This is a walkthrough of an internal linking experiment on a mid-sized content site, testing whether concentrating internal link equity on priority pages — by removing low-value boilerplate links and deliberately reallocating that link budget — would measurably improve rankings on the pages that mattered most.
What prompted the experiment
An internal linking audit found that the site's per-page link count was dominated by two low-value sources: a large footer link block repeated identically on every page (around 35 links), and a sidebar "related articles" widget that surfaced content by recency rather than topical relevance, often linking to entirely unrelated pages. Combined, these two sources accounted for the large majority of each page's total internal link count — meaning genuine, contextually relevant internal links were a small fraction of the total, diluted by boilerplate.
The hypothesis: this boilerplate was spreading link equity thin and indiscriminately rather than concentrating it where it would help most, and reallocating that link budget toward deliberate, contextually relevant links to a defined set of priority pages would measurably help those pages.
What actually changed
Footer link block reduced from roughly 35 links to 8, keeping only genuinely important sitewide destinations (not every category and subcategory that existed).
Sidebar "related articles" widget replaced with a smaller, manually curated set of 3-4 genuinely related articles per page, rather than an automated recency-based feed.
Deliberate contextual links added to a defined list of roughly 40 priority pages — inbound links placed within the body content of genuinely related existing articles, following the same publish-time linking discipline described in our internal linking system guide.
Net effect: total internal link count across the site dropped by roughly 40%, while the number of contextually relevant, deliberately placed links pointing specifically at the 40 priority pages actually increased.
Results
Of the 40 priority pages, roughly 60% showed a measurable ranking improvement for their target keywords within 6-8 weeks of the change, with the clearest gains on pages that moved from having zero genuine contextual inbound links (only boilerplate) to having 3-5 newly added ones. Pages outside the priority list — which had links removed (less boilerplate reaching them) but received no compensating new links — showed no meaningful change in either direction, neither improving nor declining.
This pattern is the core finding: the change that correlated with improvement was the addition of deliberate, relevant links to priority pages, not the removal of boilerplate on its own. Pages that only experienced removal, without redistribution, were essentially a null result.
What this doesn't show
This experiment doesn't demonstrate that internal links in general are harmful, or that sites should indiscriminately cut internal linking to improve rankings — that would misread the result. What it shows is that a large volume of low-value, non-contextual internal links (footer boilerplate, unrelated sidebar widgets) can crowd out and dilute the specific, contextually relevant linking that actually concentrates equity where it's useful. The fix wasn't fewer links — it was fewer low-value links and more high-value ones, changing the composition of the site's internal link graph rather than simply shrinking it.
Takeaway for other sites
Before assuming a site needs more internal links overall, it's worth auditing what proportion of existing links are genuinely contextual and relevant versus repeated boilerplate. A site with a large link count dominated by footer and sidebar boilerplate may benefit more from reallocating that link budget toward deliberate, priority-page-focused linking than from adding still more links on top of an already-diluted structure.
