How Many Internal Links Per Post? What Actually Matters

Ask ten SEO blogs how many internal links a post should have and you’ll get ten confident answers: “3 to 5,” “one per 100 words,” “no more than 100 per page.” The awkward truth is that none of these numbers comes from Google, and most of them trace back to advice written for a web that no longer exists. In this guide we’ll debunk the magic-number myth, look at what Google has actually said on the record, walk through the factors that genuinely determine how many links a post can carry, and give you some honest heuristics — clearly labelled as sanity checks, not rules — plus a short worked example you can copy for your own site.

The short answer: there is no magic number

The “3–5 internal links per post” figure gets repeated so often that it feels like an official guideline. It isn’t. We’ve never found a Google source for it, and when you think about it for thirty seconds, a fixed count can’t be right: it would mean a 300-word announcement and a 4,000-word ultimate guide should carry the same number of links. That’s obviously absurd. The long guide references dozens of subtopics you’ve probably written about elsewhere; the short announcement might legitimately need one link or none.

The other zombie number — “keep pages under 100 links” — comes from a real but long-retired Google guideline. In the early 2000s, Googlebot truncated pages at around 100KB, so links beyond that point risked not being crawled. Google removed the recommendation years ago and Matt Cutts publicly walked it back. Modern Google crawls and renders far larger pages without difficulty. Quoting the 100-link rule in 2026 is like planning a road trip around the range of a Model T.

So if there’s no number, what should you actually optimize? Usefulness, relevance, and distribution. Let’s take those in order, starting with the horse’s mouth.

What Google has actually said

Google’s public guidance on internal link counts is deliberately vague, but it’s vague in an instructive way. John Mueller has fielded this question repeatedly in Search Central office hours, and his answers cluster around two ideas:

Use a “reasonable number” of internal links — and make sure each one is genuinely useful to the person reading the page. If a link doesn’t help the reader, it probably isn’t helping your SEO either.

That’s a paraphrase of the recurring theme rather than a single quote, but the pattern across his answers is consistent: Google doesn’t count links against a threshold; it evaluates whether the link structure makes sense. Mueller has also made two more concrete points worth internalizing:

• Internal links are how you tell Google what matters. In a 2022 office-hours session he called internal linking “one of the biggest things you can do on a website” to signal which pages you consider important. Pages that accumulate many internal links from relevant contexts are read as more central to your site.
• Dilution is real at the extremes. Mueller has noted that if you link to essentially every page from every other page, you’ve told Google nothing — the signal flattens out. A page with 400 undifferentiated links passes less meaning through each one than a page with eight carefully chosen links.

Notice what’s missing: any number. Google’s position is essentially “link where it helps, don’t link where it doesn’t, and make your important pages look important.” Everything below is about turning that into something you can act on.

The factors that actually determine your link count

1. Content length

Longer posts naturally support more links because they cover more subtopics, each of which is a legitimate linking opportunity. A 500-word post that crams in ten links feels like a link farm; a 3,000-word guide with ten links spread through it feels sparse. Length doesn’t entitle you to more links — it creates more places where a link is genuinely warranted.

2. Topical relevance

Every internal link is a claim: “these two pages are related.” Google uses those claims to understand your site’s topical structure, which is why links inside a tight content cluster punch above their weight. If you’re deliberately organizing posts into clusters, our guide to building topic clusters and content hubs in WordPress covers the architecture side; the short version is that a link from a closely related post is worth far more than a link shoehorned in from an unrelated one. A tangential link that exists only to hit a quota is worse than no link.

3. Click-through likelihood

Here’s an underrated filter: would a real reader plausibly click this? Links that people actually follow keep visitors on your site, generate engagement signals, and — more mundanely — prove the link was useful. A link to “our complete guide to X” placed at the exact moment the reader is wondering about X gets clicked. The same link buried in a wall of eleven other links in one paragraph gets ignored by readers and, we’d argue, discounted by search engines. If you can’t imagine anyone clicking a link, cut it.

4. Dilution and competition for attention

Each additional link on a page slightly reduces the share of attention — human and algorithmic — flowing through every other link. This is not a reason to be stingy; it’s a reason to be intentional. Twenty useful links dilute nothing worth keeping. Twenty useful links plus thirty decorative ones bury the good twenty. The cost of a weak link isn’t the link itself; it’s what it steals from the strong ones around it.

Practical heuristics by post length

With all the caveats above firmly in place: people still want a starting point, and a sanity-check range is genuinely useful for spotting outliers in a content audit. Our working heuristic is roughly one contextual link per 150–300 words. Treat it like a smoke detector, not a thermostat — it tells you when something might be off, not what temperature to hold.

Two clarifications. First, “contextual” means links inside the body copy — not navigation, sidebars, footers, or related-post widgets, which Google largely treats as boilerplate. Second, being outside these ranges is a prompt to look, not a verdict. A 2,500-word tutorial on a topic you’ve never written about before might legitimately have three links. A 900-word comparison post sitting at the heart of a mature cluster might legitimately have eight.

