Chapter 23: Earning Links — Creating Content That People Want to Reference, Cite, and Share

"The best way to get other sites to create high-quality, relevant links to yours is to create unique, relevant content that can naturally gain popularity in the Internet community." — Google Search Central documentation, on link schemes

Overview

In Chapter 22 we established the thing everyone in SEO eventually collides with: after a quarter-century of algorithm evolution, links are still the strongest off-page signal Google has. A link from another site is a vote, and votes from trusted, relevant pages still move rankings more than almost anything else you can do off your own domain. So the question that ends Chapter 22 is the question that opens this one, and it is the question that quietly funds an entire shady industry: how do you actually get links?

The industry's answer, screamed from ten thousand inboxes a day, is "outreach" — a polite word for emailing strangers to ask, beg, or pay for links. It mostly does not work, and where it does work it tends to buy you exactly the kind of link Google has spent twenty years learning to ignore or punish. This chapter teaches the other answer, the one that is slower, harder, less glamorous, and the only one that compounds: you earn links by creating things that are genuinely worth referencing, and then you make sure the right people know those things exist.

That reframe — from building links to earning them — is not a motivational slogan. It is a different activity with a different center of gravity. Link building starts with the link and works backward to some pretext for deserving it. Link earning starts with something real — a piece of original data, a genuinely definitive guide, a free tool that saves someone an afternoon — and treats the links as the natural byproduct of value that already exists. One is a sales problem. The other is a publishing problem. This book, unsurprisingly, is on the side of publishing.

In this chapter, you will learn to:

• Tell the difference between building a link (asking for a vote) and earning one (deserving it), and why only the second is durable.
• Design a linkable asset — original research, a definitive guide, or a free tool — and choose the type that fits your site, your budget, and your audience.
• Use digital PR and newsjacking to get cited by journalists, and recognize when each is appropriate and when it's a waste of a week.
• Work journalist-request platforms (the model everyone still calls HARO) to earn authoritative mentions.
• Write outreach that a busy person actually reads — and understand, precisely, why the mass-email version fails.
• Measure link-earning with the one metric that matters most, referring domains over time, and ignore the vanity numbers.
• Sit honestly with the limits: this is slow, uncertain work with no guaranteed payoff, and anyone who tells you otherwise is selling something.

Learning Paths

Everyone should read §23.1 (the mindset) and §23.7 (the honest limits) — they are the frame the tactics hang on. 🏪 Local Business: §23.2 and the Strategy File are your chapter; a single well-made local data asset can out-earn a year of cold email. 📝 Content Creator: §23.2 (linkable formats) and §23.5 (outreach) are where your craft becomes distribution. 🛒 E-Commerce: weight §23.2 (buying guides, data, tools) and §23.3 (digital PR) — product pages rarely earn links; assets do. 🔧 Developer: §23.2's free tools and calculators are the highest-leverage linkable asset you can build, and often the one only you can build. 📊 Strategist: live in §23.6 (measurement) and §23.7 (limits) — your job is to set honest expectations and prove the slow thing worked.

23.1 The mindset: links are earned, not built

Start with a distinction the SEO industry works hard to blur, because blurring it is profitable.

Link earning is the practice of acquiring links to your site as the natural consequence of publishing something other people independently choose to reference. You make the thing; other people decide, on their own editorial judgment, that it's worth pointing their readers to; they link. You did not ask for most of those links, and you could not have forced them. The phrase people use for that ideal is the editorial link — a link given freely, by someone's choice, because your page genuinely helped their reader (we met the term in Chapter 22). Earned links are, overwhelmingly, editorial links.

"Link building," as the industry usually means it, inverts the order. It starts from a target — "we need forty links this quarter" — and reverse-engineers activities to hit the number: emailing bloggers, trading links, writing guest posts for the link rather than the audience, and, at the ugly end, simply buying them. Some of these activities are legitimate and we map them honestly in Chapter 24. But the mindset is backwards, and the backwardness is the problem. When the link is the goal and the content is the pretext, you produce content that exists to justify a link — which is to say, content nobody would ever link to on its own. You can feel the difference in your inbox every day.

🔎 How Search Sees It Google does not see your intentions; it sees a pattern. A page that earns links naturally tends to accumulate them from a diverse set of relevant domains, with varied anchor text, at an irregular pace that tracks real events (a study gets published, a journalist cites it, other writers cite the journalist). A page whose links were manufactured tends to show the opposite fingerprint: a burst of links from unrelated or low-quality sites, over-optimized exact-match anchor text, footer or author-bio placement, and no independent reason for any of it to exist. Google's spam systems — the machine-learned system it calls SpamBrain among them (Chapter 6) — are built to tell these patterns apart at scale. You cannot reliably fake the fingerprint of an earned link, because the fingerprint is the earning. This is theme 1 of this book applied to authority: you are not trying to trick Google into seeing votes that aren't there; you are trying to earn votes that are, and to make them legible.

