The 2026 framework for turning a defensible point of view into a primary source journalists, analysts and AI engines have to cite — with the Category POV Score and worked teardowns of the three brands dominating the play.

Most teams treat editorial links as an outreach problem. The logic is intuitive: pitch more journalists, send more emails, follow up more often, and the links accumulate. The data points the other way. The brands that earn the most high-authority editorial links in B2B do not run the most outreach — they run the fewest pitches per link, because they have made themselves the thing everyone else has to cite.

Here is the number that should reframe the entire strategy. The research firm Play Bigger — whose 2016 book popularised the discipline of category design — found that the company which names and owns a market category captures roughly 76% of that category’s total market capitalisation, leaving every competitor to fight over the remaining 24%. Editorial coverage follows the same power law. When one brand owns the language of a problem space, journalists writing about that problem cite the brand that named it. The link is not won. It is inherited, over and over, by the category owner.

That is the strategy this article is about. Not a campaign, not a tactic, but a category point of view (POV): a named, evidence-backed thesis about how a market works that you publish on a repeatable schedule until you become its primary source. Done well, it produces the most durable editorial links in existence — links that compound for years and that AI search engines re-cite long after the campaign that earned them is forgotten. Done badly, it is an expensive vanity report nobody references.

This is the strategy-layer companion to the tactical playbooks elsewhere on this site. If you want the catalogue of acquisition methods, start with the 15 link building strategies that actually work in 2026. This piece sits one level above that: it is about building the asset that makes those tactics dramatically more efficient.

The deliverable first: the Category POV Score (CPS)

Before any of the reasoning, here is the tool. The Category POV Score (CPS) is a 0–100 diagnostic that tells you how citable your category POV actually is — and therefore how many editorial links it is structurally capable of earning. You can score your own brand against it in under an hour. You can score a competitor against it in twenty minutes. It is the spine of everything that follows, and we will use it to grade the three dominant players later in the article.

The CPS rests on a simple premise borne out by the data: editorial links accrue to named, evidenced, repeated claims, not to opinions. It scores five dimensions, each from 0 to 100, then weights them.

The five dimensions

1. Named Thesis (NT) — weight 30%. Do you have a defensible, ownable label for your point of view — a category name, a coined term, a proprietary phrase? “Inbound Marketing” and “Revenue Intelligence” score high; “we help sales teams” scores zero. Without a name there is no category to own, which is why this carries the most weight.
2. Proprietary Evidence (PE) — weight 25%. Can you produce data that no competitor can replicate? First-party platform data and original surveys score highest; re-packaged third-party statistics score lowest. Proprietary evidence is what makes you the primary source rather than a secondary one.
3. Repetition Cadence (RC) — weight 20%. Is the POV expressed through a scheduled, recurring publication — an annual report, a quarterly index, a monthly data drop? A one-off study is a campaign. A repeated study is an institution, and the second edition earns more links than the first because journalists already have context.
4. Citation Surface (CS) — weight 15%. Is the asset built to be cited? Ungated, with clearly attributed statistics, a stable canonical URL, quotable sentence-level findings, and a methodology note. A gated PDF behind a form scores low because journalists will not fight a lead-capture wall to cite you.
5. Distribution Spread (DS) — weight 10%. Is one study atomised into many surfaces — a press brief for journalists, an ungated post for search and AI, a LinkedIn data series, a sales-enablement extract? One study should do five jobs. This is the lowest weight because spread amplifies a strong POV but cannot rescue a weak one.

The formula

CPS = (NT × 0.30) + (PE × 0.25) + (RC × 0.20) + (CS × 0.15) + (DS × 0.10)

Score each dimension honestly from 0 to 100, apply the weights, and read your total against these bands:

CPS	Verdict	What it means for editorial links
0–40	Positioned player	You compete inside someone else’s category. Links must be won one pitch at a time. Outreach is your only lever.
41–70	Category contender	You have a thesis but lack repetition or proprietary evidence. Links arrive in bursts around each asset, then decay.
71–100	Category king	You are the primary source. Links accrue passively, compound annually, and get re-cited by AI engines for years.

