LeadHaste
IntermediateNo variables to fill — paste & go

Generate ranked pain hypotheses for an account from public signals

Converts a pile of raw public signals into 4-6 falsifiable pain hypotheses, each graded by evidence strength, costed if true, and ranked by fit with your product — plus the discovery question and email opener that would test the top two. It replaces persona-based guessing with evidence-based hypotheses you can confirm or kill in one conversation.

The prompt
You are an outbound strategist who builds campaigns on pain hypotheses, not personas. Your rule: a hypothesis must be specific enough to be wrong. 'They probably want more revenue' is not a hypothesis; 'their 14 open CS roles suggest ticket volume is outrunning headcount' is.

I'll paste public signals I've gathered about one account — job postings, news, website copy, reviews, posts, anything. You produce:

1. SIGNAL INVENTORY — restate the signals I gave you in one line each, so we agree on the evidence base. Flag anything stale or ambiguous.
2. PAIN HYPOTHESES — 4 to 6, each with: the hypothesis in one falsifiable sentence, the signals supporting it, a confidence grade (strong / plausible / speculative), and the cost of the pain if true (time, money, risk, morale).
3. RANKING — order them by (a) evidence strength times (b) how directly my product addresses it. Show the reasoning, not just the order.
4. THE TEST — for the top 2: the discovery question that would confirm or kill each, and the email opener that leads with the observation rather than the assumption.
5. DISCARDS — hypotheses you considered and rejected, with why. This keeps me honest about what the evidence doesn't say.

Rules: every hypothesis must trace to at least one signal I actually provided — no persona boilerplate. Falsifiable phrasing only. If my signals are too thin for 4 hypotheses, say so and list exactly what to go collect.

Before you write anything, interview me. Ask me these questions ONE AT A TIME, waiting for my answer each time:
1. What does your product fix, and what does that pain cost customers before they buy?
2. What's the account, and who's your target buyer there?
3. Paste every signal you've collected — job posts, news, reviews, site copy, posts. More raw is better than summarized.
4. What pains do your closed-won customers most often admit in discovery?

Once you have my answers, build the hypotheses.

How to use it

  1. 1

    Copy the prompt into Claude, ChatGPT, or any LLM.

  2. 2

    Collect signals first — 10 minutes on their careers page, news, G2, and LinkedIn gives the model real material. Paste raw content, not your summary.

  3. 3

    Answer question 4 with real discovery patterns; it teaches the model what pain looks like in your world.

  4. 4

    Lead outreach with the top hypothesis's observation, and hold the hypothesis itself for the prospect to confirm.

  5. 5

    After discovery, tell the model what was confirmed or killed — the recalibration sharpens every future account.

Best practices

  • Respect the confidence grades: build sequences on 'strong,' test 'plausible' in discovery, and never email a 'speculative.'

  • Read the DISCARDS section carefully — knowing what the evidence doesn't support saves you from the confident-but-wrong opener.

  • Job postings are the densest pain source: what a team is hiring for is what it can't currently do. Always include them in your paste.

  • Chain with the trigger-event scan and hiring-signals prompts from this category — their outputs are exactly the signal inventory this prompt wants.

Example: what this looks like in practice

An SDR at a QA-automation startup gathers signals on a 300-person fintech: 6 open QA engineer roles, a status page showing two incidents last month, an engineering blog post about release velocity, and a G2 review of their app mentioning bugs. The model produces five hypotheses; top-ranked, graded strong: 'manual QA capacity can't keep pace with their stated weekly-release goal — evidenced by the hiring spree and incident pattern.' The email opener cites the six roles and the release-velocity post together. Reply from the VP Engineering: 'painfully accurate — who told you?' Discovery confirms hypothesis one in the first ten minutes.

Best fit

Roles
SDR / BDRAccount ExecutiveFounder / CEO
Company size
Startup (1–10)SMB (11–50)Mid-market (51–500)
Audience
B2B
Industries
Any industry
Works with
Any LLM
Difficulty
Intermediate

This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.

Apply for a Pilot Spot →
Prompt FAQ

Frequently asked questions

A specific, testable claim about a problem an account probably has, grounded in observable evidence — 'their six open QA roles suggest testing can't keep pace with releases.' It differs from a persona pain ('QA leaders struggle with coverage') by being falsifiable: discovery can confirm or kill it. Outreach built on hypotheses reads as insight; outreach built on personas reads as templates.

More account & prospect research prompts