On 21 May 2026 — yesterday, as this article goes live — Google began rolling out the May 2026 Core Update, its second broad core update of the year. The rollout is expected to take up to two weeks, and ranking volatility is already being reported across every major SEO tracking platform. Within hours of the announcement, the same question hit the inbox of nearly every consultant in this industry: “Should I be sculpting PageRank to protect my money pages?”
The honest, evidence-based answer in 2026 is uncomfortable for the SEO industry: the classic version of PageRank sculpting — slapping rel=”nofollow” on internal links to fence off link equity — has not worked since 2009. Google killed it sixteen years ago. And yet a search for the phrase still returns thousands of blog posts, half of which are recycling advice that was already wrong when the iPhone 4 launched.
This guide does three things. First, it sets out exactly what PageRank sculpting was, why it stopped working, and what Google publicly confirmed. Second, it covers the modern reality — the Reasonable Surfer model, contextual link weighting, and the link-graph signals Google actually uses now. Third, it shows what you should be doing in 2026 to control internal link equity — backed by hard case study data from sites that have done it and grown traffic through every core update Google has shipped since.
| The short answer Classic nofollow-based PageRank sculpting is dead and has been since June 2009. Modern PageRank shaping — through link placement, anchor text, link removal, and architecture — is alive, measurable, and one of the few SEO levers that consistently delivers double-digit traffic lifts during core update rollouts. The May 2026 Core Update has not changed this. It has reinforced it. |
What is PageRank sculpting? A quick definition
PageRank sculpting is the practice of trying to control how link equity (link juice, authority, PageRank — pick your favourite term) flows within a website by selectively blocking certain internal links from passing value. The goal is to concentrate authority on commercially important pages — money pages, category pages, conversion targets — and starve unimportant pages of equity they don’t deserve.
In its original, now-defunct form, the implementation was straightforward:
- Identify low-value internal links — typical targets were login pages, terms-and-conditions, privacy policies, search pages, and cart pages.
- Add rel=”nofollow” to those links.
- Assume that the PageRank that would have leaked to those pages would now be redistributed across the remaining dofollow links.
The premise was elegant. The execution was simple. And the results were imaginary, because Google changed how nofollow handles PageRank in June 2009 and never changed back.
Why classic PageRank sculpting died in 2009
The death certificate was signed publicly at SMX Advanced 2009 by Matt Cutts, then head of Google’s webspam team. The mechanism is best explained with Cutts’ own numbers, which have since been requoted in essentially every reputable SEO resource — including Ahrefs and Search Engine Land.
| How Google changed nofollow in 2009 Before 2009 (the sculpting era): Page has 10 outbound links and 10 units of PageRank. 5 links are nofollowed. Each of the 5 remaining dofollow links inherits 2 units of PageRank. After June 2009 (still the case today): Page has 10 outbound links and 10 units of PageRank. PageRank is divided across all 10 links — including the nofollowed ones. The 5 dofollow links still get 1 unit each. The 5 units allocated to nofollow links simply evaporate. |
Read that twice. Nofollowed PageRank doesn’t get redistributed. It vanishes. You are not protecting authority — you are destroying it. The link is still counted as part of the divisor; it just stops passing the share it was allocated.
This change made the entire mathematical basis of sculpting collapse overnight. As one widely-cited 2026 analysis put it: “You’re destroying value, not redirecting it” (SEOJuice, April 2026). Every major search engineer who has spoken about this since — including Gary Illyes, John Mueller, and Martin Splitt — has confirmed the same model.
Why the myth persists in 2026
Sixteen years after Google killed it, why are SEOs still discussing PageRank sculpting? Three reasons:
- Inertia in the content layer. Blog posts written in 2007–2009 are still indexed and still ranking. Junior SEOs read them, take them at face value, and the bad advice keeps re-entering the conversation.
- Confusion between sculpting and link weighting. Modern Google does weight links differently based on placement, context, and probability of being clicked. This is a real mechanism. It is not the same thing as nofollow sculpting, but it sometimes gets called that.
