Home · Business · Customer Service · Feedback & NPS analysis
As of 13 August 2026, AI can compare your NPS results over time.
This still needs a person who signs their name to it.
Can you do it?
15 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 spreadsheet analyst or customer-insight specialist is the alternative; no price is supplied here.
If this goes wrong: you treat a change caused by a different sample or survey question as a real shift in customer sentiment and act on it.
What to actually do
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.
Use a tool built for this
Second choiceDo it yourself
The distant thirdA chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.
How to actually do it
- Export the NPS results for each comparison period from your survey or customer-service system, including the response-level scores if available.
- Open a spreadsheet and record each period's dates, customer population, survey question, response scale, collection method and number of responses.
- Remove or identify duplicate and incomplete records without changing valid scores, then save the cleaned export as a CSV file.
- Paste the prompt into an AI chatbot and attach the cleaned CSV, plus any document that defines how your organisation calculates NPS.
- Ask the model to recalculate each period from the response-level scores and compare those results with the scores reported by your survey system.
- Check every count and score in the AI table against the CSV and your original survey report, including the survey question and customer population for each period.
- Send the comparison and the flagged limitations to the person responsible for customer insight before using it to change service targets or priorities.
Prompt
Compare the attached NPS survey results over time. Use only the data I provide and do not invent missing values. First identify the periods, number of responses, NPS score, promoter count, passive count and detractor count for each period. Then produce a comparison table showing the change between periods and a plain-English summary of the main movement. Flag any missing, inconsistent or duplicated data. Check whether the survey question, response scale, collection method, customer population and response volume changed between periods. Separate arithmetic findings from interpretation. Do not claim that a change is statistically significant unless the supplied data supports that conclusion. If the data is insufficient for a reliable comparison, say exactly what is missing and stop short of a firm conclusion. Treat customer comments as supporting context only, not as a replacement for the NPS calculation.
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 change in NPS reflects genuine customer sentiment or a different sample, question, channel or response rate unless you provide that context.
- AI can calculate a trend without establishing that the movement is statistically or operationally meaningful.
- AI cannot detect every data-export problem, such as an undocumented change in scoring rules or a customer group missing from one period.
- AI cannot decide which customer-service action is justified by the result without your organisation's targets, constraints and knowledge of the service.
Even on a YES, the friction has a name: verification cost, judgement under ambiguity 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.
| Axis | Score (0–2) |
|---|---|
| Output | 2 |
| Inputs | 2 |
| Verification | 1 |
| Liability | 1 |
| Effort delta | 2 |
| Total | 8 / 10 |
The methodology and its thresholds are published in full.
FAQ
- Can ChatGPT compare NPS scores from different months?
- Yes. Give it the response-level scores or reliable period summaries, along with the survey question, customer population and collection method, and it can calculate and compare the results. Check the counts and arithmetic against your source report before acting on the trend.
- Can AI tell me why my NPS went down?
- It can identify patterns in the data and comments that might explain a fall, but it cannot establish the cause from NPS alone. Changes in sampling, survey wording, response volume or collection channel can produce a misleading comparison.
- Is AI accurate for NPS analysis?
- It can be accurate at counting responses and calculating the supplied scores when the data and scoring rules are clear. It can still give a confident but weak interpretation if periods are not comparable, so check the source data and survey design yourself.
- Can AI create an NPS trend report?
- Yes. It can turn period results into a table, describe the movement and draft a report for colleagues. You still need to verify every figure and add the business context before sending or publishing it.
Nearby answers
- Can AI find product improvements from customer feedback?YES
- Can AI increase my customer feedback survey response rate?PARTLY
- Can AI analyse comments from my CSAT surveys?YES
- Can AI analyse customer feedback about my business on social media?PARTLY
- Can AI benchmark my NPS score against similar UK businesses?PARTLY
- Can AI combine customer feedback from multiple sources?PARTLY
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.