Two-thirds of all pages have zero backlinks, the average bought link runs $361, and Google’s own Gary Illyes says they need “very few links” — then wished he hadn’t. This module is the honest version of link building: what a link is worth, what earning one really costs, and the digital-PR machine that makes journalists do the linking.
Every few years someone announces "links are dead." They're wrong. Multiple independent studies (Ahrefs, Backlinko, Semrush) consistently show that the number and quality of unique referring domains is the strongest correlated factor with ranking on competitive head terms. Google's own representatives confirm links remain a top-three ranking factor.
What HAS changed: the volume of low-quality links required to outrank a competitor isn't a real strategy anymore. Modern link building is about a smaller number of high-quality links, often won through PR, content marketing, and original research rather than direct "link building."
By the numbers The market reality of links in 2026
What links cost, how rare they are, and how hard they are to earn
66.31%
of all pages have zero backlinks (Ahrefs). Links remain scarce — which is why they still differentiate.
$361
average cost of an agency-built link (Authority Hacker survey); quality outreach links run $500-1,500.
0.7%
of 600,000 outreach emails became referring domains in Authority Hacker’s own program. Volume alone is a losing game.
8.5%
typical cold outreach reply rate. The other 91.5% is the cost of an undifferentiated pitch.
A page can have a high PageRank if there are many pages that point to it, or if there are some pages that point to it and have a high PageRank.
Sergey Brin & Larry Page, “The Anatomy of a Large-Scale Hypertextual Web Search Engine” (1998) — the sentence this entire industry grew from — Stanford InfoLab, original paper
What counts as a quality link
The factors that determine link value:
Domain authority of the linking site. Higher = better. A link from NYTimes is worth thousands from blogs.
Topical relevance. A link from a site that's topically related to yours is worth more than from a random high-authority site.
Page-level signals. The linking page's own authority and topical relevance matter, not just the domain's.
Anchor text. Descriptive, natural anchor text passes more topical signal. Over-optimized exact-match looks spammy.
Link placement. In-content editorial links are worth more than footer or sidebar links.
Editorial intent. Voluntary, editorial links signal endorsement. Paid, exchanged, or solicited links are heavily discounted.
Follow vs nofollow. Nofollow links pass less direct authority but still contribute to brand mention signals.
Traffic from the link. Links that send actual referral traffic are highly valued and harder to game.
Geographic relevance. For local businesses, links from same-region sources matter more.
Decision engine Is this link worth anything? Flip the switches
Five questions we run on every prospective link
Linking page is topically relevant to yours?
Placement is editorial (a human chose to cite you)?
Linking page is itself indexed and gets traffic?
Link is followed (not nofollow/sponsored/UGC)?
New referring domain (first link from this site)?
Prize link
Relevant, editorial, on a living page, followed, from a fresh domain — this is the link every tactic in this module is chasing. One of these outweighs a hundred directory listings.
Benchmark The anchor-text mix that looks earned
Because it is what earned profiles look like — RGM analysis across client link audits
target anchor mix
~50% branded & navigational (“RGM”, “realgrowthmatters.com”) — the natural majority on every clean profile we audit~15% naked URLs and generic (“here”, “this study”) — the noise real citations produce~25% topic & partial-match (“crawl budget research”) — where relevance accrues safely≤10% exact-match commercial — the slice Penguin was built to police; if this grows, someone is buying
You do not control most anchors — and should not try. The mix is a diagnostic, not a quota: when an audit shows 40% exact-match commercial anchors, we know what the last agency was doing, and so does Google.
Tactics that work and tactics that don't
What works (in 2024+)
Digital PR. Producing newsworthy research, data studies, or commentary that media outlets cover and link to.
Original research and data. Surveys, dataset analyses, industry reports. Journalists and researchers need cite-able sources.
Skyscraper / 10x content. Building genuinely better resources than what currently ranks, then promoting to people who linked to the older versions.
Resource page mentions. Existing curated lists of resources on your topic; pitch inclusion.
Broken link building. Finding broken outbound links on relevant sites, pitching your content as replacement.
HARO / Connectively / Qwoted. Helping journalists with quotes and expertise; get cited as expert source.
Podcast appearances. Many podcasts link to guest bios; quality content + reach.