Here is the mental model to carry. A link is not a thing you obtain. It is a thing that happens to a page that deserves it. Your job is to maximize the number of pages you own that deserve it — and then to reduce the friction between "deserves a link" and "gets one" by making sure the people who would link can find you. That friction-reduction is real work (it's §23.3 through §23.5), but it is downstream of the deserving. If the deserving isn't there, no amount of outreach saves you; you're just a more efficient spammer.

FIGURE 23.1 — "Two ways to get a link"                          [schematic — not to scale]

  BUILDING (starts with the link)          EARNING (starts with the value)
  ─────────────────────────────            ─────────────────────────────
  "We need 40 links"                        "We made something worth citing"
        │                                          │
        ▼                                          ▼
  invent a pretext ──▶ email strangers      publish the asset ──▶ people find it
        │                                          │                     │
        ▼                                          ▼                     ▼
  low reply rate,                           a journalist cites it   others cite
  low-quality links,                              │                 the journalist
  scheme risk                                     ▼                     │
        │                                   an editorial link           ▼
        ▼                                          │              links accrue from
  stops the moment                                 ▼              diverse relevant
  you stop paying                            compounds over time  domains, on their own

Read the two columns side by side. The left column is a treadmill: the links stop the instant the effort (or the money) stops, because nothing about the page ever earned them. The right column is an asset: once the thing exists and gets discovered, it keeps attracting references with no further input from you, the way a good reference book keeps getting cited for decades. That difference — treadmill versus asset — is theme 5 of this book, and it is the entire strategic case for doing the harder thing. Bought authority collapses when you stop buying. Earned authority compounds while you sleep.

⚖️ Evidence Check Claim: "Earned links work better than built links." Where does this sit on the honesty scale? — Confirmed by Google (direction, not magnitude): Google's public guidance explicitly discourages links "intended to manipulate ranking" and states that the durable path is content that "naturally gains popularity." Its link-spam policies exist precisely to devalue manufactured links. So Google's stated design intent is that earned links count and manufactured ones don't (or worse). — Strong correlation, not clean proof: large-scale studies from SEO tool vendors consistently find that pages and sites with more referring domains tend to rank higher — a robust correlation across the industry. But correlation isn't proof that any specific earned link caused any specific ranking gain; links travel with quality, brand, and content, which also drive rankings. — The author's professional experience: the sites that hold rankings through core updates are, in my experience, the ones with genuinely earned link profiles; the ones that crater are disproportionately the ones that bought or schemed their way up. That's a pattern I trust — but it's judgment, not a controlled experiment, and I'll label it as such.

🔗 Connection This chapter assumes Chapter 22 — what a link is (a backlink), what a referring domain is, how PageRank models links as votes, and why authority, relevance, anchor text, and placement decide a link's value. We won't re-teach that here; we build on it. The specific tactics for pursuing links one by one — guest posting, resource pages, broken-link building, unlinked mentions, and the black-hat schemes covered only to be avoided — are Chapter 24. This chapter is about the strategy those tactics should serve: being worth linking to in the first place.

23.2 Linkable assets: build something worth citing

If earning links begins with deserving them, then the central skill of this whole chapter is creating things that deserve them. We call such a thing a linkable asset: a page or resource created specifically to be referenced by other people — not merely to rank, and not merely to convert a visitor, but to be the thing a writer points at when they need to make a point. The test of a linkable asset is a single question: when would someone else, writing their own article, need to link to this? If you can answer that question with a concrete sentence — "a journalist writing about winter heating costs needs a local price figure, and we're the only local source that has one" — you have a linkable asset. If you can't, you have a page.

Most pages are not linkable, and that is fine — most pages exist to serve your searcher, not to be cited by third parties. Your service pages, your product pages, your location pages: these earn traffic by ranking, not by being referenced, and trying to force links to them is a fool's errand. Linkable assets are a deliberate minority of your content, built on purpose to attract the references that then flow authority (through internal links, Chapter 15) to the pages that actually pay the bills. Understanding that division of labor is half the battle: you don't need every page to earn links; you need a few pages built to earn links, and an internal-linking structure that spreads the resulting authority.

There are a handful of asset types that reliably earn references. Here are the ones worth your time, with an honest accounting of what each costs and where each fails.

Notice what these have in common: each answers the question "when would someone need to reference this?" with a real occasion. And notice what's absent: "a blog post about our services," "a listicle of tips everyone already knows," "a page optimized for a keyword." Those aren't assets; they're inventory.

Original data is the most durable asset of all

Of the six types, original research — original data you gathered and nobody else has — is the most powerful, because it is the hardest to copy and the most useful to cite. A writer can paraphrase your guide, but if they want to state that "furnace replacement in the Rivertown metro runs between X and Y," they have to cite someone, and if you're the only one who measured it, they cite you. Data creates a citation obligation that prose does not. That is why the "state of the industry" report, the annual survey, the price index, and the "we analyzed 10,000 X" study are perennial link magnets across every serious content program on the web.