Distribution matters more than the per-post count

Here’s the reframe that, in our experience, actually moves rankings: stop asking “how many links should this post have going out?” and start asking “which pages should be accumulating links coming in?” The per-post outbound count is a detail. The site-wide inbound distribution is the strategy.

Your pillar pages — the big guides targeting your most valuable keywords — should be the most-linked pages on your site, because inbound internal links are the clearest signal of importance you control. If your cornerstone guide has four internal links pointing at it while a throwaway news post from 2023 has thirty, your link graph is telling Google the opposite of what you intend, and no amount of per-post link counting will fix that. When we audit a site, the first chart we build is “internal links pointing at each URL, sorted descending” — the top of that list should read like your sitemap’s greatest hits. If it doesn’t, that’s the week’s work. Our pillar guide on automating your linking workflow in WordPress goes deeper on how to fix inbound distribution at scale without editing a hundred posts by hand.

Vary your anchors while you’re at it

Count and distribution are two of the three levers; anchor text is the third. If forty posts all link to your pillar page with the identical exact-match anchor, the pattern looks manufactured — because it is. Natural linking produces natural variety: exact phrases sometimes, partial matches often, descriptive sentences occasionally. Mix “keyword research guide,” “our process for finding keywords,” and “how we research keywords” rather than repeating one phrase forty times. We’ve written a full breakdown of anchor text patterns that work in WordPress if you want the details, including the ratios we aim for.

When too many links genuinely hurts

We’ve spent most of this post arguing that link-count anxiety is overblown — so let’s be fair to the other side. There are real failure modes where “too many links” is an accurate diagnosis:

• Nav-stuffed footers and sidebars. Sites that cram 150 links into a sitewide footer — every category, every tag, every city page — dilute their link graph on every single page load. Google mostly discounts boilerplate links, so you gain little and add crawl noise.
• Exact-match anchor spam. Auto-linking every occurrence of a keyword, on every page, with the same anchor, pointing at the same target. This is the pattern that made early auto-linkers infamous, and it reads as manipulation to both users and algorithms.
• Link clumping. Six links in the first paragraph and none for the next 1,500 words. Readers hit a wall of blue text and bounce; the links compete with each other instead of supporting the content.
• Linking the same target repeatedly in one post. One good link to a page per post is generally enough. The third link to the same URL adds nothing and looks desperate.

Notice that every one of these is a quality or pattern problem, not a count problem. The fix is never “get under some number” — it’s “remove the links that serve no reader.”

This is, candidly, the failure mode that made us build tooling around it. Full disclosure: our own free plugin, JnK Linkweave, exists largely because manual discipline doesn’t scale — it applies per-keyword link rules with per-page frequency caps, so a keyword links once (or however many times you allow) per page instead of every occurrence, which prevents exactly the auto-linker spam pattern above. Weigh our recommendation accordingly, but the caps-not-counts philosophy is the same one this whole article argues for.

A short worked example

Say you run a food blog and you’ve just finished a 1,800-word post: “How to Proof Sourdough in a Cold Kitchen.” Applying everything above:

• Sanity check: 1,800 words at one link per 150–300 words suggests roughly 6–10 contextual links. That’s the range we’ll aim within, not a quota to fill.
• Pillar first: The post’s most important link goes to your cornerstone “Complete Sourdough Guide” — placed early, where you first mention starters, with a descriptive partial-match anchor like “our full sourdough walkthrough.”
• Cluster siblings: Four or five links to closely related posts, each at the moment the subtopic comes up: the starter-feeding schedule post where you mention feeding, the banneton review where you mention proofing baskets, the oven-spring troubleshooting post in the section on baking day.
• Click test: You were about to link your “10 Best Stand Mixers” roundup because it’s a money page. But nothing in a cold-proofing post makes a reader wonder about stand mixers — nobody would click it. Cut.
• Final tally: Seven contextual links, spread across the post, no repeated targets, varied anchors. You never counted toward a number; you linked where it helped, and the count landed in a reasonable range on its own.

That last line is the whole article in miniature. Get the usefulness, relevance, and distribution right, and the number takes care of itself.

Frequently asked questions

How many internal links should a blog post have?

There is no official number from Google. As a sanity check, roughly one contextual link per 150–300 words of content works well — about 3–6 links for a 1,000-word post and 8–15 for a long guide — but relevance and usefulness to the reader matter far more than hitting a count.

Can too many internal links hurt SEO?

Only in specific patterns: footers stuffed with hundreds of boilerplate links, exact-match anchor spam from aggressive auto-linking, many links clumped into one paragraph, or repeatedly linking the same URL in a single post. A large number of genuinely useful, well-distributed links does not hurt.

Is the 100-links-per-page rule still valid?

No. It came from an early-2000s technical limit where Googlebot truncated pages at about 100KB. Google retired the guideline years ago and modern crawlers handle far larger pages, so the 100-link ceiling is obsolete advice.