Find [number] journalists, newsletter writers, and podcast hosts who covered [topic] in the last [number] days. For each person, verify their name, outlet, one specific recent piece or episode, why it fits our story, and a public contact route on their own site or a linked byline. Compare against [previous outreach list] and exclude [people or outlets]. Put the findings in a table with source links and checked dates. Drop entries you cannot verify and record why. Draft a short pitch that refers to each person's actual work and uses only the approved facts I provide. Hold every pitch for my review. Send nothing without my approval.
Press and podcast list with verified relevance
Research relevant writers and hosts, verify their recent work and public contact routes, and prepare outreach drafts for review.
Copy-ready brief
Example output
Example output for illustration only: Target: A podcast host who interviewed a founder about delegated work last month. Evidence: [episode URL], checked [date]. Public contact route: [show contact page URL]. Fit: The episode asked how teams review an agent's work; our approved angle addresses that question. Draft opening: “Your conversation with [guest] raised the question of who signs off before an agent sends a message. I can share a concrete approval workflow and its limits.” Status: Draft held. Contact route will be rechecked before an approved send. Dropped: One syndicated byline and one profile without a public contact route.
Trigger and cadence
- The reader defines a story, audience, and research window.
- One research pass, then a source refresh before any approved outreach.
Inputs
- Story angle, target topics, and relevant publication window.
- Target count, exclusions, and previous contact list.
- Approved product facts, tone, and public contact criteria.
Steps
- Find writers and hosts whose recent work matches the topic.
- Verify each person, outlet, piece, and public contact route.
- Check the exclusion and previous-contact lists.
- Record the evidence and drop entries that cannot be verified.
- Draft a short, specific pitch for each eligible contact.
Approval boundary
Every pitch and recipient waits for review; research alone never authorizes outreach.
Stop conditions
- A byline is syndicated or the original author is unclear.
- No public contact route can be verified.
- A person is excluded or was contacted inside the reader's chosen interval.
Expected output
A checked list with relevant pieces, contact routes, dropped entries, and held draft pitches.
Receipt fields
- Pages visited, checked dates, and verification outcome for each person.
- Draft versions, exclusions, approval decisions, and send status.
Recovery when it fails
- Recheck a contact route and recent work before any approved send.
- Remove stale or unverifiable entries with a recorded reason.
On this page
Build the list from recent work
A press and podcast list should connect each person to a piece of work that makes the outreach relevant. Name, outlet, and contact route are necessary, but they do not explain why that person should receive your story. OperatorNest can research and draft the list, then hold every pitch for your review. The output above is an illustration, not a record of a live outreach run.
Define the story angle before searching. A broad phrase such as “AI” can return hundreds of weak matches. A narrower question, such as how people review delegated work before an agent acts, gives the operator a way to evaluate recent articles and episodes. Set a publication window and a target count. The count is a search target, not a reason to include an unverified person.
Verify four fields per person
For each candidate, record their name, outlet or show, a specific recent piece, and a public contact route. Link the article or episode itself. Check whether a byline is original or syndicated, and whether the person is still associated with the outlet. If a contact page has only a general form, record that accurately rather than inventing a direct address.
Say Sam asks for 25 contacts. The research finds 25 candidates, but three cannot be verified: one has a syndicated byline, another has no public contact route, and a third covered the topic outside the chosen window. The result should contain 22 eligible entries and a dropped log with reasons. It should not turn the target number into a claim that all 25 are ready to pitch.
Give the previous-contact list the same check as the new search. If Sam contacted a host recently, another pitch may be unwelcome even if the host fits the story. Compare the list before writing drafts and keep the exclusion rule visible in the receipt.
Draft around the person’s work
Each pitch should refer to one concrete point in the linked piece or episode. A first line that could be pasted into 50 messages is a sign the research did not shape the draft. Keep product claims inside the approved facts supplied with the task. If a claim needs a customer quote, run result, or number that has not been supplied, leave it out.
The research assistant use case covers finding and checking sources. This workflow adds the contact record and approval boundary. A draft is useful even when it never sends; it lets you judge whether the person’s work and your angle truly fit.
Review outreach as its own task
Before any send, recheck the contact route and the person’s current work. Show the recipient, source piece, exact message, and any previous contact in one review surface. You may approve some pitches, edit others, and reject the rest. No batch count should hide which individual messages will leave your account.
The receipt should show pages checked, inclusion decisions, your edits, and approved sends. If a contact route disappears or a person changes outlets, mark the entry stale and stop that pitch.
Run it on a schedule
Plan the timing and missed-run behavior with the recurring task planner. For documented schedule controls across assistants, read scheduled AI tasks compared.
Related tools
- AI follow-up email writer
Draft two or three follow-up emails for a proposal, meeting, unpaid invoice, or introduction, with a suggested send time and copy buttons.
- Email tone checker
Check an email draft for unclear asks, passive aggression, over-apology, and missing deadlines, then copy or download a rewrite in your chosen tone.
- AI agent approval policy
Set clear approval rules for an AI agent. Choose systems and risk limits, then copy a policy that says what waits and what can run.
Common questions
Is a large list better than a small verified one?
A shorter list with a specific piece and public contact route gives you a clearer basis for each pitch. Keep unverified names in the dropped log.
Can it use an address found on a third-party directory?
Prefer a contact route on the person's own site, outlet page, or recent byline. If the route cannot be confirmed there, mark it unverified.
When do pitches go out?
Only after you review the recipient and exact text for each pitch or approved batch. A research instruction does not authorize a send.
Keep reading
- An AI research assistant that turns a question into a sourced briefIt researches a question or a standing topic and delivers a sourced brief, once or on a recurring schedule.
- Approvals before anything that mattersIt asks before it sends, pays, books, deletes or posts. You approve, edit or hold from any channel.
- A receipt for everything your operator doesEvery task keeps a receipt: what was asked, what ran, which model did it, what it cost, and every approval.
- Build a press list with an AI agent, overnightBuild a press list with an AI agent overnight: verification rules, tone memory, your own model key, and every pitch held for approval before it sends.
Hand off your first task tonight.
Tell us your email and what you'd hand off first. We'll send your access details and help you set up your operator.