Link Building in 2026: What Still Works and What’s Dead

Link building has quietly changed shape over the past few years, and a lot of the advice still circulating hasn’t caught up. Google has been explicit about parts of this shift in its own policy language — the word choices in its documentation actually changed — while the rest is visible mainly in how practitioners talk about what’s working and what’s a waste of time now.

The nofollow shift that changed the incentive structure

In September 2019, Google published “Evolving ‘nofollow’ — new ways to identify the nature of links,” introducing two new attributes: rel="sponsored" for paid or affiliate links, and rel="ugc" for user-generated content like blog comments and forum posts. The more consequential change, effective March 1, 2020, was reclassifying nofollow itself from a strict directive into what Google calls a “hint” for crawling and ranking — meaning Google could, in some cases, still choose to crawl through or consider a nofollowed link, rather than ignoring it absolutely.

Google’s current guidance on when to actually use nofollow is narrower than a lot of site owners assume: use it “only when you don’t trust the source, and not for every external link on your site.” That’s a meaningfully different default than the reflexive “nofollow everything external” pattern I still see configured on WordPress sites — a pattern that, if Google is treating nofollow as a hint rather than an absolute block anyway, may not even be accomplishing what people think it is.

What Google has said about link volume specifically

Google’s spam policy language has also shifted in a smaller but telling way — an update in March 2024 reportedly removed the word “important” from language describing links as a ranking factor. That’s a subtle wording change, but it’s consistent with years of public commentary from Google’s own Search Advocates downplaying raw link count as something to chase directly. John Mueller has been reported saying his recommendation is not to focus on the absolute count of links a site has — the emphasis, consistently, has been on relevance and quality over volume.

I’d treat specific quoted phrasing from Mueller carefully — it comes through secondhand press coverage rather than a direct transcript in most cases — but the consistent theme across years of his public comments is real: link count alone isn’t the lever people treat it as.

What the industry data actually shows

Independent of Google’s own statements, industry surveys paint a picture of practitioners who still believe strongly in link building’s effectiveness, even as the tactics shift. A uSERP survey found roughly 70% of marketers rated backlinks’ impact on rankings as “big,” with the remaining 30% calling it “moderate” — essentially none rated it irrelevant. Aira’s industry survey found about 91% of respondents involved in link building called it effective or very effective. Separately, backlink-profile analysis comparing SERP positions has consistently found the #1 result holds roughly four times more backlinks on average than positions two through ten combined — a pattern that’s held up across multiple studies over the years, even if it says more about correlation than proof of direct causation.

None of that data comes from Google — it’s industry research, and worth labeling as such rather than treating it as confirmation of exactly how Google’s algorithm weighs links today. But it does explain why link building hasn’t disappeared as a practice, even as the specific tactics that count as “safe” have narrowed.

What still works, in practice

• Genuine coverage and citations — being mentioned by journalists, industry publications, or other sites because you’re a real source of information or data, not because you paid for placement.
• Original research and data — content that other sites want to cite because it contains something they can’t get anywhere else. This remains one of the most durable link-earning strategies precisely because it doesn’t rely on outreach tricks.
• Genuine relationships and guest contributions — real, relevant contributions to sites your audience already reads, disclosed appropriately with rel="sponsored" where compensation is involved.
• Digital PR built around a real hook — a genuinely newsworthy angle, not a manufactured “study” designed purely to bait links.

What’s dead, or close to it

• Paid link networks and PBNs — the risk-to-reward ratio has only worsened as detection has improved.
• Mass directory submissions — largely ignored as a ranking signal at this point, though a handful of genuinely relevant, curated directories still have value.
• Reciprocal link exchanges at scale (“link to me and I’ll link to you”) — an old pattern that’s easy for both algorithms and manual reviewers to spot.
• Chasing raw link count as a KPI — given Google’s own documented de-emphasis of link volume as a described ranking factor, treating “number of backlinks acquired” as the success metric misses what’s actually being measured now.

How this changes what a link-building budget should actually fund

If link volume is a weaker signal than it used to be while relevance and genuine earned coverage matter more, that has a direct budget implication: money spent on acquiring a large number of low-relevance links is increasingly poorly spent, even where it isn’t outright risky. A smaller number of genuinely relevant, hard-to-get placements — a mention in an actual trade publication, a citation in someone else’s original research, a link from a site your actual customers already read — is a better use of the same budget than paying for volume from sites with no topical connection to what you do. That’s a harder thing to scale and report on in a simple monthly dashboard, which is probably part of why the volume-based approach persisted as long as it did even after Google’s own language shifted away from emphasizing it.

The honest summary: link building isn’t dead, but link volume as a strategy is on much shakier ground than it used to be, both because Google has said as much and because the sites doing it well have shifted toward earning citations rather than acquiring placements. If your current link strategy is measured purely in links-per-month, that’s worth reconsidering regardless of what number you’re hitting.