Crucially, you do not need a research department to have original data. You need a proprietary vantage point. A plumbing company sits on thousands of real invoices — that's a price dataset no journalist can get elsewhere. A software company sits on aggregate usage patterns. A newsletter sits on a survey-able audience. The data you already generate by operating is, very often, the most linkable thing you will ever publish — you've simply never packaged it. (Chapter 32 returns to original research as a link engine for B2B, and Chapter 13 is where we're careful about the line between analyzing your real data and fabricating a convincing-looking "study" — the latter is worse than useless, as §23.7 and Case Study 2 will show.)

📄 Read the Report

text FIGURE 23.2 — "Why one page earns links and its neighbor doesn't" [constructed teaching example] THE PAGE Two pages on the same home-services site. Page A: "5 Signs You Need a New Furnace" (a competent blog post). Page B: "2026 Metro Furnace-Replacement Cost Report" (real price ranges from the company's own 1,200 jobs, by city, with methodology). WHAT'S THERE Page A: solid advice available on a thousand other sites. Page B: numbers that exist nowhere else, a chart, a stated sample size and date, and a downloadable summary. WHAT IT SHOWS Page A has ~2 referring domains after a year (a scraper and a directory). Page B has ~30, including two local news sites and several realtor blogs — because each of them needed a local cost figure and B was the only source. WHAT IT DOESN'T It does NOT show that B "went viral" (it didn't) or that B was easy (gathering and verifying the data took real work). And B's links took months, not days, to appear. THE MOVE Stop pouring effort into more Page-A posts hoping for links. Build one Page B per year, properly, and give it the promotion §23.3–23.5 describe. THE LESSON Links follow citation obligation. Prose can be paraphrased; original data must be credited. Be the thing that must be credited. (All figures illustrative; the mechanism — data creates citations — is the real lesson.)

Definitive guides and free tools

If you can't generate original data, the next most reliable assets are the definitive guide and the free tool. A definitive guide earns links by being so complete that another writer would rather link to it than re-explain the topic — "for the full breakdown, see [this guide]." The bar, though, is genuinely high and routinely underestimated: "definitive" means more complete, more current, and more useful than everything else that ranks, not merely "long." A 4,000-word article that merely restates what the top ten results already say is not definitive; it's redundant, and redundant content earns nothing. (This is why the "skyscraper" idea from Case Study 1 — find a popular resource, build something better, tell the people who linked to the original — works in principle and disappoints in practice: most people do the "better" part lazily.)

A free tool or calculator is often the single highest-leverage asset a developer can build, because utility gets reused, and reuse gets cited and bookmarked far more than an article ever will. A mortgage calculator, a "which permit do I need" wizard, a load-size estimator — anything that turns a question the audience keeps asking into an instant answer becomes a reference other sites point their own readers toward. The catch is real: tools cost real development time, they need maintenance, and a broken tool is worse than none. But when they work, they are the most durable asset of all, because their usefulness doesn't decay the way a "2024 guide" does.

🛠️ Try It on Your Site Open Google Search Console (free), go to the Links report, and look at Top linked pages — the pages on your own site that have earned the most links from other sites. This is a mirror: it shows you, from real data, what your audience has found reference-worthy. Two things to notice. First, is anything there at all? For most small sites, the honest answer is "barely" — which is the problem this chapter exists to fix. Second, whatever is there — a tool, a data post, an unusually thorough guide — is a pattern: it tells you the kind of asset your niche links to. Build more of that kind on purpose. (The report is approximate and updates slowly; treat it as direction, not a ledger.)

🔄 Check Your Understanding A local bakery wants links. Its owner asks whether to (a) write ten blog posts about baking tips, or (b) survey fifty local cafés about what they pay for flour and publish a "Local Wholesale Ingredient Cost Index." Which is more likely to earn links, and why — in one sentence about citation obligation?

Answer

(b). Baking tips are available on thousands of sites, so no one has any obligation to cite this bakery for them; a local ingredient-cost index is data that exists nowhere else, so any local writer covering food costs must cite the bakery to use the figure — that citation obligation is what turns into links. The tips might get traffic; only the data reliably earns references.

23.3 Digital PR and newsjacking: getting cited by journalists

Building the asset is half the job. The other half is that the right people have to find out it exists, and the highest-authority "right people" on the open web are journalists and editors. A single link from an established news outlet can carry more authority — and, through syndication, spawn more downstream links — than months of small-blog outreach. Earning those links has a name: digital PR.

Digital PR is the practice of earning links and mentions from journalists, editors, and established publications by giving them something genuinely newsworthy — most often, original data or expert commentary. It is the collision of two old disciplines: traditional public relations (building relationships with the press and pitching stories) and SEO (caring that the resulting coverage includes a link). Done well, it is simply being a useful source. Done badly, it is spamming the press desk with a "story" about your product launch that no reader outside your company cares about — and journalists, who receive hundreds of pitches a day, delete that instantly.

The engine of digital PR is almost always a linkable asset from §23.2, usually data. You don't pitch a journalist "please link to our website." You pitch them a story your data makes possible: "Home-heating costs in the metro rose sharply this winter — we have the local numbers, broken down by city." The data is the story; the link is how they attribute it. This is why §23.2 comes first: digital PR with nothing to say is just noise. Digital PR with a genuinely useful dataset is a service to the reporter, and reporters reward services to them with exactly the thing you want — an attributed link.

