Get quoted where the models are reading.

Assistants trust sources with no stake in the answer. We make you one of them.

Book a callCheck your AI visibility

What gets quoted

Not this

“We’re thrilled to announce our innovative, industry-leading platform, which empowers businesses to unlock synergies at scale.”

No claim, no attribution, nothing an assistant can lift and stand behind.

This

“Teams of 20–200 switch CRMs for forecasting accuracy, not features, it’s the top cited reason in 4 of 5 migrations we handle,” says [name], [role] at [company].

A specific claim, attributable to a named person, in one paragraph that survives being extracted on its own.

The difference a paragraph makes

01 Why this exists

Your own site can only say so much

Both are press copy. Only one can be lifted into an answer and attributed, and that difference is most of the job.

There's a ceiling on what a model will take from a company about itself. Your pages establish what you do and let you be parsed correctly, but when an assistant is asked who's best, it leans on sources with no stake in the answer.

This is why technically excellent sites still don't get named. The markup is right, the content is quotable, and the model still reaches for a roundup on a publication it has cited a hundred times before.

The work, then, is placement, but placement chosen by which publications the engines actually cite for your category, not by domain authority scores designed for a different era of search. And the copy has to be written to be lifted: a quotable claim, attributed, in a paragraph that survives being extracted on its own.

02 Definition

What is Digital PR for AI citations?

Digital PR for AI citations
Digital PR for AI citations is the practice of earning placements on the publications that AI assistants already draw from when answering questions in a category. It differs from traditional digital PR in its target: outlets are chosen by how frequently engines cite them rather than by domain authority, and copy is written as short, attributable claims that survive being extracted from the article and quoted on their own.

  • From domain authority to citation frequency

    The right target is not the strongest domain. It is the publication the engines actually quote when answering your category's questions.

  • From reach to extractability

    A placement's value here is whether one paragraph of it can be lifted and attributed. Impressions do not predict that.

  • From coverage to citation tracking

    The measure is whether the placement shows up in answers afterwards, which means re-running the prompts, not counting pickups.

03 What’s included

Four things land. All of them are built, not filed.

  • 01

    Target list from the source map

    The publications the engines already cite in your category's answers, ranked by how often, so effort goes where a citation is likeliest to be read rather than where a metric says the domain is strong.

  • 02

    Copy written to be extracted

    Releases and contributed pieces built around a specific, attributable claim, in paragraphs short enough to be quoted whole. Approval-ready, so legal review isn't a rewrite.

  • 03

    Syndication that lands

    Distribution to the outlets where the pickup is indexable and citable, rather than to a wire whose copies are noindexed and invisible to the engines.

  • 04

    Citation tracking

    Whether the placement is being cited in answers afterwards, which is the only measure that matters here, and the one most PR reporting replaces with impressions.

04 How it runs

In order, and in the open.

  1. Step 01

    Map the cited sources

    Run the category's prompts, record which publications the engines draw from, and rank them by frequency.

  2. Step 02

    Build the angle

    A claim worth quoting that's actually yours to make. Data you hold, a position you can defend, or a practitioner view your team has.

  3. Step 03

    Place and syndicate

    Pitch, place and distribute, then verify the pickup is indexable rather than assuming it.

  4. Step 04

    Track the citation

    Re-run the prompts afterwards to see whether the placement is showing up in answers, and feed that back into the next target.

05 The difference

Traditional digital PR vs PR for AI citations

 Traditional digital PRPR for AI citations
Target selectionDomain authorityHow often engines cite the outlet
CopyWritten for a readerWritten to be lifted and attributed
SyndicationWire, often noindexedOutlets where pickup is indexable
Success measureImpressions and linksWhether the answer cites it

06 Who it’s for

Four markets, four different questions.

  • B2B software

    Buyers research through assistants long before a demo. The questions are comparative, alternatives, integrations, fit for team size, and the answer names two or three vendors.

  • DTC and ecommerce

    Category questions now start with an assistant. “Best X for Y” returns a shortlist, and being absent from it costs the consideration set, not just the click.

  • Healthcare

    Trust and provenance dominate. Assistants lean hard on sources they can attribute, which makes structured, citable content and clean entity signals disproportionately valuable.

  • NDIS and disability services

    Participants and coordinators ask assistants for local providers. The answers lean on directories, reviews and clearly structured service pages, which most providers have never marked up.

07 Questions

What buyers ask about this one.

Something not covered? Ask us directly and we’ll answer straight.

How do AI assistants choose sources?
They retrieve from pages they can fetch and have reason to trust, then quote what is extractable. In practice that means a relatively small set of publications gets cited repeatedly for any given category, which is why the target list is built from observed citations rather than from domain metrics.
Do press releases still work?
Wire releases mostly don't, for this purpose: the copies are frequently noindexed or duplicated, and engines rarely cite them. What works is a placement on a publication the models already draw from, written so one paragraph can be lifted and attributed.
What makes copy quotable?
A specific claim, attributable to a named person, in a paragraph that stands on its own. If lifting the sentence out of the article leaves it unintelligible, pronouns pointing at earlier paragraphs, a claim with no source, an assistant has nothing to use.
Can you guarantee coverage?
No. Editors decide, and any agency guaranteeing placements is either buying them or describing a wire release most engines ignore. We commit to the target list, the angle, the copy and the outreach, and report honestly on what landed.
How is this measured?
By re-running the category's prompts after a placement and checking whether the engines cite it. Impressions and links are the traditional measures and they don't predict this, a modest placement on a frequently cited outlet outperforms a large one the models never read.
Is this different from normal PR?
The target is. Traditional digital PR optimises for reach and links; this optimises for being retrieved and quoted, which changes both the outlet list and how the copy is written. A placement with modest traffic on a publication the models cite constantly is worth more here than a bigger one they never read.
Can you guarantee placements?
No. Editors decide, and any agency guaranteeing coverage is either paying for it or describing a wire release that most engines ignore. What we commit to is the target list, the angle, the copy and the outreach, and we report honestly on what landed.
How long until a placement shows up in answers?
It has to be indexed and then retrieved, so weeks at the fast end and a quarter at the realistic one. It also may not happen at all for any single placement, which is why the work is a programme rather than a one-off.

Next step

Start with what your site already says.

The free checker fetches your site the way the assistants do and scores what comes back. It takes a minute, and it makes the first call a conversation about your findings rather than about our process.

Book a callRun the free check

08 The rest of the work