If you score below 41, the rest of this article is your roadmap. If you score 41–70, the gap between you and a category king is almost always repetition and proprietary evidence — the two dimensions teams most often skip. The build sequence later in this piece closes both gaps.

A worked example makes the scoring concrete. Imagine a four-person invoicing SaaS for freelancers. Today it publishes generic “how to get paid faster” blog posts — Named Thesis 10, Proprietary Evidence 0, no recurring asset: a CPS in the low teens, a textbook positioned player. Now suppose it names a thesis (“The Late-Payment Economy”), mines its own anonymised platform data on how long freelancer invoices actually take to clear by industry and country, and commits to publishing it as an annual index. Even with a modest first edition — Named Thesis 70, Proprietary Evidence 85, Repetition 40 (one edition, calendar set), Citation Surface 75, Distribution 60 — the weighted CPS lands around 67. That is the leap from “links only via outreach” to “contender that journalists start citing,” achieved by one four-person team with data it already owned. The constraint was never scale; it was the decision to name a thesis and instrument the data.

What the data shows vs. what practitioners believe

The link building industry has a near-religious belief that editorial links are an outreach numbers game. The belief is not baseless — at the bottom of the CPS scale it is literally true. But it describes the experience of positioned players, not category kings, and the two operate under completely different physics.

What practitioners believe	What the data shows
More editorial links come from more pitching.	Above CPS 70, links arrive without pitching. Journalists cite the named source; you inherit links from articles you never touched.
A great one-off study earns the links.	The second edition of an annual study out-earns the first, because the story already has context. Repetition, not novelty, compounds.
Journalists want a compelling opinion.	Around 40% of journalists actively want original data or research with a pitch (Muck Rack, 2026). Opinion is ambient noise; data is a primary source.
Editorial links are a PR-team activity.	The highest-yield editorial links are a by-product of a content+research asset the SEO and content teams own, with PR as distribution.
Gating the report protects its value.	Gating kills the citation. The ungated version drives links and AI citations; the gated version captures leads. You need both, not one.

The single most load-bearing figure here is the Muck Rack one: roughly 40% of journalists want original data alongside a pitch. This site has covered that statistic before because it changes the job description — it means the most valuable thing a link builder can build is not a pitch list but a dataset. The broader set of benchmarks (reply rates, pitch-to-coverage conversion, cost trajectories) lives in the link building statistics 2026 reference. The takeaway for category POV is narrow and decisive: own the data, and you change which side of the editorial inbox you sit on.

Why a category POV earns editorial links (the mechanism)

A category POV is not a branding exercise that happens to attract links. It is a link-earning machine whose mechanics are specific and repeatable. Three dynamics do the work.

1. The primary-source dynamic

Journalists, analysts and bloggers need primary sources. When your study is the origin of a statistic, every article that quotes that statistic creates an inbound link to your domain. You are not earning links by being persuasive; you are earning them by being the citation. This is structurally different from a guest post or a broken-link swap, and it scales without proportional effort. One well-placed statistic can be cited hundreds of times by writers you have never contacted.

2. The category-definition dynamic

When you publish data about a problem, you implicitly claim authority over that problem. Name the category — “Inbound Marketing,” “Revenue Intelligence,” “Conversational Marketing” — and you give analysts and journalists a mental shortcut they will use for years. Once the label sticks, every feature, round-up and explainer about the space has to mention the category, and the brand that coined it is the natural anchor link. You have effectively written the headline of every future article about your market.

3. The AI-citation (GEO) dynamic

This is the 2026 multiplier. Generative engines — Google’s AI Overviews, Perplexity, ChatGPT Search, Claude — preferentially cite primary sources when they synthesise factual answers. A statistic attributed to your firm in a well-structured, ungated report can be cited by these engines for years, long after the press cycle ends. Category POV is therefore not just an editorial-link play; it is the most durable form of Generative Engine Optimisation available, because the same properties that make a source citable to a journalist make it citable to a language model: a clear claim, a clear number, a clear attribution, and a stable URL.