Newsjacking is a specialized, time-sensitive form of the same move. Newsjacking means inserting your expertise or data into a breaking news story while it's still developing, so that a journalist writing that story right now cites you. A heat wave hits and every outlet is writing "how to keep your home cool" — the HVAC company that can, that day, offer a technician quote and local data becomes a source. The key word is timing. Newsjacking has a brutally short window: it is worth almost nothing the day before the news breaks (there's no story yet) and almost nothing a week after (the story is filed and forgotten). You are useful only inside the window when the story is hot and the journalist is actively looking for sources.

FIGURE 23.3 — "The newsjacking window"                          [schematic — not to scale]

  value of your pitch
      ▲
      │                 ██
      │              ██ ██ ██          ◀── the window: hours to a couple of days,
      │            ██ ██ ██ ██ ██          while the story is hot and reporters
      │         ██ ██ ██ ██ ██ ██ ██       are actively hunting for sources/quotes
      │  ▁▁▁▁▁██ ██ ██ ██ ██ ██ ██ ██ ██▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁▁
      └────────────────────────────────────────────────────────▶ time
        before:          NEWS          after: story is filed;
        no story yet     BREAKS        your pitch is now late and useless

The diagram carries the whole discipline: newsjacking rewards speed and relevance over polish. A rough, genuinely useful quote sent within the window beats a beautiful pitch sent two days late. This is why newsjacking suits businesses with real, on-the-ground expertise — a technician who can explain why heat pumps struggle in a cold snap is a gift to a reporter on deadline — and why it fails for businesses with nothing timely to add.

🔎 How Search Sees It A link from a news article is not magic dust, and it's worth being precise about what it does and doesn't do. Many large publishers mark outbound links from certain sections as nofollow, sponsored, or ugc (Chapter 22) — attributes that tell Google not to pass full authority. So some hard-won press links pass less ranking value than SEOs hope. But — and this matters — a genuine editorial mention in a reputable outlet still helps in ways that don't depend on the link attribute: it builds brand, it gets seen by other writers who then cite you with followed links, it can feed Google's understanding of your business as a real entity (Chapter 4), and it diversifies your traffic beyond search (theme 6). Chase press coverage for the whole value, not just the single link's PageRank — and never assume a big-name link is a followed one without checking.

🔗 Connection The mechanics of evaluating a specific link opportunity — is this outlet relevant, authoritative, actually read by humans, and does it link editorially? — belong to Chapter 24 (§24.1). Local press and community links specifically (chambers of commerce, local sponsorships, event coverage) are developed in Chapter 24 (§24.5) and pay off in Chapter 25 (local SEO). Here we're establishing the strategy: data plus a real story equals a reason for a journalist to cite you.

23.4 HARO and the journalist-request platforms

Digital PR usually means you going to journalists. There is a mirror-image channel where journalists come to you, and for many small businesses it is the single most accessible way to earn a high-authority mention: the journalist-request platform.

The archetype — the service whose name became a generic verb in the SEO world — was HARO, short for Help A Reporter Out. HARO was a free service, founded in 2008, that connected journalists who needed sources with experts willing to be quoted. A reporter working on a story would submit a query ("I need an HVAC expert to explain heat-pump efficiency in cold climates by Thursday"); the platform emailed those queries out to a huge list of subscribers; experts replied with a useful quote; and the reporter, if they used your answer, quoted you and — often — linked to your site. For a plumber or a consultant or a founder with real expertise and no PR budget, it was a genuine on-ramp to mentions in outlets they could never have pitched cold.

Here the book's honesty rule requires a note that most SEO articles won't give you, because most SEO articles are older than the fact: HARO, as a specific product, no longer exists. It was acquired over the years (ending up under the media-intelligence company Cision), rebranded as "Connectively," and then shut down entirely at the end of 2024. People in the industry still say "HARO" the way people still say "Google it" — as a generic term for the category — but if you go looking for the original service, you won't find it. This is a small, useful lesson in itself, and we'll draw it out fully in Case Study 2: a tactic that depends on a single platform is a tactic that can vanish overnight.

The category, though, is alive and well, served by a rotating cast of alternatives — Qwoted, Featured (formerly Terkel), SourceBottle, ProfNet, and, increasingly, journalists posting requests directly on social platforms under tags like #JournoRequest. The specific product names will keep changing; the model is durable, and the skill transfers to whichever platform is current when you read this. So we teach the model, not the brand.

