Audience Development
Rankings Are a Flywheel: How Share-Worthy Data Replaces Link Building
Two-thirds of web pages have no backlinks, and referring domains remain the strongest ranking correlate there is. Rankings solve both problems at once: the ranked have a reason to share, the local press has a reason to write, and each release compounds the last. The mechanics, the methodology rules that make rankings defensible, and what one release looks like at scale.
By Josh McWilliam, Co-Founder
Every publisher knows the two facts that make organic growth hard. First, links still decide who ranks. Second, nobody wants to give you one. Traditional link building fights the second fact with volume — templated pitches to people who receive hundreds of them — and the returns have been falling for a decade.
There is a category of content that inverts the problem. When you publish a ranking, the entities you rank have a reason to talk about you. The university that placed first, the hospital in the top ten, the employer named best in its state — each has a communications office, a newsletter, a local reporter who covers it, and an audience that wants to hear it. You do not have to ask for the link. You have to give them something worth linking to. Done once, that is a campaign. Done systematically, it is a flywheel.
1. Why links still matter, and why almost nobody has them
Ahrefs’ analysis of roughly one billion pages in its index found that 66.31% had no backlinks whatsoever and another 26.29% had links from three or fewer referring domains. Fewer than one page in thirteen had links from more than three sites. The same body of research finds the number of referring domains to be the factor most strongly correlated with a page’s organic traffic — and that most pages with no referring domains get no traffic at all, which is a large part of why 96.55% of pages receive no Google visits.
Distribution of pages by number of referring domains (Ahrefs Content Explorer, ~1 billion pages)
Source: Ahrefs search traffic study, 2020 edition (~1 billion pages in Content Explorer), as reported by Search Engine Roundtable.
Links matter more, not less, because search is winner-take-all. Compiling click-through data across our own sites’ keywords, the top two results captured roughly 50% of clicks, the top three 62%, the top five 73%, and the top ten 89%. Below the fifth position, a page is present but effectively unvisited. The links that move a page from eighth to third are worth more than any amount of content added to a page that stays eighth.
Cumulative share of clicks captured by search position
Source: Media Factual analysis of click-through rates across keywords for the sites it operates; compiled for internal investment analytics.
2. The mechanics of the flywheel
The flywheel has five stages, and every one of them can be systematized.
Publish a ranking worth being in. The ranking must be something the ranked would be proud to display: independent, data-driven, with a published methodology and a plausible claim to fairness. That is a content decision, but it is also the whole outreach strategy. Nobody shares a ranking they are embarrassed by.
Give the ranked something to share. A badge they can put on their site, a media kit with the numbers behind their placement, pre-written social images, a short article they can republish. The easier it is to say “we ranked,” the more often it is said.
Tell the specific people who care. Not “the university” — the nursing department that just ranked, the communications office, the alumni magazine, the local newspaper that covers the campus, the high-school counselors who advise its applicants. Each of those is a distinct micro-audience with a distinct reason to link.
Turn every release into news. Rankings that update annually — or whenever the data changes — are a recurring news event. The story is the change: who moved up, who is new, what the data says this year. Each story links to the permanent ranking page, which is where the authority accumulates.
Let authority lift the cluster. Links to the ranking lift the entity profiles it links to, the comparisons beneath them, and the head-term pages above them. Next year’s release starts from higher ground, which earns more coverage, which earns more links.
3. What it produced
The flagship site built on this model — College Factual — publishes thousands of rankings by major, degree level, location, and audience, refreshed as the underlying data updates. By 2022 it had accumulated 3.79 million backlinks from 19,107 referring domains and a domain rating of 69, with 11,374 keywords in Google’s top three. Hundreds of colleges publish the rankings they have earned on their own sites, which is the flywheel’s most visible output: an institution’s own domain vouching for yours, unprompted, every year.
The model transferred. A ten-site network we built and operate for a partner in the same vertical, launched from zero a few years earlier, had reached 173,028 backlinks from 8,328 referring domains by the same measurement — without the flagship’s decade of accumulated authority, running on the same ranking releases and the same outreach system.
Referring domains accumulated, 2022 measurement
Source: Ahrefs Site Explorer, 2022, as compiled for Media Factual's internal reporting.
4. What one release looks like at scale
The difference between a campaign and a flywheel is whether the work is repeatable without a proportional increase in people. Here is one release, for one ranking, on one of the network’s sites — a single nursing-related major:
- 9,000 recipients, each receiving an individually personalized message generated from the same data as the ranking (their institution, their placement, their cohort comparison).