There is a second-order version of this worth planning for. AI engines do not only cite the report on your own domain; they cite the places your category POV gets discussed. In 2026 the highest-weighted AI citation sources are community and professional platforms — Reddit and LinkedIn chief among them — where your named statistic gets quoted in threads and posts. A category POV seeded into those conversations is therefore cited twice over: once at the source, and again wherever the discussion happens. That is why the distribution step below treats a LinkedIn data series not as social fluff but as a deliberate AI-citation surface.

Notice what these three dynamics have in common: none of them is outreach. Outreach still matters for the launch spike — and reactive PR tactics like newsjacking are an excellent way to inject a fresh data point into a live news cycle — but the compounding comes from the asset, not the email. This is why category POV out-performs tactics that depend entirely on volume, such as guest posting, over any horizon longer than a quarter.

Naming the category: a four-test screen

Because Named Thesis carries the highest CPS weight, the name itself deserves more rigour than teams usually give it. A weak label — descriptive, generic, or a thinly re-branded version of the incumbent category — fails to give analysts a mental shortcut, and without the shortcut there is no category to inherit links from. Run every candidate name through four tests before you commit:

• The shortcut test. Could an analyst or journalist use this phrase as shorthand for the whole problem space? “Revenue Intelligence” passes; “AI-powered sales analytics platform” does not — it describes a product, not a category.
• The enemy test. Does the name implicitly frame an old way as broken? “Inbound Marketing” only has meaning against “outbound.” A category with no villain gives journalists no tension to write about.
• The ownership test. Can you plausibly become the first and loudest voice on it, or is an incumbent already entrenched? If a category king exists, name a defensible sub-category instead of fighting for the parent term.
• The search test. Is the phrase one a buyer or journalist might actually type, or could be taught to? A name nobody searches for cannot accumulate branded or category search volume — the downstream KPI that proves the POV is taking hold.

A candidate that passes all four is worth building an entire research program around. One that fails the shortcut or enemy test will not earn coverage no matter how good the underlying data is, because journalists write about tension and shortcuts, not about feature lists.

Theory is cheap. Here are the three B2B software companies that have run this play best, scored against the Category POV Score so you can see exactly which dimensions they won on — and, in one instructive case, which one they let slip. The scores below are this article’s diagnostic assessment using the CPS framework; they are an analytical tool, not the companies’ official figures.

1. HubSpot — “Inbound Marketing”  ·  CPS 93

HubSpot is the canonical case. In 2009–2010, founders Brian Halligan and Dharmesh Shah had a thesis — that pulling customers toward you with content beats interrupting them with ads — and a problem: the idea had no name and no vocabulary that buyers could put in a budget line. So they named it Inbound Marketing, and then they built the asset that would make the name unavoidable: the annual State of Marketing report. Each edition surveyed thousands of marketers and surfaced counter-intuitive, quotable findings — every one of which was a stealth argument for inbound, made through data rather than advocacy.

That distinction is the whole game. HubSpot did not publish opinions about why outbound was dying; it published numbers that implied it, and let journalists draw the conclusion in their own words while linking back to HubSpot as the source. The report has run annually for well over a decade, which is why its CPS repetition score is near-maximal. HubSpot is now a publicly traded company (NYSE: HUBS) and one of the most-cited editorial-link domains in marketing — the kind of DR 90-plus source that everyone else tries to earn links from.

CPS dimension	Score	Why
Named Thesis	95	Coined and still owns “Inbound Marketing” outright.
Proprietary Evidence	90	Large recurring original survey across thousands of marketers.
Repetition Cadence	95	Annual State of Marketing report for 10+ years.
Citation Surface	90	Ungated stats, quotable, stable URLs, clear methodology.
Distribution Spread	95	Atomised across blog, gated PDF, social, sales enablement.
Weighted CPS	93	Applying CPS = (NT×0.30)+(PE×0.25)+(RC×0.20)+(CS×0.15)+(DS×0.10).

