Sales & Outreach prompt
ChatGPT Prompt for Sales Discovery Call Prep
ChatGPTClaudeGemini
The short answer
Turns your account research into a one-page discovery brief: hypotheses, sharp questions, red flags. Walk in knowing more than they expect. Copy the prompt below, replace the 9 [BRACKETED] variables with your own, and paste it into ChatGPT, Claude or Gemini.
You are a senior account executive known for discovery calls that uncover real pain instead of running a checklist.
CONTEXT
- What I sell and its typical price range: [PRODUCT_AND_PRICE]
- Problems it solves best: [CORE_PROBLEMS]
- Prospect company: [COMPANY], [INDUSTRY], approx. size [COMPANY_SIZE]
- Attendees and roles: [ATTENDEES]
- How this call came about (inbound form, referral, outbound reply): [LEAD_SOURCE]
- Research notes (website, news, job posts, LinkedIn, tech stack): [RESEARCH_NOTES]
- Call length: [CALL_LENGTH]
TASK
Build a discovery call brief that I can read in five minutes before the call.
RULES
- Base every hypothesis on the research notes or clearly label it as an assumption. Do not invent company facts, numbers, or quotes.
- Questions must be open-ended and specific to this company, not generic ("What keeps you up at night?" is banned).
- Prioritise questions that uncover: current process, cost of the problem, who decides, timeline, and what happens if they do nothing.
- Keep the whole brief under 450 words.
OUTPUT FORMAT
1. Snapshot (3 bullets): who they are and what likely matters to them right now.
2. Top 3 pain hypotheses, each with the evidence behind it and a confidence level (low/medium/high).
3. Opening (2–3 sentences): how to set the agenda and get permission to ask questions.
4. 10 discovery questions grouped by: Situation, Pain, Impact, Decision process, Timeline. Mark the 3 must-ask questions with *.
5. Likely objections or red flags to listen for, with a one-line way to probe each.
6. Ideal next step to propose at the end of the call, plus a fallback.
If my research notes are too thin to form hypotheses, first list the 5 things I should find out before the call and where to look.What to fill in
Everything in [GOLD CAPS] is yours to replace. Leave one blank and the machine will guess — and the machine guesses like a meatbag on a Monday.
[PRODUCT_AND_PRICE][CORE_PROBLEMS][COMPANY][INDUSTRY][COMPANY_SIZE][ATTENDEES][LEAD_SOURCE][RESEARCH_NOTES][CALL_LENGTH]
How to use it
- Run it again after the call with your notes pasted in and ask what you failed to uncover. Humbling. Useful.
- Each attendee role is a different call. Add the champion and the economic buyer separately if you know both.
- Pair it with Perplexity for the research step, but check anything it finds before you repeat it to a prospect's face.
Example
PRODUCT_AND_PRICE: warehouse scheduling software, mid five figures per year
COMPANY: regional 3PL, ~200 staff
ATTENDEES: Ops Director, Warehouse Manager
LEAD_SOURCE: inbound demo request
RESEARCH_NOTES: hiring 12 warehouse associates, opened second facility this springWorks with
Tested for ChatGPT, Claude, Gemini. Any modern chat model that follows long instructions will cope; the weaker ones will need you to restate the output format if they drift.
Weapons from the Arsenal that pair well with this prompt:
- ChatGPT — OpenAI's general-purpose assistant: writing, research, images, voice and agents in one chat box.
- Perplexity — AI answer engine that searches the web and replies with cited sources.
- Claude — Anthropic's assistant: the strongest long-form writer and careful reasoner in the consumer AI pack.