In this article
Point your internal links at the pages you actually want to rank, using anchor text that describes what sits on the other end. A solid internal linking strategy is mostly that one habit, applied consistently across hundreds of URLs instead of a handful. The rest is plumbing — crawl depth, orphan pages, redirect chains, and knowing which of your existing pages has authority to lend. I've rebuilt link structures on 200-page blogs and on catalogues with 400,000 SKUs. The mechanics scale better than you'd expect.
What Makes an Internal Linking Strategy Actually Work?
An internal linking strategy works when it does three jobs at once: crawlers can reach every important URL within about three clicks of the homepage, authority flows from your strongest pages into your priority pages, and descriptive anchor text tells search engines what each destination covers. Miss any one of those and the structure underperforms.
Most sites fail on the second job. They have plenty of links, but the links go everywhere — nav menus, footers, sidebars, "related posts" widgets — and nothing is prioritised. A page with 180 outbound internal links spreads whatever equity it has very thin.
Think in terms of intent instead. Which twelve URLs make you money? Which twenty pieces of content earn the most referring domains? Now connect the second group to the first, deliberately, in body copy.
Google's own Search Central documentation is blunt about this: links help discovery, and anchor text helps Google understand context. That's not a growth hack. It's the crawl model the whole system runs on.
One more thing that separates working structures from decorative ones — placement. A contextual link inside the third paragraph of a well-read guide behaves differently from the same link buried in a footer that appears on 40,000 templates. Editorial links inside content carry intent signals. Boilerplate links mostly carry noise.
Map Your Site Architecture Before You Add a Single Link
Never start linking blind. Crawl first.
Run Screaming Frog (or Sitebulb, if you prefer the visualisations) against your full domain, then open the Crawl Depth report. You want the distribution: how many URLs sit at depth 1, 2, 3, and — the ugly part — depth 6 and beyond. On one B2B site I audited, 34% of blog URLs sat at depth 5 or deeper because the only path to them was paginated archive pages. Those posts were effectively invisible.
Set a rule you can enforce: every commercially important URL within three clicks of the homepage. Everything else within four.
Next, look at your hierarchy honestly. Does your URL structure match how buyers actually think about your products? Category pages should sit above subcategories, which sit above detail pages, and links should travel both directions along that spine. Flat isn't automatically better — a totally flat 50,000-page site gives Google no hierarchy signals at all.
Then export your top pages by referring domains from Ahrefs or Semrush. That list is your budget. It's the authority you have to distribute, and pretending otherwise wastes months.
Pair the crawl data with a visualisation of link equity flow. Sitebulb's internal PageRank calculation is imperfect but directionally useful. When your "About" page outranks your top service page in internal authority, you've found your first fix.
Build Topic Clusters Around the Pages That Earn Revenue
Clusters are the practical shape of a good link structure. One pillar page covers a broad topic. Eight to fifteen supporting articles cover subtopics in depth. Every supporting article links up to the pillar, the pillar links down to each supporter, and siblings link across where the reader genuinely benefits.
Here's a real example. A payments client had a "chargeback prevention" guide and eleven scattered posts on fraud rules, representment deadlines, and issuer codes. None of them linked to the guide. We added contextual links from all eleven, plus a curated block of nine links from the guide down. No new content. Within seven weeks the guide moved from position 14 to position 4 for its head term.
Why does this work so reliably? Because clustering does two things simultaneously. It concentrates internal authority on one URL, and it demonstrates topical coverage around that URL.
Some ground rules I'd defend:
- One pillar per topic. Two competing pillars split your links and cannibalise your rankings.
- Link from strength to need. Your best-linked post is a donor, not a destination.
- Cap sibling links. Three to five in-cluster links per article is plenty.
- Update old posts when you publish new ones. Publishing without retro-linking is how orphans are born.
Clusters also give your team a repeatable brief. Map them into your wider SEO strategy for 2026 so content and links get planned together instead of bolted on afterwards.
Write Anchor Text People and Crawlers Both Understand
Anchor text is the cheapest relevance signal you own, and most teams squander it. "Click here", "read more", "this post" — all wasted. Describe the destination in three to seven words instead.
