thinkhard.SI

Prompt repository Sales

Write a sales proposal from discovery call notes

You need a proposal that uses the client's own words for the problem, scope, price and next steps.

Registry no.
SAL-007
Collection
Sales
Laid out for
Any chat model
Score
10 out of 10
Blanks to fill
4
Added

SalesAny chat

The prompt

Score 10 of 10, Excellent

Use it when You need a proposal that uses the client's own words for the problem, scope, price and next steps.

  • {{client_name}}
  • {{call_notes}}
  • {{offer_and_pricing}}
  • {{about_us}}
Full prompt and why it works
Prompt
Write a sales proposal for {{client_name}} based on my discovery call notes below. The readers are the people at {{client_name}} who will decide, including some who were not on the call, so the proposal must make sense on its own and show that we understood their problem before it talks about us.

Discovery call notes:
"""
{{call_notes}}
"""

What we are offering, with prices and terms:
"""
{{offer_and_pricing}}
"""

About us, in a few lines: {{about_us}}

Use these sections, with headings:
1. Your situation: the problem, why it matters now, and what it is costing them, in their own words. Quote them where the notes have their exact phrasing.
2. What success looks like: the outcomes or measures they named.
3. Our approach: how we will deliver, in steps or phases.
4. Scope: what is included, and a short "Not included" list, so both sides agree on the edges.
5. What we need from you: access, people, decisions or materials.
6. Investment: the price and payment terms exactly as given above. If there are options, show them in a table.
7. Next steps: the decision needed, by whom, and what happens after a yes, with dates only where the notes give them.
8. About us: two or three sentences.

Length: 600 to 1,000 words, with short paragraphs a busy executive can skim.

Rules:
- Use only facts from the notes and the offer. Do not invent results from other customers, return-on-investment figures, discounts, start dates or guarantees.
- Copy every price, quantity and term exactly. Do not round or restate them.
- Leave out anything in the notes marked internal, such as my read on their budget or the competition.

A good proposal reads as if it could only have been written for this client. For example, "You told us onboarding a new driver takes three weeks and needs to take one" rather than "Companies like yours face onboarding challenges."

If anything is missing from the notes that a section needs, such as who signs off or what success means to them, write "[TO CONFIRM: ...]" in that spot and list every open question at the end under "Questions before we send this".

Variables

{{client_name}}
The client company's name.
{{call_notes}}
Your notes or the transcript from the discovery call. Mark anything internal.
{{offer_and_pricing}}
What you are proposing: deliverables, options, prices and payment terms, exactly as approved.
{{about_us}}
Two or three facts about your company that matter to this client.

Why it works

  • Background facts. The situation and success sections come from the client's own words on the call, which is what makes a proposal feel written for them rather than adapted from the last one.
  • What to avoid. Prices and terms are copied exactly and no outside results or guarantees are added, so the proposal never commits you to something you did not approve.
  • If unsure. Gaps such as who signs off are marked to confirm and collected in a list, so you can close them before the proposal goes out.

How to use it

  1. Press Fill in variables and type your details into each blank. What you type stays in this browser.
  2. Copy the filled prompt into the AI tool you already use, such as ChatGPT, Claude or Gemini, or open it in the Grader or the Prompt Tester here.
  3. Before you rely on it, try it on three or four real examples. The Prompt Tester runs them all at once and shows which ones go wrong.
Use it from code

Get this prompt as JSON from the free REST API, with your own text in the blanks, or let an AI assistant fetch it through the MCP server. No key.

curl -s https://thinkhard.si/api/v1/prompts/proposal-from-call-notes

With the MCP server added to Claude, ChatGPT, Cursor or VS Code, call get_prompt with {"prompt": "proposal-from-call-notes"}, or just ask for it by name.