Home · Business · Customer Service · Feedback & NPS analysis

YES

As of 13 August 2026, AI can find out why customers recommend your business.

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 costsA customer-insight analyst or survey platform is the alternative; no price is supplied in the available tool information.

If this goes wrong: you prioritise a loud or misleading theme, spend money changing the service and miss the reason most customers actually recommend you.

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. Export the relevant survey answers, recommendation scores and review comments from your survey, CRM or review system, keeping the question wording and date range with the export.
    2. Remove names, email addresses, phone numbers, order references and other details that could identify a customer, then separate each response clearly.
    3. Open a chatbot and paste the prompt, replacing each bracketed slot with your business description, response details and anonymised feedback.
    4. Ask the chatbot to analyse the complete dataset in batches if the export is too large, using the same prompt and asking it to keep theme names consistent across batches.
    5. Compare every reported quotation and theme count with the original export, and remove any quotation or count that cannot be found there.
    6. Compare the final themes with a simple tally of responses and with your knowledge of customer groups, then choose only actions supported by more than one clear source of evidence.

    Prompt

    Analyse the anonymised customer feedback below to find out why people recommend this business.

Business and offer: [brief description]
Customer groups or segments, if known: [description]
Question wording and response scale: [paste it]
Recommendation scores or NPS data, if available: [paste it]
Customer comments: [paste anonymised comments, one response per line or row]
Date range and number of responses: [details]

Use only the information supplied. Do not invent customer motives, demographics, counts or causes. Group comments into no more than six clear themes, using plain language. For each theme, give its name, what customers mean, the number of comments supporting it if that can be calculated, two representative quotations with identifying details removed, and whether the evidence is strong, mixed or weak. Separate reasons customers explicitly state from interpretations you are making. Identify comments that are contradictory, ambiguous or too isolated to support a conclusion. If recommendation scores are supplied, compare the themes in higher-scoring and lower-scoring responses without claiming that a theme causes recommendation. Finish with three practical questions I should investigate next and a short summary of what the data does not tell me.

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

What it gets wrong

  • AI cannot know whether a theme is representative when your feedback comes from a biased group of respondents.
  • AI cannot establish that a stated reason caused a customer to recommend you rather than merely appearing in the same comment.
  • AI cannot resolve sarcasm, local expressions or ambiguous comments reliably without your business and customer context.
  • AI cannot obtain missing feedback or safely decide whether customer data was collected and shared for this analysis.
  • AI cannot take responsibility for changing your service, pricing or customer treatment based on the findings.

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

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
Liability1
Effort delta2
Total8 / 10

The methodology and its thresholds are published in full.

FAQ

Can ChatGPT analyse customer feedback to find out why people recommend my business?
Yes. Give it anonymised comments, the question customers answered and any recommendation scores, and it can group recurring reasons and quote the supporting evidence. Check every quotation and count against the source data before making a business decision.
Can AI analyse my NPS comments?
Yes, it can compare themes in responses from different recommendation-score groups and summarise what customers say. It cannot prove that a theme causes a higher score, or correct a sample that does not represent your customers.
Can AI tell me what my customers value most?
It can identify the topics mentioned most often in the feedback you provide. That is not automatically the same as what all customers value most, so compare the result with response counts, customer groups and other evidence.
Is it safe to upload customer feedback to AI?
Remove names, contact details, order references and other identifying information before uploading it, and check your organisation's data policy and the tool's handling terms. Do not include private customer details that are not needed for the analysis.

Nearby answers

All Feedback & NPS analysis 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.