone tells you breadcrumb markup will produce a visible change in mobile results, they are promising more than the documentation does.
What does a passing audit tell you about linking?
Less than it seems to. In the checks our toolkit uses, the breadcrumb rule passes if any audited page carries a BreadcrumbList. It does not compare the markup to the visible trail, check the position numbers, or test that the URLs resolve. One page with a breadcrumb passes the whole site.
The internal-link rules count every anchor that points to the same host, excluding bare fragment links. That includes navigation, header, footer and sidebar links. A site whose pages share a ten-link menu can pass both link rules with zero contextual links in the body copy. A pass says the site is not a link desert. It does not say the linking is good. There is also no orphan-page detection in these checks, so finding orphans takes a manual step.
Google's breadcrumb documentation says a BreadcrumbList must define at least two ListItem entries, each with a position, a name and, except on the last item, a URL. You do not need an entry for the homepage or for the page itself. (Google Search Central: Breadcrumb structured data)
For the wider point that audit findings need checking before you act on them, see the article on audit false positives.
Why audit findings need checkingHow do you find and handle orphan pages?
On a small site, take the URLs from your sitemap and, for each one, check whether any other page links to it. A URL that is in the sitemap but linked from no other page is an orphan. On a larger site, use a crawler that reports inbound links per URL and filter for zero or one.
What you do next depends on why the page is an orphan. Not every orphan is a mistake.
| The orphan is | What to do |
|---|---|
| A page that should exist and rank | Link it from relevant pages |
| A near-duplicate of another page | Do not link it. Consolidate the two pages instead |
| Old, replaced content | Redirect or remove it deliberately, with the owner's say-so |
| Intentionally unlinked, such as an ad landing page or a thank-you page | Leave it alone. This is not a problem |
Linking a duplicate to save it is the common mistake. If two pages answer the same question, linking both makes them compete with each other. Decide which one stays first.
How do you improve internal linking and add breadcrumbs, in order?
Use real anchors
Write links as anchor elements with an href. Links built from click handlers, span elements or a javascript: address are not reliably crawled. Keep them in the HTML the server sends.
Write descriptive anchor text
Describe the destination in a few words. For an image link, the image's alt text becomes the anchor text, so give it one.
Add contextual links where a reader would want them
Link from a general guide to the detailed page, from a comparison to each product page, and from old posts to a newer, better one. A few relevant links per substantial page is a reasonable range. That is a guideline, not a measured threshold. Never add links for volume.
Link up and across the hierarchy
Child to parent, sibling to sibling, hub to children. If many pages cover one subject, one page should act as the hub, linking to each and linked back from each. Build a hub only where the content genuinely clusters.
Link every new page the day it ships
Put new pages in at least one existing, crawled page, such as a hub or the blog index. Do not rely on the sitemap alone.
Add breadcrumbs only where there is a hierarchy
Render a visible trail, add BreadcrumbList markup that mirrors it, and generate both from one data source so they cannot drift apart. Follow the typical path a visitor takes, which is not always the URL path.
What does valid breadcrumb markup look like?
This trail has three levels. The last item has no URL because Google uses the page's own address. The visible breadcrumb on the page must show the same three labels.
{
"@context": "https://schema.org",
"@type": "BreadcrumbList",
"itemListElement": [
{ "@type": "ListItem", "position": 1, "name": "Shop", "item": "https://example.com/shop" },
{ "@type": "ListItem", "position": 2, "name": "Shoes", "item": "https://example.com/shop/shoes" },
{ "@type": "ListItem", "position": 3, "name": "Runner" }
]
}The rule that matters most is that markup describes what is visible. If you cannot say what the visible trail is, do not add markup for it. Structured data that does not match the page is a liability, not a signal.
Where is the full implementation guide?
This article explains what breadcrumbs and internal links do and when to bother. The full procedure, including the orphan-detection commands, per-framework notes and the verification checklist, is in the documentation.
Read the site structure implementation guideThe same guide is an agent skill, so a coding agent can make the changes in your repository. The skill source is public.
View the skill source on GitHubThe skills can be installed into Claude Code and other coding agents. Install details are in the skills index.
See how to install the AEO skillsCommon questions
Do breadcrumbs help my site appear in AI answers?
There is no verified evidence that breadcrumb markup changes whether an AI assistant cites a page. Google documents breadcrumbs as an eligibility feature available on desktop. Their dependable value is helping visitors and crawlers understand where a page sits in a real hierarchy.
How many internal links should a page have?
There is no measured number. A few relevant contextual links per substantial page is a reasonable range, and link where a reader would genuinely want to go next. Hundreds of footer links add count and no meaning.
Is an orphan page always a problem?
No. An ad landing page or thank-you page is intentionally unlinked and should be left alone. An orphan is a problem when it is a page you want found and ranked, in which case link it from relevant pages. If it duplicates another page, consolidate instead of linking.
Should my breadcrumb follow the URL path?
Not necessarily. Google recommends a trail that represents a typical path a user takes to the page, not a mirror of the URL structure. If a product lives in two categories, pick the usual path or mark up more than one trail, and make sure it matches the visible breadcrumb.
Sources
Related guides
Want it done for you
Want this done and re-checked every month? Our monthly maintenance re-runs your buying questions in ChatGPT, Perplexity and Google, ships the fixes that keep you out of the answers, and reports what moved. Or run the free check on your site first at advancelabs.dev/tools/audit.
See how it works