📄 Read the Report

text FIGURE 23.4 — "Two answers to the same journalist request" [constructed teaching example] THE REQUEST "Looking for a licensed HVAC pro to explain, in plain English, why a heat pump can struggle below freezing — and what a homeowner can realistically expect. Need it by 10am tomorrow. Please include name, title, and company." (deadline: ~18 hours) ANSWER A (loses) Sent 2 days later. Three paragraphs of generic marketing about "our award-winning service." No plain-English explanation. A link to the homepage. No credentials. ANSWER B (wins) Sent in 40 minutes. Four tight sentences that actually answer the question, in plain English, with one vivid concrete detail only a working tech would know. Signed with a real name, "licensed HVAC technician, 14 years," and the company. Offers a follow-up. WHAT IT SHOWS The reporter used B verbatim and linked the company. B won on the three things these platforms reward: SPEED (inside deadline), RELEVANCE (answered the actual question), and CREDIBILITY (a real, named, credentialed human). WHAT IT DOESN'T B is not guaranteed a link — the reporter might not use it, or might quote without linking, or the outlet may nofollow. You control the pitch quality, not the outcome. THE LESSON Be a genuinely useful source, fast. These platforms reward expertise made easy to use, and punish self-promotion made slow.

Two disciplines make the difference between people who earn links this way and people who waste an hour a day on it. The first is speed: reporters work on deadline, and a perfect answer sent after the deadline is worth nothing. The second is genuine expertise expressed plainly: the answers that get used are the ones that actually help the reporter's reader, written by someone who obviously knows the subject, not dressed-up sales copy. This is E-E-A-T (Chapter 5) operating in the wild — the "Experience" and "Expertise" that these outlets are, quite literally, checking for when they pick a source. A licensed technician answering a heating question is the expertise; a marketer paraphrasing a Wikipedia article is not, and reporters can tell.

⚖️ Evidence Check Claim: "Journalist-request platforms are a reliable link source." Honest sorting: — The author's professional experience: they genuinely work — I've earned mentions in outlets I could never have pitched cold — but the hit rate is low (you'll answer many requests per link earned) and it is labor you must sustain. It rewards consistency, not one heroic afternoon. — Speculation, flagged as such: whether these mentions move rankings directly is unknowable at the single-link level; their reliable value is brand, referral traffic, and the followed links that sometimes follow from being seen. Treat the ranking effect as a possible bonus, not the reason you do it. — What's NOT true: that you can automate it. Templated, mass-submitted answers are the fastest way to get filtered out; the whole value is a real human being genuinely useful, which brings us to §23.5.

23.5 Outreach that works — and the myth that it "just works"

Sooner or later, earning links involves telling a specific human being that your thing exists. That is outreach: the practice of contacting a relevant person — a writer, an editor, a site owner, a resource-page curator — to make them aware of something you've published that would genuinely help their audience. Notice every qualifier in that sentence, because every one of them is load-bearing, and the industry ignores all of them. Relevant person. Genuinely help. Their audience. Outreach that respects those words is a service. Outreach that ignores them is spam, and it is the single most hated genre of email on the professional internet.

Let us dispose of the belief that funds most of it.

🚫 SEO Myth: "You can just email people and they'll link to you." This is the load-bearing fantasy of the entire link-building industry, and it is false in the way that matters most: at scale, it doesn't work, and the version that "works" produces links Google is built to ignore. Here is the reality. Blindly emailing hundreds of site owners "I loved your article, would you add a link to my page?" earns reply rates so low they round to zero, because the recipients get dozens of identical emails a day and can smell a template instantly. The tiny fraction who do respond often want payment — and a paid link is a link scheme under Google's guidelines (Chapter 24), a liability, not an asset. So the mass-email model fails twice: it mostly gets ignored, and where it succeeds it buys you the wrong kind of link. What actually earns a link is not the email; it's the thing the email is about. If you have a genuinely useful asset and you tell exactly the right person, personally, why it helps their specific readers, a small number will link — not because you emailed, but because you gave them something worth linking to and made it easy to find. The email is the last five percent of the work. People who believe the myth do only the email, skip the ninety-five percent, and then conclude "outreach doesn't work." They're half right: their outreach doesn't, and can't.

With the myth cleared, here is what real outreach looks like. It is low-volume, high-relevance, and personal. You are not blasting a list; you are contacting a short list of people for whom your asset is genuinely, specifically useful, and you're proving in the first sentence that you know who they are and why you're writing to them.

FIGURE 23.5 — "The same outreach, spam vs. earned"               [constructed teaching example]

  SPAM VERSION (ignored)                    EARNED VERSION (gets read)
  ──────────────────────                    ─────────────────────────
  "Hi Dear, I loved your article!           "Hi Priya — your Rivertown Gazette piece on winter
   I have a great resource that would         utility bills cited the state average for heating
   be perfect for your readers. Would         costs. We just published local numbers: furnace-
   you consider adding a link to it?          repair and replacement costs for the metro, broken
   [generic URL]. Thanks!"                     out by all five cities, from 1,200 of our own jobs.
                                              The Cedar Hills figure surprised us. Data + method
   • Wrong/no name                             here if it's useful for a follow-up: [link].
   • No evidence they read anything            No ask — just thought it might help your beat."
   • "would be perfect for your readers"
     (they have no idea who the readers are)  • Real name, real article referenced specifically
   • A pure ask, zero value                    • Offers DATA that fits the writer's exact beat
   • Identical to 500 other emails            • Leads with THEIR reader's benefit, not a link ask

Study the earned version. It references a specific piece the writer actually wrote. It offers something of genuine value to that writer's beat — local data they can't get elsewhere. It leads with the reader's benefit, not with a request. And, tellingly, it barely asks for anything: it makes the useful thing easy to find and trusts a good writer to do the obvious. That restraint is not a trick; it's a consequence of actually having something worth citing. When your asset is real, you don't have to beg — you just have to inform. When it isn't, no email is polite enough to fix that.