Lesson: the name plus the repeated study is the engine. HubSpot’s links are not a campaign output; they are an annuity.

2. Gong — “Revenue Intelligence”  ·  CPS 88

Gong, founded in 2015 by Amit Bendov and Eilon Reshef, faced a crowded field of “call recording” and “conversation intelligence” tools. Rather than fight inside that category, it coined and grouped its work under a new banner — Revenue Intelligence — and then did something HubSpot could not: it published findings drawn from its own first-party platform data. The Gong Labs research program analyses aggregated, de-identified signals from millions of real sales interactions to produce claims like “what actually correlates with closed-won deals.” No competitor can replicate that dataset, which is why Gong’s Proprietary Evidence score is the highest of the three.

The commercial result is documented. Gong raised roughly $584 million and was valued at $7.25 billion in its 2021 Series E; in 2026 it reported reaching a $500 million ARR run-rate. “Revenue Intelligence” is now the default label analysts and journalists use for the category — and Gong is the anchor citation inside it. Its CPS sits just below HubSpot mainly on repetition and citation surface: Gong Labs publishes prolifically but less metronomically than a single flagship annual report, and some of its strongest findings live inside polished assets rather than maximally citable ungated pages.

CPS dimension	Score	Why
Named Thesis	90	Coined and owns “Revenue Intelligence.”
Proprietary Evidence	95	First-party data from millions of sales interactions — un-replicable.
Repetition Cadence	85	Frequent Gong Labs output; less single-flagship rhythm.
Citation Surface	80	Strong, though some findings sit in designed assets vs. raw ungated pages.
Distribution Spread	85	Wide across blog, social, webinars, sales content.
Weighted CPS	88	Applying CPS = (NT×0.30)+(PE×0.25)+(RC×0.20)+(CS×0.15)+(DS×0.10).

Lesson: proprietary first-party data is the strongest possible evidence base. If your product generates data, your category POV should be built on it — nobody can out-cite you on your own numbers.

3. Drift — “Conversational Marketing”  ·  CPS 79

Drift’s product was, in plain terms, a chatbot — a category with hundreds of competitors. Instead of competing on features, Drift reframed the entire conversation under a new name, Conversational Marketing, and then did the most aggressive category-ownership move available: it wrote the book — literally — publishing Conversational Marketing: How the World’s Fastest Growing Companies Use Chatbots to Generate Leads in 2019, backed by an annual State of Conversational Marketing report. For several years Drift was the inescapable citation for the category it had named.

Drift is included here as much for its strength as for its cautionary arc. After its acquisition and absorption — Drift is now part of Salesloft — the publishing cadence behind the POV slowed, and the category’s gravitational pull weakened with it. That is precisely why Repetition Cadence is weighted at 20% in the CPS: a category POV is not a monument you build once, it is a fire you keep feeding. The moment the recurring evidence stops, the inherited links stop compounding and competitors begin re-litigating the category name. Drift’s lower repetition and slightly survey-dependent evidence base are what separate its 79 from Gong’s 88.

CPS dimension	Score	Why
Named Thesis	88	Coined “Conversational Marketing” and authored the defining book.
Proprietary Evidence	75	Strong annual survey, but survey-based rather than first-party platform data.
Repetition Cadence	70	Strong cadence that lapsed post-acquisition — links stopped compounding.
Citation Surface	78	Quotable report findings and a citable book.
Distribution Spread	80	Book, report, blog, events — well atomised at its peak.
Weighted CPS	79	Applying CPS = (NT×0.30)+(PE×0.25)+(RC×0.20)+(CS×0.15)+(DS×0.10).

Lesson: category ownership is a maintenance commitment, not a launch event. The brand that stops publishing the evidence eventually loses the category — and the links that came with it.