Guest posts on legit editorial sites. Real publications with editorial standards (not paid placement networks).
Partnerships and integrations. Customers, vendors, partners often link to integration pages.
Community / sponsorship. Local sponsorships, industry event partnerships, conference participation often yield organic links.
What doesn't work (or is risky)
Buying links from link networks — manual action risk.
Reciprocal link exchanges at scale.
Footer links on partner sites en masse.
Spammy directory submissions.
Comment spam and forum signatures.
Private blog networks (PBNs) — high risk of detection and penalty.
Article syndication on low-quality sites with exact-match anchor text.
"Press release" distribution services that publish to hundreds of low-quality outlets.
Sponsored content without rel=sponsored markup — violates Google's policies.
The board Every tactic, graded — effectiveness, risk, and when we use it
The honest tactic matrix
Effectiveness
Risk
Cost curve
RGM verdict
Digital PR / data studies
HIGH
LOW
HIGH UPFRONT
CORE
Unlinked mention reclaim
HIGH
NONE
LOW
ALWAYS-ON
Editorial guest posts
MEDIUM
MEDIUM
MEDIUM
SELECTIVE
HARO / expert quotes
MEDIUM
LOW
LOW
ALWAYS-ON
Broken link building
LOW-MED
LOW
HIGH PER LINK
OPPORTUNISTIC
Niche edits (paid inserts)
SHORT-TERM
HIGH
$$$ RECURRING
NO
PBNs / link farms
DECAYING
SEVERE
$$$ + CLEANUP
NEVER
Two tactics are always-on because they monetize work already done (mentions you earned, expertise you have). One is core because it compounds. The bottom two are how sites end up in module 8’s recovery section.
We need very few links to rank pages… Over the years we’ve made links less important.
Gary Illyes, Google, SERP Conf 2024 — followed days later by his tweet: “I shouldn’t have said that…” — Search Engine Roundtable
Digital PR: the dominant pattern
Digital PR has overtaken traditional link building as the dominant strategy for earning high-authority links. The pattern:
Identify newsworthy angle. Original data, surprising finding, timely commentary on news, expert perspective on industry trends.
Produce the asset. Research report, data visualization, interactive tool, original analysis.
Pitch journalists and influencers. Personalized outreach to relevant reporters covering the topic.
Earn coverage. Articles, mentions, citations — with links back to your asset.
Amplify. Promote coverage via social, newsletters, internal channels.
What journalists need
A hook — the story angle, not the data.
Data they can cite with confidence (methodology, sample size, transparency).
Quotes from real experts (with credentials).
Visual assets (charts, images) ready to use.
Fast response when they reach out with follow-up questions.
Exclusivity or first-look access for top-tier outlets.
RGM EXPERT TRICK
Pitch the journalist’s next paragraph, not your content
Our outreach emails contain the stat, pre-written, in quotable form — one sentence with the number, the source, and the so-what. The journalist can paste it into their draft in ten seconds.
We stopped asking people to “check out our study.” Nobody has time to check anything out. We do the extraction for them: three bullet-ready findings, one chart they may embed, attribution line included.
Reply rates do not measure list quality; they measure how much work the recipient must do. Make the work zero and the 8.5% industry average stops applying to you.
WHY IT’S RARE · Everyone optimizes subject lines. Almost nobody optimizes the paste-ability of the asset itself — which is the actual conversion event.
Linkable assets
Asset type
What works
Original research
Surveys (200+ respondents minimum), proprietary dataset analyses, industry benchmarks
Calculators & tools
Mortgage calculators, ROI calculators, salary tools, comparison tools
Industry reports
Annual state-of-the-industry reports with primary research
Comprehensive resources that other writers reference as citation
Expert roundups
Curated insights from 10+ industry experts on a topic
Templates and frameworks
Practical artifacts professionals share and reference
News-jacking content
Timely commentary on industry news from credible experts
Awards and rankings
Recognition programs that recipients announce and link to
RGM EXPERT TRICK
Linkable assets come from sales calls and your own database
Before inventing a campaign, we mine two places: the questions prospects ask on sales calls (demand nobody has quantified) and the client’s own operational data (supply nobody else owns).