A few honest points about outreach's limits, because it is not a magic wand even done right:

• Most of your good, personal, relevant emails will still be ignored. Busy people are busy. A reply rate that's a fraction of your sends is a good day. Volume of quality helps; volume of spam hurts.
• You cannot control the outcome, only the inputs. You control who you contact, how relevant your asset is, and how respectfully you write. You do not control whether they link, whether they nofollow it, or whether they even open the email. Anyone promising a link count from outreach is promising something they can't deliver.
• It does not scale the way spam scales, and that's the point. The moment you try to make outreach a numbers game — templates, automation, buying lists — you've rebuilt the thing the myth describes, and you get the myth's results. Its non-scalability is a feature: it's what keeps the links editorial.

🔄 Check Your Understanding Two outreach emails go out. Email 1: sent to 800 site owners, identical text, "please add my link." Email 2: sent to 12 named local reporters, each referencing a specific article they wrote, offering relevant local data. Email 1 gets 1 link (a paid one); Email 2 gets 3 links (editorial). Which campaign was more successful, and why is the raw send count the wrong scoreboard?

Answer

Email 2, decisively. It earned three editorial links — freely given, from relevant sources — versus one paid link that is a scheme liability Google may devalue or penalize (Chapter 24, 26). Send count measures effort spent spamming, not value earned; the right scoreboard is quality referring domains earned, and by that measure the 12-email campaign beat the 800-email one while also not putting the site at risk.

23.6 Measuring link-earning: referring domains over time

Link earning is slow and uncertain enough that you must measure it, or you will either quit too early (because you can't see progress) or fool yourself with the wrong number (because it's flattering). The good news: there is one metric that matters most, and it's conceptually simple.

Track referring domains over time. A referring domain (Chapter 22) is a unique website that links to you at least once; it is a far better measure than raw backlink count because one site linking to you a hundred times is roughly one vote, not a hundred. What you want to watch is the count of unique referring domains, plotted month over month, trending up. That single line — are more distinct, relevant sites vouching for me than last quarter? — is the closest thing link earning has to a north star.

FIGURE 23.6 — "Referring domains, month over month"             [constructed teaching example]

  unique referring domains
   60 │                                          ╭──● 58
      │                                     ╭────╯
   40 │                          ╭──────────╯
      │              ╭───────────╯   ▲
   20 │   ●──────────╯               │
      │  18 (baseline)         data report published here — links
    0 └──────────────────────────────────────────────────────────▶
        Jan   Mar   May   Jul   Sep   Nov      (illustrative numbers)

  The shape is the lesson: flat for months, then a step up after a real asset
  lands and gets cited, then slow accrual as others cite the citers. Earned
  links arrive in irregular steps, not a smooth line — and almost never fast.

Read the shape, not the specific numbers. Earned-link growth is lumpy: long flat stretches punctuated by steps when an asset lands, followed by slow secondary accrual as new writers discover the now-established resource. If you expected a smooth diagonal and you see a flat line for three months, you might panic and quit right before the step. Knowing the shape is what lets you hold your nerve through the flat parts — which is most of link earning.

Alongside the trend, judge three qualities of the domains you're earning, because not all referring domains are equal (Chapter 22 taught why):

🛠️ Try It on Your Site In Google Search Console (free), open Links → Top linking sites and Top linked pages, and write down two numbers today: your total linking sites (a rough referring-domain count) and your single most-linked page. Put a calendar reminder for 90 days out to check the same two numbers. That's your baseline and your trend — the only honest way to know if link earning is working. Two caveats the pros know: GSC's link data is approximate and updates slowly, and it counts links Google has found, not necessarily all that exist. For sharper referring-domain tracking, paid tools (Ahrefs, Semrush, Moz) do it better — but they cost money, and we save the tool decision for Chapter 30. GSC is free and honest enough to start.

⚖️ Evidence Check Claim: "More referring domains cause higher rankings." The honest version: — Strong correlation, repeatedly observed: vendor studies across the industry consistently find referring-domain count among the metrics most correlated with ranking. This is about as robust as correlational SEO evidence gets. Confirmed by Google (in spirit): Google has never disputed that links are a major signal and has said so for years. — Not proven causal at the margin: the correlation is inflated because good sites earn both links and rankings for the same underlying reasons (quality, brand, relevance). Buying 100 referring domains does not reliably buy the ranking the correlation implies — because the correlation was never about purchasable links. Use referring domains as a health metric for earning, not as a dial you can crank.

