Digital PR for links: an honest playbook

There are two kinds of link building. One is an industry of guest-post farms, "skyscraper outreach" templates, and directory submissions that Google's spam updates spent the last three years publicly dismantling. The other is digital PR: making something genuinely worth citing, then telling the people who cite things for a living. This playbook is about the second kind — what we run for clients, what it actually costs, and the conversion rates nobody puts in the case study because they're humbling.

The honest framing first: a good digital PR campaign earns links from a low single-digit percentage of the journalists you pitch. If that number makes the program look hopeless, the barbell holds: one link from a national publication or a topically authoritative outlet outweighs fifty directory links, and the compounding effect of a few genuinely strong referring domains on a mid-authority site is visible in rank movement within a quarter. It's a low-hit-rate, high-value channel, like good hiring.

Asset first, outreach second — never the reverse

Every failed PR-for-links program we've inherited made the same mistake: it wrote the pitch first and reverse-engineered an asset to justify it. Journalists detect that shape instantly, because their inbox is full of it. The programs that earn links make something a person would cite without being asked, and then make sure the right people see it. Four asset types actually pull:

Original data studies. Your product or client's domain yields data nobody else has: a budgeting app aggregates anonymised savings behaviour; an e-commerce platform publishes delivered-vs-promised shipping times by carrier. The bar is a finding, not a dataset — "Nightside nurses check work email at 3am at twice the rate of day-shift staff in our sample" is a story; a table of forty columns is not. Methodology section, honest sample caveats, an embargo press window for tier-one outlets, and a press page with charts journalists can lift. The convert rate on well-targeted pitches of a real finding runs high single digits — exceptional for this channel.

Expert commentary at speed (reactive PR). Journalists on deadline need a quotable expert within hours. Winning is a logistics problem, not a creativity one: a standing bench of two or three named experts with pre-approved positions on their topic, a monitoring habit over morning news in the client's industry, and a 90-minute response SLA. Two tight paragraphs, a specific stance ("this rate change will hit first-home buyers in apartments hardest — here's why"), credentials, done. Most clients comment too slowly and too blandly; speed is the entire game.

Tools and assets worth citing. A genuinely useful free calculator, a salary benchmark, a public dataset, a style guide other teams steal — these earn links passively for years. Slower, less headline-worthy than a data study, but the compounding is real: the evergreen asset you ship this quarter is still earning citations when the news cycle has forgotten every pitch you sent. Programmatic-flavoured asset thinking applies here: publish only what deserves to exist.

Contrarian research with skin in the game. "We analysed 40,000 of our own landing-page A/B tests and found the industry's favourite pattern underperforms" works because it's falsifiable and costly to fake. Base-rate contrarianism ("everything you know about X is wrong!") without proprietary data is just opinion, and journalists can smell the difference by the second paragraph.

The outreach mechanics that actually matter

Lists are the work. Sixty well-researched contacts — right journalist, right beat, recent evidence they cover this shape of story — outperform six hundred scraped names by every measure we track. Read what they wrote last month and say something true about it. Templates get filtered; personalisation gets read. The pitch email itself is five sentences: the finding, why their readers care, the expert available on deadline, the link, the embargo terms if any. Attachments never.

Follow up once. Once. A polite nudge four business days later recovers a meaningful fraction of busy-week misses; a second follow-up converts a "not this time" into a "not ever". This channel runs on reputation, and reputation accrues over quarters. The agency that spams a newsroom on your behalf isn't spending your budget; it's spending your domain's name.

The pitch is a service, not a request. The journalist's job is to serve readers; your data's job is to make their story better and faster to file. Pitches framed as "coverage for our launch" fail; pitches framed as "a finding your readers will argue about, with a quotable expert and ready charts, embargoed for you exclusively until Wednesday" get answered. Exclusivity windows for tier-one outlets, then a wider release: stagger, don't blast.

Measuring without lying to yourself

Count what counts. Referring-domain quality (a national news link and a "submit your site" directory are not both +1), link placement editorially within the article (not a company-directory page), relevance of the linking page's topic to yours, and downstream rank movement on the target pages — not a DA chart presented as a business result. Expect the impact to land in months, not weeks, in line with what we say about paid vs organic timelines: PR-for-links is organic moat building, and moats fill slowly.

A worked example from our own fictional universe: for Meridian Climate, we published "The Suburb Heat Divide" — a study correlating tree-canopy coverage with summer peak temperatures across 40 council areas, drawn from open sensor networks, with an interactive explorer journalists could embed. Three national outlets ran it inside a fortnight; the interactive earned links from 30+ domains across planning blogs, university departments and local news over the following year. The full engagement write-up — clearly illustrative, as always — lives in the Meridian Climate case study. The pattern generalises: the interactive asset out-earned the press release by an order of magnitude.

What this doesn't replace

Digital PR builds authority; it does not fix a technically broken site, thin content, or a product nobody wants to talk about. It sits beside technical hygiene and a content strategy that compounds — the three legs of the same stool. A campaign earning major-outlet links to a site that can't crawl its own pages is pouring water into a colander. Sequence the work.

Key takeaways

  • Assets earn links; pitches just deliver them. Build the data study, benchmark, tool or interactive first, and never write the pitch before the finding exists.
  • Reactive expert commentary is a logistics game: named experts, pre-approved stances, 90-minute SLA. Speed beats eloquence.
  • Sixty well-researched contacts beat six hundred scraped ones; five-sentence pitches, one follow-up, embargo tiers for the outlets that matter.
  • Measure referring-domain quality, editorial placement, topical relevance and downstream rank movement — months, not weeks. Anyone reporting DA gains in week two is reporting the weather.
  • Fund it as moat-building beside technical SEO and content, not a replacement for either.

FAQ

How much should a data study cost? A properly instrumented study — data wrangling, analysis, write-up, charts, press page, outreach — runs roughly the budget of three to five standard content pieces. If your data is already clean, less; if the "study" is a survey of 200 people on the internet, you're paying for a press page and shouldn't bother. The asset's durability is the amortisation.

Do we need newsworthy data, or can we buy a survey? Both can work, but only the proprietary kind is defensible. A commissioned survey earns links until every competitor commissions one; your own platform data earns links only you can earn. Survey or not, the finding has to survive a journalist's "so what?"

How do we find the right journalists? Work backwards from coverage: who wrote the last three stories adjacent to your finding, at which outlets, on what cadence. Tools help build the canvass list; the final sixty are chosen by actually reading. A media database without reading is a spam cannon with a subscription fee.

What about link reclamation — asking sites that mention us to link? High hit-rate, low effort, do it monthly. Unlinked brand mentions, dead redirect chains from old campaigns, and image reuse without credit are the three buckets. It's maintenance, not strategy — but it's the cheapest authority you'll ever earn.

Can a small brand without big data do digital PR? Yes — small brands win on commentary speed and specificity. A niche operator's tight, quotable read on their sliver of the industry beats a generic take from a big name, because journalists filing on deadline quote whoever answered with substance first. Expertise is portable; data is rented.


Keep reading