“Ultimate guides” do not earn links — journalists cite numbers, not summaries. One proprietary benchmark (“median response time across 4,000 service tickets”) outearns ten guides.
Our cadence: one data study per quarter, refreshed annually so the citation keeps pointing at a live number. The refresh costs 10% of the original and protects 100% of the links.
WHY IT’S RARE · Most teams brainstorm assets from keyword tools, producing what already exists. The data you alone hold is the only asset competitors cannot brief an agency to copy.
Outreach mechanics
What makes outreach work
Personalization. Reference the specific article they wrote, the specific resource page, the angle they cover.
Brevity. Two short paragraphs. The pitch and the value, nothing more.
Concrete value. Specific data, specific finding, specific reason it's relevant to their work.
No demand. Don't ask for a link in the first email. Make the case and let them decide.
Right time of day. Tuesday–Thursday, 9–11am in the recipient's time zone for most professional outreach.
One follow-up. A single, polite follow-up 5–7 days later. After that, move on.
Outreach response rates by tactic
Tactic
Typical response rate
HARO/Connectively responses
15–30% pickup rate when responses are useful
Cold journalist pitches with data hook
5–15% response, 1–5% coverage
Resource page outreach
5–10% inclusion rate with good content
Broken link building outreach
2–5% conversion to link
Generic guest post outreach
1–3% — often not worth the effort
Field data What 600,000 outreach emails actually produced
Authority Hacker’s published funnel — the honest math of cold outreach
Emails sent
600,000
↓ deliverability, relevance, attention
Replies (~8.5%)
~51,000
↓ negotiation, editorial fit
Referring domains
4,300
Source: Authority Hacker, 600,000 emails analyzed. End-to-end: ~0.7% of emails became linking domains. The number is not an argument against outreach — it is an argument against treating outreach as a volume game instead of an asset game.
Calculator Your real cost per link — run your own funnel
Four inputs from your program; the market benchmark is $361
$627per link · 6.4 links/mo
ABOVE MARKET
Program cost = hours × loaded rate + tools + content. Benchmark: $361 average agency link (Authority Hacker survey); quality earned links commonly $500-1,500. If your cost per link beats the market AND quality holds, scale; if not, the lever is usually the asset, not the volume.
Case study · Authority Hacker · the 600,000-email program
Authority Hacker published the funnel most agencies hide: across their own campaigns, 600,000 outreach emails to ~150,000 site owners produced 4,300 referring domains — roughly one link per 140 emails. Their published lessons match what this module teaches: personalization moved replies more than any template, the asset being pitched dominated every other variable, and follow-ups roughly doubled response. The often-cited takeaway: outreach works as the distribution arm of something worth citing — and only as that. (Authority Hacker study)
Unlinked mentions and brand monitoring
Many sites mention your brand without linking. Tools like Ahrefs Alerts, Google Alerts, Mention, Brand24 surface these. Outreach to add the link is one of the highest-ROI link tactics:
Site is already willing to mention you (they did).
Editorial relationship doesn't need to be built.
Simple ask: "Could you add a link to our site at the mention?"
Conversion rates 20–40% on these requests when polite.
Toxic links and the disavow tool
Most sites accumulate spammy backlinks they didn't build — from scraper sites, automated link networks, foreign-language spam. Google says it largely ignores these and disavowing isn't necessary in most cases.
Exceptions:
If you have a manual action against your site for unnatural links.
If you have evidence (algorithmic ranking drop coincident with a spam attack) that spammy links are affecting you.
If you bought links in the past and need to clean up the legacy.
For most sites, the disavow tool is a non-issue. Don't reflexively disavow.
RGM EXPERT TRICK
The disavow file is almost always a placebo — we bill those hours elsewhere
Unless there is a manual action in Search Console, we do not disavow. Google says it ignores most spam links algorithmically, and our before/after tests on inherited disavow files show no movement when we remove them.
The toxic-link panic is vendor-manufactured: scores from tools, not from Google. We have watched teams disavow their way through a quarter while competitors earned twelve real links.
The exception is real: manual action, or a documented paid-link history that needs cleanup before a migration. Then the file is surgical, with evidence, not a tool export.