Vary it, though. If forty pages all link to your pricing page with the identical phrase "enterprise SEO pricing", the pattern looks manufactured. Natural variation reads better and covers more query space: "how our enterprise pricing works", "compare plan tiers", "what enterprise SEO costs".
A gotcha nobody warns you about: when a page links to the same URL twice, testing has repeatedly suggested the first anchor is the one that counts. So if your header nav links to /services with the word "Services", and your body copy later links to /services with "technical SEO audits for SaaS", the descriptive anchor may be ignored entirely. Fix it by keeping nav anchors generic and unique, or by ensuring the contextual link comes first in the DOM.
Keep the surrounding sentence relevant too. Google reads the words around the link, not only the link itself. A link to your keyword tool inside a paragraph about keyword mapping is worth more than the same link dropped into an unrelated closing line.
Last point: never link on a word that describes something else. Anchoring "content strategy" to a page about backlink audits confuses both readers and crawlers. Match the promise to the page, every time — the same discipline you'd apply on any on-page SEO checklist.
How Many Internal Links Should a Page Have?
There's no fixed limit. Google dropped its old "keep links under 100" guidance years ago, and John Mueller has said publicly there's no hard cap. For practical purposes, aim for 3–10 contextual body links on a standard 1,500-word article, and keep total internal links per page — including navigation — under roughly 150.
The real constraint is dilution, not penalties. Every link you add divides that page's outgoing equity. A 2,000-word guide with 45 body links gives each destination a sliver.
Scale with length. One contextual link per 200–300 words feels natural to readers and rarely triggers the "link-stuffed" reading experience. Go much denser and users start ignoring links altogether — which quietly damages the click data those links exist to generate.
Inbound links matter more than outbound counts. When I audit a priority page, the number I care about is how many unique internal URLs point to it. Under five? It's under-supported. Your homepage might have 4,000 inbound internal links while your highest-margin service page has three.
Watch template bloat as well. A mega-menu with 90 links appearing sitewide means every single page spends most of its equity on navigation before body copy gets a look in. Trim the menu to the categories that matter, and let contextual links do the heavy lifting where they're actually relevant.
Hunt Down Orphan Pages and Crawl Dead Ends
Orphan pages have zero internal links pointing at them. Search engines may still find them through sitemaps or external links, but they inherit almost no authority — and they routinely include high-value URLs nobody noticed had been stranded.
Finding them takes ten minutes. In Screaming Frog, connect the Google Search Console and Google Analytics 4 APIs, tick "Crawl Linked XML Sitemaps" in configuration, then run the crawl and open Reports → Orphan Pages. Any URL that shows impressions or sessions but appears in no crawl path is an orphan.
Server log analysis catches a second tier of problems. If Googlebot is spending 60% of its requests on faceted filter URLs and parameterised sorts, your crawl budget is being burned on pages you'd never want ranked. That's a linking problem as much as a robots problem.
Dead ends deserve attention too. Common culprits:
- Deep pagination with no path back into category hubs
- Old campaign landing pages, live but unlinked
- Blog tag archives that link out but receive nothing
- Product variants dropped from a category feed after a stock update
One retailer I worked with had 1,100 discontinued-then-restocked products sitting orphaned because their CMS removed links on stock-out and never restored them. Reconnecting those URLs recovered organic revenue within a quarter — no new content required. Catching that pattern early needs software that monitors link structure continuously, so if you're shopping around, start with these criteria for picking an SEO analytics platform that scales.
A Repeatable Internal Link Audit Workflow
Strategy dies without process. Here's the quarterly workflow I run, and it takes roughly a day for a mid-sized site.
- Step 1 — Crawl. Full Screaming Frog crawl with GSC and GA4 connected. Export Inlinks, Outlinks, Crawl Depth, Orphan Pages.
- Step 2 — Rank your priorities. List the 20–30 URLs that drive revenue or pipeline. Note current inbound internal link counts against each.
- Step 3 — Find donors. Sort pages by referring domains and organic clicks. These are your link sources.
- Step 4 — Match. For each priority URL, find five to ten topically relevant donors. Use a
site:search on a target phrase to surface pages already discussing the topic. - Step 5 — Fix redirect chains. Update internal links to point at final destination URLs, not 301s. Every hop leaks a little and wastes crawl requests.
