PR & Earned Media

PR & Earned Media

Get coverage you did not pay for, without an agency. The work is not writing a press release — it is deciding whether you have a story, proving you can back it, aiming it at twenty of the right people, and having the materials ready when one says yes.

When to use

The user wants press, is about to pitch a journalist, is drafting a launch announcement, or is asking whether a PR agency is worth it at their stage.

Do not use this to write a press release for a mass send. That is the practice this skill exists to replace.

Before you start

1. Read the brand/product context first (.agents/product-marketing.md / .agents/aaj-brand.md) for audience, positioning and what the company can honestly claim.
2. Get the job of the coverage. Credibility for sales, visibility for hiring, proof for investors, or authority in search and AI answers. One job, written as one line: "We want [audience] to see us in [outlet] so that [outcome]." Every later step serves that line.
3. Get the evidence. For any figure: where it came from, over what period, and whether the method is written down. For a customer result: whether written permission exists. If the user cannot produce these, say so and stop — an unbacked figure costs the relationship, not just the story.

Method

Earned media is borrowed trust: a journalist gives you their audience's attention because you gave their audience something useful. That exchange only works when the story survives without your company name in it.

Four story types work for a startup:

Two tests decide everything. The company-removed test: strip the company name out; if nothing is left, there was never a story. The forward test: would a reader send this to a colleague?

Workflow

1. Name the job (one line, as above). Refuse to proceed without it — it is what makes an outlet list arguable rather than arbitrary.
2. Score the story. Run the engine below. Report the verdict and every failed check verbatim. Do not soften a "Do not pitch yet".
3. Fix or replace the story if it is blocked. Usually the fix is a different angle on the same quarter's work, not a new event.
4. Build the list — twenty names, not two thousand. For each: what they cover, what they wrote recently, and why this story fits their beat. If the third column is empty, the name does not belong on the list. Never invent outlets or reporter names; if you do not have real ones, say so and describe the type of outlet to look for instead.
5. Become a source before pitching. Publish the data on the company's own site first, answer journalist requests in the field, engage with their work honestly.
6. Draft one pitch per person, under 200 words. Subject line is the story, not the company. First two lines: what is new and why their readers care. Then the offer — the data, an interview, a customer who will talk. No attachments; link instead. Run checkPitch on the draft.
7. Ready the materials before the first send: press page, current headshots, logo pack, the written method, and any customer who has agreed in writing to speak.
8. Follow up once, after a few days, with something new. Then stop.
9. Set up the tracking sheet: who, when, angle, result. It is the only way to learn which stories land.
10. Make coverage work twice. Link it, quote it in the deck, post what you learned rather than "we were featured", and send the journalist your next useful idea.

Run the tool

node resources/newsworthiness.js --demo

scoreStory(story) returns a score, a verdict, every check with its reason, and the materials that are not ready. checkPitch(pitch) applies the Step 5 rules to a draft: subject line, length, personalisation, attachments, named offer, and how many people it is going to.

Use the engine's output rather than your own judgement of newsworthiness — it is the same rubric every time, which is the point.

Present the result

Report in this order:

1. The one-line job of the coverage.
2. The verdict and score, then each failed check with its reason. Lead with the blockers.
3. What is missing from the materials.
4. The outlet list, with the "why this fits" column filled in for every row.
5. The pitch draft, with the pitch check output beneath it.
6. What you will track, and when to judge it.

Keep the verdict plain. "Do not pitch yet" is more useful than an encouraging score.

Guardrails & common mistakes

• Never invent a journalist, outlet, publication or quote. If real names are not available, say so and describe the beat to search for.
• Never pitch a figure the user cannot source. One figure that falls apart costs the source relationship for every future story.
• No customer result without written permission — no logos, no numbers, no name.
• Paid placement is advertising, even when it looks like an article. Label it sponsored and do not count it as press.
• Mass sends reach nobody. One pitch, one person, one beat.
• Do not promise coverage. Nothing here makes an editor publish. Reply rate is the only part of this the user controls.
• Judge it on the user's own numbers — reply rate, pieces published in the Step 1 outlets, and how many buyers mention coverage when asked how they heard of you. Industry averages describe other people's lists.
• An agency is rarely right at Seed to Series B. The founder is the most credible source the company has. Say that plainly if the user asks.

Related AAJ resources

• Full method: https://aajconsult.com/playbooks/earned-media-playbook
• A worked pitch, before and after: https://aajconsult.com/blog/how-to-pitch-a-journalist
• Borrowing an audience instead: https://aajconsult.com/playbooks/partner-co-marketing-playbook

Related skills

• content-repurposing — make one piece of coverage work across channels
• brand-voice-governance — keep the pitch and the commentary sounding like the company
• copywriting — sharpen the subject line and the first two lines