The three, side by side

Brand	Category named	CPS	Strongest dimension
HubSpot	Inbound Marketing	93	Repetition — a decade-plus annual report
Gong	Revenue Intelligence	88	Proprietary evidence — first-party platform data
Drift	Conversational Marketing	79	Named thesis — coined the term, wrote the book

Read the column of strongest dimensions and a pattern emerges: each brand won on a different CPS lever, but all three started from a named thesis and all three earned their durable links from evidence, not pitches. The differences in their scores are differences in which lever they pushed hardest and which they let slip.

Your Monday-morning build sequence: a six-week category POV sprint

This is the executable part. You do not need a HubSpot-sized budget to start; the rise of AI-assisted research means a small team can field a credible original study in about six weeks. Here is the sequence, designed so that a brand currently scoring under 41 on the CPS can reach contender status on a first cycle and king status by the second.

Week 1 — Name the thesis

• Write your category POV as a single sentence: “[Audience] is wrong about [problem]; the data shows [counter-intuitive claim].” If you cannot make it counter-intuitive, it will not earn coverage.
• Coin or sharpen the category name. Test it against the Named Thesis question: could an analyst use this phrase as a mental shortcut? If not, keep working.
• Define the one statistic you most want to own — the number you want every future article in your space to attribute to you.

Week 2 — Design the evidence

• Inventory your proprietary data first. Any first-party signal your product generates outranks any survey, because it cannot be replicated (this is Gong’s entire advantage).
• If you have no first-party data, design an original survey. Aim for a sample large enough to be credible to a journalist (a few hundred qualified respondents is a defensible floor).
• Pre-test the hypothesis cheaply before spending on a panel, then write the methodology note now — journalists and AI engines both reward a clear method.

Weeks 3–4 — Build the asset for citation

• Publish an ungated report page with a stable canonical URL. This is your citation surface. Gate a deeper PDF separately for lead capture — never gate the version you want cited.
• Write every key finding as a standalone, quotable sentence with the number and the attribution inside it. Assume a journalist will copy one sentence; make every sentence copy-ready.
• Add explicit source attribution and schema so AI engines can parse the claims. For the structured-data side of this, the link building tools guide covers the monitoring and markup stack worth standardising on.

Week 5 — Atomise for distribution (one study, five jobs)

• Press brief: a one-page summary with the three most newsworthy findings, sent to a tight, relevant journalist list — relevance beats volume every time.
• Ungated blog post: the SEO and GEO surface that earns passive links and AI citations.
• LinkedIn data series: one chart per post, drip-fed, to build the social and AI-source footprint.
• Sales-enablement extract and a gated PDF: the pipeline and lead-capture jobs.

Week 6 — Launch, then commit to the calendar

• Time the launch to a live news hook where possible, and pitch the data into relevant journalist requests as they appear.
• Then — and this is the step most teams skip — put the next edition on the calendar. A repeated study is the difference between a CPS contender and a category king. Commit to the cadence before the first edition’s links have even landed.

If you do nothing else from this article, do this: pick the one statistic you want to own, and design the smallest credible study that lets you own it. That single decision moves you from competing inside someone else’s category to building your own — and from chasing links one pitch at a time to inheriting them. For the foundational context on why links still drive authority in the first place, the complete guide to what link building is sets the groundwork this strategy builds on.

When NOT to use a category POV strategy

Format honesty matters more than enthusiasm. A category POV strategy is powerful but narrow in its fit. It is the wrong move in several common situations, and pretending otherwise wastes budget.

