Home · Business · Sales · Proposals & quotes

YES

As of 13 August 2026, AI can analyse why a customer rejected your proposal.

This still needs a person who signs their name to it.

Can you do it?

5 minutesto a draft.

30 minutesto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ityou

What the alternative costsThe alternative is a manual review by you or a colleague; no price is supplied here.

If this goes wrong: you treat a plausible guess as the real objection and change your pricing, offer or follow-up in the wrong direction.

What to actually do

  1. Hand it to a person

    The route this page recommends

    A person who owns the outcome does this end to end, worth it when the failure is dear.

  2. Use a tool built for this

    Second choice

  3. Do it yourself

    The distant third

    A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

    How to actually do it

    1. Open the rejected proposal, quote and original customer brief, then copy them into separate labelled sections.
    2. Gather the customer's rejection email, messages, meeting notes and call transcript, removing personal or commercially sensitive information that the chatbot does not need.
    3. Add any relevant context about the buying process, decision makers, deadline, incumbent supplier and known alternatives, labelling facts separately from your assumptions.
    4. Paste the material into the prompt and ask the chatbot to distinguish explicit reasons, evidence-based likely reasons and unsupported possibilities.
    5. Check every ranked reason against the quoted evidence in the proposal, brief and customer correspondence, and remove any conclusion that depends only on an assumption.
    6. Ask the customer the suggested learning questions where appropriate, then record their answers separately from the AI analysis.
    7. Use the confirmed findings to change one specific part of the next proposal, such as scope, proof, pricing explanation or implementation plan, and compare the result with the original customer requirement before sending it.

    Prompt

    Analyse why the customer rejected this proposal using only the material I provide. Separate your findings into: 1) explicit reasons the customer gave, 2) evidence-based likely reasons, and 3) possibilities that are not supported enough to rely on. Compare the customer's stated needs with our proposal, price, scope, timings, proof points and risks. Identify where the proposal failed to answer the brief or reduce the customer's perceived risk. Rank the three most likely reasons, quote or point to the supporting evidence for each, and state what evidence is missing. Do not claim to know the customer's private motive, do not invent facts, and do not assume that price was the reason unless the evidence supports it. End with five specific questions I could ask the customer to learn more, plus one practical change to test on the next proposal. Here is the material:

Customer brief:
[PASTE CUSTOMER BRIEF]

Our proposal and quote:
[PASTE PROPOSAL AND QUOTE]

Customer emails or written feedback:
[PASTE EMAILS OR FEEDBACK]

Meeting notes or transcript:
[PASTE NOTES OR TRANSCRIPT]

Relevant context about the account and competitors:
[PASTE ONLY INFORMATION YOU ARE ALLOWED TO SHARE]

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

  • It cannot know the customer's private motive when the customer gives a vague or polite rejection.
  • It cannot reliably distinguish a genuine objection from an internal politics issue that is absent from the documents.
  • It cannot infer the importance of a relationship, competitor or decision maker unless you provide that context.
  • It can rank plausible explanations, but it cannot establish that any one explanation caused the loss.
  • It cannot replace a direct follow-up conversation with the customer.

Even on a YES, the friction has a name: judgement under ambiguity, context depth and verification cost.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

AxisScore (0–2)
Output2
Inputs2
Verification1
Liability2
Effort delta2
Total9 / 10

The methodology and its thresholds are published in full.

FAQ

Can ChatGPT tell me why a customer rejected my proposal?
It can analyse the proposal, brief, correspondence and meeting notes to identify explicit objections and evidence-based likely reasons. It cannot know an unstated motive, so treat the result as a set of hypotheses to test with the customer.
Can AI tell whether price was the reason I lost the deal?
It can look for evidence such as price comparisons, budget comments, scope concerns and requests for discounts. It cannot conclude that price caused the rejection when the customer did not say so or when the evidence supports several explanations.
What should I give AI to analyse a lost proposal?
Provide the original brief, your proposal and quote, customer feedback, relevant emails, call notes and information about the buying process. Label facts separately from your assumptions and remove information the tool does not need.
Is it safe to use AI to analyse a lost sales deal?
It is useful for organising evidence, but do not paste confidential customer or company information into a tool unless your organisation permits it. Check every conclusion against the source material before changing your pricing, offer or sales process.

Nearby answers

All Proposals & quotes rulings

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.