WHY IT’S RARE · Saying “do nothing” feels negligent, so agencies sell the file. Knowing when inaction is the expert move is what the audit trail is for.
Measuring link building program effectiveness
Referring domains acquired per month. Track new unique referring domains over time.
Quality of referring domains. DR/DA distribution. Are the new links from high-authority sites?
Topical relevance of referring domains. Categorize by industry/topic; track relevance ratio.
Referral traffic from link building. Links that send traffic are higher-value; track in GA4.
Ranking improvements on target topics. Link investments should correlate with cluster ranking lift.
Cost per qualified link. Total program cost / number of high-quality links earned.
Coverage volume from PR campaigns. Articles, mentions, syndication from each PR push.
Advanced playbook
Annual data study. Plan one large primary-research study annually. Survey, dataset, industry analysis. Maximum effort, maximum payoff. Often produces 100+ links per study.
Newsroom-style content team. Embed PR, content, and SEO in one team rather than separate functions. Speed of response to news cycles becomes a competitive advantage.
Topic-aligned PR pitches. Don't pitch the same data to every journalist. Slice the data differently for each beat. Same study, ten different angles for ten different outlets.
Reactive expert source positioning. Track industry news daily; offer expert commentary within hours. Journalists on deadline link to fast experts.
Internal linking from new PR-earned pages. When external authority pages link to your site, ensure your internal linking distributes that authority through the topic cluster.
Quote-source content for journalists. Create "ask the expert" pages with quotable, citation-ready sound bites on key topics. Journalists discover and cite these.
Strategic conference speaking. Conference talks generate guest post invitations, podcast appearances, and citations. Build a roster of speakers.
Awards and recognition programs. Hosting your own industry awards generates inbound links from winners. Apply for industry awards yourself for the same effect.
Educational partnerships. Partnerships with universities, accreditation bodies, certification programs often yield .edu links — among the most authoritative.
Wikipedia and Wikidata source citations. Be the cite-worthy source on your topic. Wikipedia editors link to authoritative original sources.
Backlink gap analysis quarterly. Use Ahrefs/Semrush to find sites linking to multiple competitors but not you. Targeted outreach with high relevance fit.
Newsletter and substack linking ecosystem. Build relationships with industry newsletter writers; they cite frequently and influence professional buyer audiences.
Step by step A digital PR campaign, the RGM sequence
From dataset to links in seven moves
Start from data only you hold.Client database, survey you can field, public data nobody joined before. The test: can a journalist get this number anywhere else? If yes, restart.
Find the tension before producing anything.A finding that surprises, reverses an assumption, or localizes a national story. “Rents rose” is weather; “rents rose fastest in the 10 cities building the most” is a story.
Build the asset page to be cited.Methodology visible, key numbers in copy-paste form, charts embeddable, one URL forever. This page is infrastructure — module 4’s information-gain logic, weaponized.
Cut the list by beat, not by DA.Fifty journalists who cover exactly this beat their editors expect — not five thousand “relevant sites.” Read three recent pieces by each; note the angle they favor.
Pitch the paragraph, not the page.Subject = the finding. Body = three bullet-ready stats with attribution, offer of exclusive cuts or expert quotes. Zero “just checking out our new study.”
Follow up once, with a new angle.One follow-up, 3-4 days later, leading with a different cut of the data. Follow-ups roughly double response; nagging with the same email does not.
Reclaim and refresh.Monitor mentions of the study; convert unlinked citations (the easiest links in the program). Refresh the data annually so the URL stays the live source of record.
Common mistakes
Outsourcing link building to spammy services that buy or fabricate links.
Treating link count as the goal instead of link quality and relevance.
Generic outreach with no personalization.
Pitching journalists without a real news hook.
Asking for links in first email instead of building the case.
Reciprocal linking with weak partners at scale.
Sponsored content without rel=sponsored markup.
Disavowing aggressively without evidence of harm.
Building links to homepage only and ignoring deep page authority distribution.
No internal linking strategy to distribute link equity from new external links.
Treating links as goal, not as one signal among many.
Forgetting brand monitoring — missing unlinked mentions that are easy wins.
Ten questions, CASE method (Context · Analysis · Strategy · Execution). Pass at 90% to unlock this module’s completion passcode — retake as many times as you like.