Two temptations to name and refuse. First, the vanity backlink count — "we have 40,000 backlinks!" — which usually means a handful of sites linked you site-wide (in a footer or sidebar) and inflated the number; forty thousand links from three domains is three votes, not forty thousand. Second, chasing a third-party authority score (the Domain Authority or Domain Rating metrics from tool vendors, Chapter 22) as if it were the goal itself: those scores are useful estimates invented by companies that are not Google, and optimizing your effort to move a vendor's number rather than to earn relevant real links is measuring the thermometer instead of the temperature. We develop the full separation of KPIs from vanity metrics in Chapter 29; for now, the discipline is simple — count distinct, relevant, real domains, and distrust any metric that gets big fast.

23.7 The honest limits: slow, uncertain, unglamorous

Every section so far has told you what a tactic can do. This one collects, in one place, what none of them can do — because a chapter on earning links that oversold the payoff would betray the whole book.

It is slow. An asset published today might earn its first meaningful links in a month, its best links in six, and its long tail over years. Nothing here produces links this week. If you need traffic this week, link earning is the wrong tool, and paid search (Chapter 39) or existing rankings are the honest answer. Link earning is a compounding investment, and like any investment its returns are back-loaded and its early months look like failure.

It is uncertain. You can do everything right — build a genuinely original dataset, pitch the perfect reporters, write flawless outreach — and earn very little, because a bigger story broke that week, or the outlets that would cite you happened not to, or the topic simply didn't catch. There is no input you can guarantee produces an output. You are improving your odds, meaningfully, but you are not buying a result, and anyone who quotes you a guaranteed number of links per month is either lying or selling schemes.

It is unglamorous. The real work is 95% "make something genuinely good and useful" and 5% "tell the right people politely." It is not the hacker-genius fantasy the industry sells. It is closer to being a good reference librarian than a growth wizard: patient, useful, and mostly invisible.

And it can be actively harmful when faked. The dark mirror of everything in §23.2 is the fabricated "study" — a survey of fifty people dressed up as authoritative research, or an "index" built from numbers someone made up — pitched hard enough that credulous outlets cite it. It can even work in the short term. But it fails the book's ethics twice over: it pollutes the information commons, and it torches your credibility the moment a journalist or a competitor checks the methodology and finds nothing there. Real data earns durable links and trust; fake data borrows both and pays them back with interest. Chapter 13's guardrails on originality and Chapter 24's on manipulation both apply here — the honest version is the only one that compounds.

🔗 Connection Because link earning is slow and uncertain, it must never be your only source of traffic — the argument for traffic diversification (email, direct, social, brand) runs through Chapter 6 and comes to a head in Chapter 36, where AI Overviews make single-source dependence riskier than ever. A useful, genuine linkable asset is also, not coincidentally, exactly the kind of authoritative, attributable content that AI systems tend to cite (the "GEO" discussion of Chapter 36) — another reason the earned path ages well while the built path ages badly.

The limits are not a reason to skip link earning. They are the reason to start early, hold expectations honestly, and never bet the whole business on it. Slow, uncertain, and unglamorous is still the only authority strategy that lasts — and "lasts" is the whole game.

📈 The Strategy File

Rivertown Home Services reaches the point in its rebuild where it needs authority, and Chapter 22 gave the sober baseline: a thin link profile, mostly branded and local, plus the inherited spammy links the previous "SEO guy" left behind (we decide what to do about those in Chapter 26). The founders — Marisa and Tony Delgado, running the company their father Ray started in 1984 — have heard the pitch a hundred times: pay a few hundred dollars a month and "we'll build you links." This chapter's Strategy-File task is to reject that and design the honest alternative: one linkable asset only Rivertown could make.

The asset writes itself once you ask the §23.2 question — when would someone need to reference this? Every winter, local reporters, realtors, and "cost of living in the metro" writers need a number for what home repairs cost around here, and there is no authoritative local source. National aggregators publish generic "cost of furnace replacement" ranges, but nobody has the Rivertown-metro figures — except Rivertown, which sits on thousands of real invoices across five cities. That is a proprietary vantage point, and it is the seed of original data.

📄 Read the Report

text FIGURE 23.7 — "Rivertown's linkable asset: the annual metro cost report" [the Strategy File] THE ASSET "Rivertown Metro Home-Services Cost Report 2026" — real, anonymized, aggregated price ranges for the jobs people search most (AC repair, furnace replacement, water-heater replacement, panel upgrade, drain cleaning), broken out by all five cities: Rivertown, Cedar Hills, Northgate, Westbrook, Millhaven — built from Rivertown's own completed jobs. WHAT MAKES IT (1) ORIGINAL DATA no one else has at the local level → citation obligation. LINKABLE (2) A stated METHODOLOGY: sample size, date range, "aggregated from N completed jobs," so it's defensible and trustworthy (not a made-up "study" — see §23.7). (3) E-E-A-T baked in (Chapter 5): real company, real address, named licensed technicians, the Delgados on the record — Experience you can point at. (4) A clear local angle Rivertown can actually WIN, where national aggregators can't. WHAT IT SETTLES It gives Rivertown, for the first time, something a journalist or realtor MUST cite — an earned-link engine that compounds each year the report is updated. WHAT IT DOESN'T It earns NOTHING on its own. It needs promotion (the outreach + local-PR plan is Chapter 24), it needs measurement (referring domains in GSC — Chapters 27, 30), and it will take MONTHS, not weeks. It also won't outrank aggregators for generic national "cost of furnace replacement" — only for the LOCAL query, which is the point. THE LESSON Rivertown's most linkable asset was hiding in its invoices the whole time. Earning links started with having something worth citing — not with buying anything. (All figures/data illustrative; the strategy — package proprietary data as a local reference — is real.)