- The nofollow-as-hint update of 2019. Google reclassified nofollow as a “hint” rather than a directive in March 2020. This made the picture more complex and gave sculpting advocates a thin reed to cling to. It did not bring sculpting back.
| Important nuance Since 1 March 2020, Google treats rel=”nofollow” as a hint, meaning Google reserves the right to crawl and pass equity through nofollowed links anyway. This means even the residual “protection” some SEOs hoped nofollow gave them is not guaranteed. Your nofollow may be ignored entirely. |
What actually controls internal link equity in 2026
If nofollow sculpting is dead, what does work? The honest answer is: a combination of architectural decisions and signals that Google has publicly documented, plus mechanisms inferred from patents. Here is the full modern picture.
1. The Reasonable Surfer model
In May 2010 Google was granted US Patent 7,716,225 — the “Reasonable Surfer” patent — and updated it in 2016. The core idea: not all links on a page carry the same weight. The amount of PageRank a link passes is modulated by the probability that a real user would actually click it.
Factors that the patent and its updates explicitly list as influencing this probability include:
- Link position on the page (above the fold links carry more weight than below)
- Font size, colour, and visual prominence
- Anchor text relevance to the linking page’s topic
- Whether the link is contextual (in-content) or boilerplate (nav, footer, sidebar)
- Position within the surrounding paragraph
- User behaviour data — actual click-through rates on similar links
This is the closest thing to modern sculpting that exists, and it works in the opposite direction from the 2007-era version. You don’t block link equity. You amplify it by making the links you care about more clickable to a real human reader.
2. Link count dilution
PageRank on a page is divided by the total number of outbound links — internal and external, dofollow and nofollow combined. This means the single most powerful sculpting lever is not which links you mark up, but how many links exist in the first place.
| The link-count math A page with 40 outbound links transfers ~1/40th of its PageRank through each link. Remove 20 of those links (e.g., trim global navigation to essentials). Each remaining link now transfers ~1/20th — double the equity, with no nofollow trickery and no Google update risk. |
This is sculpting that actually works in 2026. It is also conservative, fully Google-approved, and survives every core update — including the one rolling out right now.
3. Contextual link placement
Contextual links — links embedded inside the main body content, surrounded by topically relevant text — pass meaningfully more equity than boilerplate links to the same destination. This has been confirmed both by patent analysis and by independent SEO testing for over a decade.
Practical implication: a single contextual link from a 2,000-word article on a related topic will move the needle more than the same link repeated 47 times in your sidebar.
4. Anchor text
Anchor text doesn’t change the volume of PageRank flowing through a link, but it dramatically changes what the destination page ranks for. A keyword-rich anchor pointing to your target page tells Google explicitly what query that page deserves to rank for. This is covered in depth in the upcoming Article #132 on anchor text distribution.
5. Click depth and crawl frequency
Pages closer to the homepage receive more PageRank and get crawled more often. Pages buried four, five, or six clicks deep see exponential equity decay. Reducing maximum click depth from six to three is one of the most reliable sculpting interventions you can make.
How the May 2026 Core Update changes the picture
Google’s official description of the May 2026 Core Update, posted to the Search Status Dashboard on 21 May 2026, reads: “a regular update designed to better surface relevant, satisfying content for searchers from all types of sites.” Sources tracking the rollout — Search Engine Journal, Search Engine Roundtable, and Search Engine Land — all confirm the rollout is global, affects all content types, and is expected to complete by around 4 June 2026.
From everything we have seen in the first 18 hours of volatility, the update has three implications for how you should think about internal link equity:
- Sites with bloated, low-utility internal link counts are losing. Early ranking trackers (Semrush Sensor, Mozcast, Advanced Web Rankings) show notable volatility on sites with deep, link-heavy navigation and orphan-heavy structures. The dilution problem is becoming a ranking problem.
- Topical clusters with tight bidirectional linking are gaining. Sites that have invested in hub-and-spoke architecture — pillar pages with contextual links to and from cluster content — are showing the strongest stability through the rollout’s first day.
- Nofollow sculpting is, if anything, doing more damage than usual. Sites that have aggressively nofollowed internal links to “protect” money pages are seeing those pages drop, not rise. Authority that should be flowing to them is evaporating, not concentrating.
| Practical action for the next 14 days Do not make sudden changes. Google’s own guidance is to wait until at least one full week after the rollout completes before reviewing Search Console data — that means baseline data from the weeks before 21 May, compared with the period after roughly 4 June. Then audit internal linking, not nofollow attributes. |
Sculpting tactics in 2026: what works, what doesn’t
Here is the consolidated state of every sculpting tactic SEOs are currently using, scored against the May 2026 algorithmic landscape.
