The problem this solves
PR targeting still runs on domain authority, readership stats, and relationships - measures of human reach. AI answer engines have quietly created a second axis: some publications get cited in machine answers constantly, others with equal prestige almost never. Coverage in a machine-read publication keeps working long after publication day, resurfacing in answer after answer. No media database tells you which outlets those are, because the only way to know is to ask the engines and count.
The result is PR budgets optimized for a distribution channel that is shrinking relative to the one nobody measures. A placement choice between two similar outlets is now a citation decision, and teams make it blind. The fix is empirical: run your category's real buyer questions through the engines, tally which publications the answers actually cite, and aim outreach - with real editor contacts - at the outlets machines demonstrably trust.
How the mission runs
- Test the category's questions. AI Visibility runs a broad panel of your category's buyer questions through the live answer engines and captures every response with its citations. Breadth matters here - the goal is a stable ranking of sources, which takes a wide sample of questions to produce.
- Rank the cited publications. The agent tallies citations by publication across all captured answers, producing an evidence-based ranking: which outlets the engines rely on for your category, how often, and for what question types. Outlets you assumed mattered often rank surprisingly low, and vice versa.
- Build the outreach targets. For each high-ranking publication, the agent identifies the editors and writers behind the cited pieces - the people who cover your space in the formats engines quote. Email Finder locates and verifies their addresses, so the list is actionable the moment it arrives.
- Ship the CSV. The target list lands as a CSV: publication, citation frequency and contexts, the specific cited articles, contact name and role, verified email, and a note on what each outlet's cited coverage has in common. It loads directly into whatever outreach flow you already run.
The prompt
This is the exact objective the agent receives. Swap the obvious placeholders for your own domain, segment or channel and run it as-is from the console, Slack, or the API.
What comes back
A CSV ranking the publications that AI engines actually cite in your category - with citation counts, the question types each outlet wins, links to the cited articles, and verified editor contacts for each - ready to import into your outreach tooling. The captured answers behind the ranking stay attached to the mission run, so the methodology is auditable end to end. This mission builds the list; the pitching stays yours.
Make it yours
- Segment the ranking by question type - educational, comparative, pricing - since different outlets win citations at different stages of the buyer's research.
- Re-run before each PR cycle to catch new outlets entering the citation mix, which often signals an engine source refresh worth exploiting early.
- Add your existing media list as an input and ask for a gap analysis: which relationships you already hold at machine-read outlets, and which effort is going to outlets the engines ignore.
Frequently asked questions
How many questions does a reliable ranking need?
Enough for publications to accumulate meaningful citation counts - a broad panel across your category's themes, typically dozens of questions. The mission reports sample sizes alongside the ranking, and outlets with thin counts are marked provisional until repeated runs confirm their standing.
Are the email addresses verified?
Email Finder locates each contact and verifies deliverability before the address enters the CSV. Contacts that cannot be verified are included with a flag and without a guessed address, so your sender reputation never rides on a fabricated email. The verification status ships as its own column.
Does citation frequency really translate to PR value?
A citation places your story inside the answers buyers read at decision time, and cited articles keep resurfacing across many questions for months. The ranking measures observed machine trust in each outlet; paired with your judgment about human audiences, it makes placement decisions two-eyed.