- Seven distinct audiences — the ranked department, institutional communications, local media near each campus, counselors, and others — each with its own message template and its own reason to care.
- Custom media kits per institution, plus auto-generated social images.
- Delivery at a natural cadence: throttled, randomized sending windows, per-organization caps, and persona management, which is why deliverability held above 80% on cold outreach.
Behind that one release is a prospecting system that at last count maintained 132 audience definitions across 42 roles, 121,000 organizations, and 921,000 individual prospects — refreshed and re-validated on a schedule — and a publishing calendar with more than 850 scheduled releases, each with its campaigns attached. That is how “10+ campaigns a month” becomes a routine rather than a project, and how, between 2019 and 2021, the team went from one or two campaigns a year to ten or more a month while shrinking.
5. The rules that make a ranking defensible
A ranking earns links in proportion to how much the ranked trust it. Six rules, learned the hard way:
- Publish the methodology. Every factor, every weight, every data source. Institutions will scrutinize it; journalists will ask; the ones who rank well will cite it as evidence of rigor.
- Rank outcomes, not reputation. Completion, earnings, placement — things that happened, from public data — rather than surveys of who is admired. Outcomes are harder to argue with and more useful to the reader.
- Include everyone. Every accredited institution, every program, not the ones who answered a questionnaire. A ranking that omits half the field is a ranking half the field can dismiss.
- Go bottom-up. Rank the program, then roll up to the institution. A school is a collection of its programs, and the program-level ranking is the one the department shares.
- Update on the data’s schedule, and say so. A ranking with a visible vintage is a ranking that can be trusted this year and anticipated next year.
- Never sell the position. Badges and media kits can be licensed after the fact; a placement can never be bought or changed by a payment. The moment a ranking is for sale, its link value goes to zero, and everyone can tell.
Methodology and caveats
Public figures. The backlink distribution is from the 2020 edition of Ahrefs’ search traffic study (~1 billion pages in Content Explorer), which is the most recent edition to report it; the 2023 edition (~14 billion pages) reports the 96.55% zero-traffic figure and the referring-domain correlation but not the distribution. Ahrefs’ index over-represents English-language and higher-quality pages.
Our figures. Backlink and referring-domain counts are Ahrefs Site Explorer measurements from 2022, compiled for internal reporting; they are point-in-time and both sites have changed since. The click-share curve is our own compilation across keywords for sites we operate and will differ by vertical and by results-page layout, particularly since AI Overviews. Campaign figures (recipients, audiences, deliverability) are from our outreach system’s records for one release; prospecting-system totals are as of 2022. The partner network is not named.
What to do about it
- Find the ranking your vertical is missing. Most categories have a reputation-based ranking and no outcomes-based one. The gap is the opportunity, and the data usually exists in public sources.
- Build the release kit before the release. Badges, media kits, social images, and a republishable article, generated from the same data as the ranking. If the ranked have to make their own graphics, most will not.
- Map the micro-audiences. For each ranked entity, list the five kinds of people who would want to know, and build a prospect list for each. This is where the work is, and where the return is.
- Make the ranking page permanent and the release a story. One URL per ranking, updated in place; a new article each time it updates. Links go to the permanent page; news goes to the story.
- Automate the cadence. Throttling, personalization, caps, and follow-up are what keep deliverability high and your domain off blocklists. They are also what make ten releases a month possible.
- Protect the methodology. Publish it, defend it, and keep commerce out of it. Trust is the asset; everything else is a multiplier on it.
Sources
- Search Engine Roundtable — Over 65% of Web Pages Have No Backlinks (reporting Ahrefs' 2020 search traffic study) — 66.31% zero backlinks; 26.29% three or fewer referring domains; ~1 billion pages
- Ahrefs — 96.55% of Content Gets No Traffic From Google (Soulo, December 2023) — Referring domains vs. traffic correlation; ~14 billion pages
- Ahrefs — How long does it take to rank in Google? (Stox, May 2025) — Age of top-ranking pages
- Google Search Central — Spam policies: link spam — Why paid placements and link schemes are excluded from the model
- Media Factual — internal reporting: backlink profiles (Ahrefs Site Explorer, 2022), click-share compilation, outreach and prospecting system records
- rankings
- backlinks
- link building
- outreach
- digital PR
- authority
- badges
- media kits