| Tactic | Status in 2026 | Risk level | Recommended |
| rel=”nofollow” on internal links | Obsolete — destroys equity | Medium | No |
| rel=”nofollow” on external sponsored/UGC | Required (use sponsored/ugc) | High if missing | Yes |
| Reducing global nav link count | Works | Low | Yes |
| Removing footer link bloat | Works | Low | Yes |
| Contextual in-content linking | Works | Low | Yes |
| JavaScript-cloaked links | Risky, easily reversed | High | No |
| robots.txt disallow on low-value pages | Works for crawl budget | Medium | Selective |
| noindex on thin pages | Works for index quality | Low | Yes |
| Iframing low-value content | Obsolete and brittle | High | No |
| Hub-and-spoke architecture | Works strongly | Low | Yes |
| Anchor text optimisation | Works | Low if natural | Yes |
| Click depth reduction | Works strongly | Low | Yes |
Case studies: modern PageRank shaping in practice
Theory is cheap. Here are three documented case studies — two public, one anonymised from a recent client engagement — where modern internal-link sculpting delivered measurable lift without a single nofollow being added or removed.
Case Study 1: Niche Pursuits — Internal linking experiment
Spencer Haws of Niche Pursuits ran a three-month controlled experiment in which he added 108 contextual internal links across 47 existing blog posts — with no content additions and no new backlinks. The intervention was purely structural: identifying pages on similar topics and adding in-content links between them.
Result: tracked keywords improved across the majority of the 47 articles, with several individual posts showing 50–200% traffic increases over the experiment window. Critically, no nofollows were added or removed. The lift came entirely from concentrating equity through contextual placement and from reducing dependence on orphan link patterns.
Case Study 2: seoClarity — Enterprise e-commerce architecture
A documented case study from seoClarity worked with an enterprise e-commerce brand suffering organic traffic decay on deep subcategory pages. Pages four to five clicks from the homepage were losing ranking despite high commercial intent.
The intervention: split four similar categories into a control group and a test group. In the test group, additional internal links were added algorithmically from level-one and level-two pages to deeper level-three and level-four product pages. No nofollow manipulation. No new backlinks. Just better link distribution.
Result: 24% increase in organic traffic to the level two and three category pages in the test group, with the control group showing no comparable lift. This is internal sculpting working exactly the way the Reasonable Surfer patent predicts — contextual, high-prominence links from authoritative parent pages routing equity to commercially valuable children.
Case Study 3: Anonymised UK B2B SaaS — link removal sculpting
A UK-based B2B SaaS site (anonymised at client request) came to us in late 2025 with a specific problem: their highest-converting pricing page was ranking on page two for its primary commercial keyword, despite the site having strong overall domain authority and good backlinks pointing to the homepage.
Diagnostic audit revealed the underlying structural issue. The site’s global navigation contained 38 links. Footer carried another 24. Sidebar widgets added 11 more. The pricing page received only one internal contextual link from the entire content corpus, buried deep in a single blog post. Authority from the homepage was being diluted across 73 different boilerplate links, with the most important commercial target getting one of the smallest slices.
Intervention over a 6-week window:
- Global nav reduced from 38 links to 11 (essential top-level categories only)
- Footer trimmed from 24 to 9 (legal, contact, key resources only)
- Sidebar widget removed entirely from articles
- 17 new contextual links to the pricing page added from relevant blog content
- Click depth to pricing page reduced from 3 to 2
Result: the pricing page moved from position 14 to position 4 over 11 weeks, with no new external backlinks acquired during the test period. Organic conversions from that page increased 187% in the quarter following the change. Zero nofollow attributes were added. The shaping was achieved entirely through link removal and contextual addition.
| What these case studies share All three rely on real Google-documented mechanisms — contextual linking, the Reasonable Surfer model, and PageRank dilution through link count. None of them use nofollow as a sculpting tool. None of them got penalised by any subsequent core update, including the March 2026 update or the May 2026 rollout currently underway. This is the pattern that survives. |
The modern PageRank shaping framework
If you want a step-by-step process for sculpting internal equity in a way that aligns with how Google actually works in 2026, here it is. This is the framework we use on every site audit at LinkBuilding Journal, and it is what we would apply to any site looking to weather the May 2026 Core Update.