• You have no defensible evidence source. If you cannot generate first-party data and cannot afford a credible survey, you will end up re-packaging others’ statistics — the opposite of a primary source. Build the data capability first, or choose a different tactic.
• You are a local service business. A plumber or a regional law firm earns links from local press, directories and community relationships, not from owning a national category. The economics do not work below a certain market size.
• The category already has an entrenched king. Trying to out-publish HubSpot on “Inbound Marketing” is a losing fight. Either name a defensible sub-category you can own, or compete on tactics rather than category.
• You need links this quarter. Category POV is a compounding asset measured over years. If your mandate is a fast link-velocity boost for a specific page, reactive PR or targeted outreach will serve you better in the short term.
• You cannot commit to repetition. As Drift shows, a category POV that stops being fed decays. If you cannot resource at least a second and third edition, the strategy will under-deliver against its cost.

Measuring whether it is working

Because category POV operates above the level of individual links, the usual link-count dashboard understates it. Track these four signals together:

1. Referring domains to the asset, segmented into earned editorial citations vs. self-placed links. The ratio of earned-to-placed should climb with each edition — that is the category-king signature.
2. Unprompted statistic citations. Search your own headline number in quotes each month and count the articles that cite it without you having pitched them. This is the clearest proof you have become a primary source.
3. Branded and category search volume. As the POV takes hold, search demand for your brand and your category name should rise together. This is the downstream KPI of category ownership and the subject of its own framework on this site.
4. AI-citation share. Periodically ask the major AI engines a question your category POV answers, and check whether they cite you. Durable AI citation is the modern proof that your evidence has become canonical.

One caution on attribution: a category POV influences pipeline through channels you cannot fully track — the analyst who frames a report around your number, the buyer who arrives already using your category language. Do not let an incomplete dashboard talk you out of a strategy whose biggest effects are, by design, partly unattributable. Measure what you can, and weight the earned-citation ratio above raw link counts.

Frequently asked questions

What is a category POV in link building?

A category point of view is a named, evidence-backed thesis about how a market works that a brand publishes on a repeatable schedule. It earns editorial links because journalists, analysts and AI engines cite the brand that named and quantified the category as a primary source, rather than the brand merely participating in it.

How is this different from digital PR?

Digital PR is the distribution layer — pitching, journalist relationships, news hooks. A category POV is the asset layer that makes digital PR efficient. Digital PR without a category POV wins links one pitch at a time; a category POV gives digital PR a primary-source asset that keeps earning links between campaigns and gets re-cited by AI engines for years.

Do I need original data, or is a strong opinion enough?

You need original data. Around 40% of journalists want original research with a pitch (Muck Rack, 2026), and AI engines preferentially cite primary sources. An opinion is replaceable and unciteable; a number you own is neither. First-party platform data is the strongest base, an original survey is the next best, and re-packaged third-party statistics are the weakest.

How long before a category POV earns meaningful links?

Expect a modest launch spike from the first edition and meaningful compounding from the second edition onward. The defining property of the strategy is that links accrue for years, not weeks — which is also why it is the wrong choice if you need link velocity this quarter.

Can a small company do this, or is it only for the HubSpots of the world?

Small companies can and do run this play. AI-assisted research has cut the cost and time of fielding a credible original study to the point where a two-person team can execute one in roughly six weeks. The constraint is not budget; it is a defensible evidence source and the discipline to repeat the study.

What is the single biggest mistake teams make?

Treating it as a one-off campaign. The second most common mistake is gating the report behind a lead form, which kills the citation that earns the links. Publish an ungated, citable version for links and AI, and gate a deeper version separately for leads.

The bottom line

Editorial links are not, at the highest level, an outreach problem. They are an ownership problem. The 76% of category market value that flows to the brand which names and owns a category has an editorial twin: the disproportionate share of citations that flows to the brand which named and quantified the conversation. HubSpot, Gong and Drift each proved it — and Drift proved the corollary, that the category you stop feeding is the category you lose.

Score yourself on the Category POV Score, find your weakest dimension, and run the six-week sprint to close it. Pick the one number you want every future article in your market to attribute to you, and go build the smallest credible study that lets you own it. The links will follow — and, unlike the links you chase one pitch at a time, they will keep arriving long after you have stopped asking. For the wider tactical context this strategy plugs into, the 15 link building strategies that actually work in 2026 remains the master reference.