Home · Business · Customer Service · Feedback & NPS analysis

YES

As of 13 August 2026, AI can identify your NPS detractors.

Most people should hand this to a purpose-built tool.

Can you do it?

5 minutesto a draft.

15 minutesto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ityou

What the alternative costsBotpress is an open platform for building LLM chatbots and agents; no human alternative price is provided here.

If this goes wrong: you miss a genuine complaint or send an inappropriate follow-up to the wrong customer, damaging trust and wasting the service team's time.

What to actually do

  1. Use a tool built for this

    The route this page recommends

  2. Do it yourself

    Second choice

    A chat interface, chat-fluent skill, and roughly 15 minutes until you can act on the result.

    How to actually do it

    1. Open your survey or CRM export and create a copy containing a respondent ID, the NPS score from 0 to 10 and the written feedback.
    2. Remove names, email addresses, telephone numbers and any other personal data that the identification task does not need, while retaining a non-identifying respondent ID.
    3. Check the copy for blank scores, scores outside 0 to 10, duplicate IDs and broken rows before pasting it into a chatbot.
    4. Paste the export into the prompt and ask the model to return the detractor table, data-quality issues and feedback themes separately.
    5. Compare every row in the returned detractor table with the source export and confirm that every listed score is between 0 and 6.
    6. Save the verified list and send it to the customer-service owner, who can decide whether and how each customer should be contacted.

    Prompt

    Analyse the survey data below to identify NPS detractors. Treat a detractor as a respondent whose NPS score is from 0 through 6 inclusive. Do not infer a score from comments when the score is missing, and do not invent or correct any data. Return: 1) a table of all detractor records with respondent ID, score and verbatim feedback, 2) the number of detractors and the total number of valid scored responses, 3) recurring themes in the detractor feedback with the supporting respondent IDs, and 4) a separate list of missing scores, duplicate IDs, invalid scores and other data-quality issues. Keep comments verbatim in the first table and anonymise names, email addresses, telephone numbers and other unnecessary personal data. Do not recommend contacting any customer or make a decision about their priority. Survey export: [PASTE CSV OR TABULAR DATA HERE]

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

  3. Hand it to a person

    The distant third

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

What it gets wrong

  • AI cannot access your survey or CRM unless you export and provide the relevant data.
  • AI cannot know whether a respondent ID maps to the correct customer when your source data is duplicated or stale.
  • AI cannot decide which detractors matter most without your rules about account value, complaint severity, timing and consent.
  • AI can summarise feedback themes but cannot establish that a theme is commercially important without your customer and service context.

Even on a YES, the friction has a name: private data access and context depth.

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
Verification2
Liability2
Effort delta2
Total10 / 10

The methodology and its thresholds are published in full.

FAQ

Can AI identify my NPS detractors?
Yes. Give it a clean survey export and define detractors as scores from 0 to 6, then check the returned rows against the original scores. It cannot access your customer system or decide which detractors deserve personal follow-up without your rules.
What data does AI need to find NPS detractors?
It needs a respondent ID and the NPS score, with feedback included if you want themes or complaint summaries. Remove names, email addresses and other unnecessary personal data before uploading the export.
Can ChatGPT analyse my NPS survey results?
Yes, it can filter scores, list detractor feedback and identify recurring themes from a pasted export. Check the output against the source file, especially where scores are missing, duplicated or outside the valid 0 to 10 range.
Is it safe to upload customer feedback to AI?
Only upload data your organisation permits you to share, and remove personal data that is not needed for the analysis. Check your employer's supplier, retention and confidentiality rules before putting customer feedback into an external AI service.

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.