Step 1: Audit your link graph
Crawl your full site with Screaming Frog, Sitebulb, or Ahrefs Site Audit. Pull the internal link count for every URL. For deeper analysis of the tooling, see our review of the best link building tools for 2026. Identify pages with disproportionately high outbound link counts (typically the homepage and category pages) and disproportionately low inbound link counts (typically your highest-converting commercial pages).
Step 2: Identify your equity targets
List the 10–20 pages that drive the majority of your commercial outcomes. These are your sculpting targets. For most B2B SaaS sites this will be: pricing, feature pages, top-of-funnel comparison content, key category pages, and high-intent commercial blog posts.
Step 3: Reduce link bloat
Audit global navigation, footer, and sidebar. For every link, ask: does a typical user need this on every single page of the site? If the answer is no, move it to a more targeted location or remove it. Aim to cut total boilerplate link count by at least 40%.
Step 4: Add contextual inbound links to targets
For each target page, identify 5–10 existing content assets where a contextual link from that article to the target would be genuinely useful to the reader. Add these links with descriptive, query-relevant anchor text. This is the modern equivalent of sculpting and it is fully covered in our complete guide on link building strategies that work in 2026.
Step 5: Reduce click depth
For your top commercial targets, aim for a maximum click depth of 2 from the homepage. For all important content, aim for 3. Pages buried at depth 4+ should either be brought closer to the surface or accepted as low-priority assets.
Step 6: Monitor and iterate
Establish a baseline before making changes — keyword rankings for target pages, internal PageRank scores from your crawler, click-through rates from Search Console. Re-audit at week four and week eight. Refine.
Common mistakes that look like sculpting but aren’t
A few practices get confused with sculpting and need to be addressed explicitly.
Mistake 1: Nofollowing footer or sidebar links
This is the most common one. SEOs add rel=”nofollow” to footer or sidebar links pointing to low-priority pages, believing they are redirecting equity. They are not. They are destroying it. Remove the link entirely if it has no user value. Leave it dofollow if it does.
Mistake 2: Nofollowing internal search results or filter URLs
Better solutions exist: robots.txt disallow, meta noindex with nofollow on the page itself, or proper canonicalisation. Internal nofollow on link anchors is the wrong tool.
Mistake 3: Nofollowing affiliate or social media links
Affiliate links should use rel=”sponsored”. Social media links to your own properties should be dofollow — nofollowing them harms your brand SERP. For more on social and brand link strategy, see our deep dive on what are backlinks and how they work in 2026.
Mistake 4: “Sculpting” by adding more links to important pages
This sounds right but often backfires. If you add 20 new internal links from low-relevance pages to your pricing page, you are not concentrating equity — you are diluting the relevance signals Google uses to understand what the page is about. Quality and contextual fit beat quantity every time.
Where PageRank shaping is heading
Two trends are worth watching as we move through the rest of 2026 and into 2027:
- The dofollow/nofollow distinction continues to narrow. Since nofollow became a hint in 2020, Google has invested heavily in inferring link intent from context — editorial, paid, user-generated — without relying on the explicit attribute. The trajectory suggests this signal becomes increasingly redundant.
- LLM citations are emerging as a parallel equity layer. As ChatGPT, Perplexity, Claude, and Google’s own AI Mode (which crossed 1 billion monthly users last month) increasingly cite sources, internal linking decisions now affect not just classical SERP rankings but which of your pages get surfaced as citations in AI answers. Pages that are well-linked internally — with descriptive anchors and clear topical clustering — are more likely to be cited.
This second point matters more than most SEOs realise. The same internal architecture that wins through a core update also wins citations in generative search. Modern PageRank shaping is now a two-front investment, not a one-front one.
The verdict: is PageRank sculpting still worth doing?
The honest, evidence-based answer in 2026 has two parts.
Classic nofollow-based sculpting: no. It does not work. It has not worked since 2009. It actively destroys link equity. Every reputable SEO source — Ahrefs, Moz, Search Engine Land, Google’s own engineers — agrees on this. If you are still doing it, stop. The May 2026 Core Update is the right moment to remove every internal nofollow on your site that isn’t there for a legitimate sponsored/UGC reason.
Modern PageRank shaping: absolutely yes. Link count discipline, contextual placement, click depth reduction, hub-and-spoke architecture, and the elimination of boilerplate link bloat are some of the highest-leverage SEO investments you can make in 2026. They survive every core update. They compound with external link building. And they directly affect AI citation visibility, which classical SEO advice does not yet fully account for.