- Step 6 — Implement and log. Record every link added in a spreadsheet with date, source, destination and anchor.
That log is the unglamorous part, and it's the part that lets you prove causation eight weeks later.
Automation helps with discovery. Ahrefs Site Audit surfaces internal link opportunities, Semrush flags orphans, and Link Whisper suggests placements inside WordPress. My honest position: let tools suggest, never let them auto-insert. Bulk-inserted links produce laughable anchor text and awkward sentences. Human review takes seconds per link and prevents embarrassment.
Measuring Whether Any of It Worked
Internal linking is measurable, which makes it one of the easier SEO wins to defend in a budget meeting.
Start with Search Console's Links report. It shows your top internally linked pages — and if your priority URLs aren't near the top, the structure disagrees with your business. Track that list monthly.
Then watch Crawl Stats in Search Console. After a structural fix, you should see Googlebot requests shift toward the sections you reconnected, and average discovery time for new URLs drop. On a news publisher, reworking category hub links cut time-to-index for new articles from about nine hours to under two.
For ranking impact, isolate it. Pick 15 priority URLs, add links to only ten, leave five untouched as a control, and change nothing else on any of them. Compare average position and impressions over the following six to eight weeks. Rough rule from experience: meaningful movement shows up in three to eight weeks, faster on frequently crawled sites.
Don't ignore engagement data either. GA4 event tracking on internal link clicks tells you which placements readers actually use. Links nobody clicks still pass equity, but a link with a 12% click rate is doing double duty.
Build these views into your reporting rhythm so link work stays visible — this breakdown of what belongs in an SEO reporting dashboard covers the metrics worth pinning.
Mistakes That Quietly Wreck Good Structures
Even careful teams repeat the same handful of errors. These are the ones I flag in nearly every audit.
Nofollowing internal links to "sculpt" PageRank. This stopped working in 2009 when Google changed how nofollow equity is handled. The value evaporates rather than redistributing. Remove those attributes.
Linking everything to the homepage. Your homepage almost never needs help. Send links where they're scarce.
Leaving internal links pointing at redirects. After a migration, thousands of body links often still hit old URLs. Cheap fix, real gain.
JavaScript-only links. Links built with onclick handlers or non-anchor elements may not be followed. Use real <a href> markup, always.
Identical "related posts" blocks everywhere. Automated three-link widgets pulled by tag rarely produce topical relevance. Curate the important ones manually.
And the mistake nobody admits to: linking new content into old posts, then never revisiting the old posts again. Internal linking is maintenance, not a launch task. Schedule it quarterly, and pair it with the broader tactics for improving search rankings.
Where to Start This Week
Pick your five most valuable pages. Find ten relevant existing posts for each. Add contextual links with descriptive anchors, fix any redirect chains you hit along the way, then log everything with today's date.
That's fifty links and maybe three hours of work. It'll outperform most content you publish this month, because the pages you're strengthening already have traffic, conversions and history behind them.
Then repeat next quarter. Compounding beats cleverness here.
Frequently Asked Questions
How long does it take for internal linking changes to affect rankings?
Expect three to eight weeks for measurable movement. Google must recrawl the linking pages, process the new anchor signals, and reassess the destination. Large, frequently crawled sites see shifts inside two weeks; smaller sites with slower crawl rates can take two months. Submit updated source pages in Search Console to speed recrawl slightly.
Should I use nofollow on internal links to control crawl budget?
No. Since 2009, nofollowed internal links waste equity rather than redirecting it elsewhere, and since 2019 Google treats nofollow as a hint anyway. To manage crawl budget, use robots.txt disallow rules for parameterised or faceted URLs, canonical tags for duplicates, and simply remove links to pages that don't deserve discovery.
Do navigation and footer links count as internal links?
They count, but they carry less weight than contextual links inside body copy. Boilerplate links repeated across every template give search engines little information about relative importance. Keep navigation focused on genuine categories, trim footer link dumps, and rely on editorial in-content links to signal which specific pages matter most.
How many internal links should point to one important page?
For a priority commercial page, target at least 10 to 20 inbound internal links from topically related URLs. Compare against your homepage count for context. There's no upper penalty, but relevance beats volume — ten links from genuinely related articles outperform fifty from unrelated posts with generic anchor text.