What this adds to the running strategy, and what it deliberately leaves for later. We have designed the asset: its data source, its structure (city-by-city), its methodology discipline, and its honest local scope. We have not built the outreach and local-link plan that turns the asset into actual links — that is Chapter 24 (§24.5's local sponsorships and supplier links, plus the digital-PR push for this exact report). We have not touched the inherited spammy links (Chapter 26). And we have not promised Rivertown a single link or a ranking — only a genuinely referenceable asset and honest odds. That restraint is the strategy: give the company something worth citing, then, in the next chapters, reduce the friction between "worth citing" and "cited."

Your Strategy-File task (on your own site, using Appendix C's link tracker): answer the citation-obligation question for your business — "what data, tool, or definitive resource could I publish that a writer in my field would have to reference?" Write one concrete asset idea and the one proprietary vantage point that makes it yours. Don't build it yet. Name it. That single sentence is the seed of every link you will ever earn.

Conclusion

We started where Chapter 22 left off — links still matter most off-page — and turned the practical question, how do you get them?, on its head. You do not build durable links; you earn them, by creating things worth citing and making sure the right people can find them. That reframe organized the whole chapter: the linkable asset (original data above all, then definitive guides and free tools) is the thing that deserves links; digital PR and newsjacking put it in front of journalists; journalist-request platforms let reporters come to you; outreach — real, relevant, personal, low-volume outreach — makes the right individual aware; and referring domains over time is how you prove, honestly, that the slow thing is working.

We were equally clear about what none of it can promise. Earning links is slow, uncertain, and unglamorous; it improves your odds without guaranteeing a single link, let alone a ranking; and its faked shortcut — the fabricated "study," the mass-email blast, the bought link — is worse than doing nothing, because it borrows trust it can't repay. The myth that "you can just email people and they'll link to you" survives only because the people who believe it skip the ninety-five percent of the work that is being worth linking to. Do that ninety-five percent, and the last five percent — the polite email to the right person — starts, occasionally, to work.

That is theme 5 of this book, made concrete: bought authority collapses when the money stops; earned authority compounds while you sleep. In Chapter 24, we get tactical — a frank, one-by-one map of the link tactics: guest posting done right versus link farms, resource-page and broken-link building, unlinked mentions, local and relationship links, and the black-hat schemes we cover only so you can recognize and avoid them. The asset you designed here is what makes those tactics work honestly; without it, they're just more efficient spam.

→ Continue to Chapter 24: Link Building Tactics — What Still Works, What's Penalized.

Key Terms

• Link earning — acquiring links as the natural consequence of publishing something others independently choose to reference; the durable alternative to "link building."
• Linkable asset — a page or resource created specifically to be referenced by others (original data, a definitive guide, a free tool), not merely to rank or convert.
• Original research — data you gathered that no one else has, which creates a citation obligation because writers must credit the source to use the figure.
• Digital PR — earning links and mentions from journalists and established publications by giving them something genuinely newsworthy, usually original data or expert commentary.
• Newsjacking — inserting your expertise or data into a breaking news story while it's still developing, so a journalist covering it cites you; rewards speed and relevance within a short window.
• HARO (Help A Reporter Out) — the archetypal journalist-request platform (2008–2024); now defunct as a product, but its name persists as the generic term for the category (Qwoted, Featured, SourceBottle, and others).
• Outreach — contacting a relevant person to make them aware of a genuinely useful asset; effective only when low-volume, personalized, and led by the recipient's benefit — not mass email.
• Editorial mention — a reference to your business in someone's editorial content, whether or not it carries a followed link; valuable for brand, entity signals, and downstream links even when the link itself is nofollowed (compare the editorial link of Chapter 22).

Spaced Review

Retrieval practice. Try each before revealing the answer.

1. In one sentence, explain the difference between building a link and earning one — and why only the second compounds.
2. Name the linkable-asset type that is hardest for competitors to copy, and state the one-word reason data earns links that prose doesn't.
3. (From Chapter 22.) Why is referring domains a better measure of a link profile than raw backlink count? Use the "one site links you a hundred times" example.
4. (From Chapter 22.) A link from a high-authority, topically relevant page is worth more than a link from a random low-quality one. Name two of the four factors from Chapter 22 that determine a link's value.
5. (From Chapter 5.) When answering a journalist-request platform query, which two components of E-E-A-T is the reporter most directly checking for when they pick you as a source, and how do you demonstrate them?