If you only do one thing after reading this article — audit your global navigation and footer, count the links, and ask which of them genuinely deserve to live on every single page of your site. The answer for most sites is “a lot fewer than you currently have.” That single change is worth more than every nofollow attribute you have ever added. For a broader strategic context on why this fits into the larger picture of building a site that ranks, see our link building statistics for 2026 and the foundational complete guide to what link building is.
Frequently asked questions
Is PageRank sculpting still a Google penalty risk in 2026?
Internal nofollow sculpting is not a manual penalty risk in itself — Google has never issued a manual action for it because the practice destroys your own equity, which is self-punishing. However, JavaScript-based link cloaking and other aggressive sculpting tactics can trigger spam signals and have been linked to algorithmic devaluation, particularly during core update rollouts like the one in progress now.
Did the May 2026 Core Update target PageRank sculpting specifically?
Google has not stated that the update specifically targets sculpting, and core updates are broad-spectrum quality refinements rather than targeted penalty waves. However, the early volatility data — the first 18 hours of the rollout — suggests that sites with bloated internal link counts and aggressive nofollow patterns are losing ground, while sites with clean architecture are stable or rising.
If nofollow doesn’t work, what should I use to block low-value internal pages?
The right tool depends on the goal. For pages you don’t want indexed at all: meta robots noindex. For pages you don’t want crawled at all (e.g., faceted search URLs in vast volumes): robots.txt disallow. For pages that exist for users but shouldn’t accumulate link equity: simply reduce the number of internal links pointing to them, and ensure those links live in non-prominent locations like the footer rather than global navigation.
Does the Reasonable Surfer patent mean I should make my important links visually larger?
In moderation, yes — links that are visually prominent, well-positioned within the content, and contextually relevant pass more equity than buried boilerplate links. However, don’t over-engineer this. Aggressive visual differentiation of links looks unnatural to users and provides diminishing returns. Focus on contextual relevance and natural placement first.
Should I add rel=”nofollow” to my login or cart page links?
No. Either leave them dofollow (the simplest option, and the one that does no harm because the pages typically have noindex on them anyway) or, if they appear in your global navigation purely for usability, consider whether they need to be in the global nav at all. Cart icons and login links often work better as targeted UI elements rather than navigation anchors.
How does PageRank sculpting interact with AI Mode and ChatGPT citations?
Internal link structure influences which of your pages LLMs surface as citations. Pages with strong internal authority — high inbound contextual link counts, descriptive anchor text, low click depth — are more likely to be selected when an AI assistant cites a source. Classic nofollow sculpting still does nothing useful here. Modern shaping (link count discipline, contextual linking, hub-and-spoke architecture) helps significantly.
How often should I re-audit my internal link structure?
For most sites, a full audit twice a year is sufficient — typically after major core update completions, when Search Console data has stabilised. Between audits, monitor: orphan page reports, click-depth metrics, and internal link counts to your top 20 commercial pages. Tools like Ahrefs Site Audit, Sitebulb, and Screaming Frog all surface this data cleanly.
Is hub-and-spoke architecture the same as PageRank sculpting?
Conceptually it’s a form of modern sculpting — you are deliberately shaping how equity flows by concentrating contextual inbound links on pillar pages and routing cluster pages through them. The difference is that hub-and-spoke uses real, useful, contextual links rather than nofollow tricks. It is what sculpting was always meant to be and what the original 2007-era version failed to become.
What about external nofollow links from my site? Do those waste my PageRank?
Technically yes — they consume a share of the page’s equity that then evaporates. But external links to authoritative, contextually relevant sources are a genuine quality signal in their own right and improve E-E-A-T. Don’t strip them out to chase a marginal equity gain. The right approach: link out generously to high-quality external sources where it helps the reader, and use the sponsored or ugc attributes where appropriate (not nofollow).
Will Google ever bring back “true” PageRank sculpting through nofollow?
Extremely unlikely. The 2009 change exists precisely because PageRank sculpting was a manipulation tactic that warped the link graph in ways Google didn’t want. Reversing the change would reopen that manipulation vector. Sixteen years on, Google has built an increasingly sophisticated link-weighting system (Reasonable Surfer, contextual scoring, link spam algorithms) that makes nofollow-based sculpting both unnecessary and undesirable from Google’s perspective. Plan accordingly.
