Analyse customer reviews and find recurring complaints
Turn a review set into evidence-backed themes, frequencies and improvement priorities.
Best for
- Small businesses with a meaningful review sample
What you need
- Anonymised review export
- Date range and business question
Limitations
A small or biased sample does not represent every customer.
Recommended tool
Claude
Claude can classify a long review set while retaining evidence quotes.
AI4E may earn a commission from partner links. This never affects which tool we recommend. Read the affiliate disclosure →
Pick your stack
Do it for nothing, or move faster with a paid tool.
AI4E may earn a commission from partner links. This never affects which tool we recommend. Read the affiliate disclosure →
Your copy-paste prompt
Fill in the blanks on the left and the prompt updates as you type.
Fill in your details
0 of 2 fields filled — every matching placeholder updates as you type.
Act as a customer-insight analyst. Analyse only [ANONYMISED REVIEWS] from [DATE RANGE]. Build themes from repeated evidence, count reviews per theme, separate complaint, request and praise, and quote representative text with review IDs. Do not infer customer identity or claim the sample represents all customers. Separate facts, interpretations and assumptions; state sample limitations and confidence. Ask questions if rating scale or business question is unclear. Return methodology, theme table, priority matrix and items needing human review.
Step-by-step instructions
- 1
Remove names and personal details, then number reviews.
- 2
Run the analysis and verify a sample in each theme.
- 3
Merge overlapping labels and recalculate counts.
- 4
Choose improvements using frequency, severity and business context.
Check the result
- Counts reconcile with the sample
- Quotes support each theme
- Minority experiences remain visible
- Priorities are not based on frequency alone
If the result is poor
- If themes overlap, define inclusion rules and reclassify.
- If sentiment dominates, request complaint type and operational impact separately.
Privacy and accuracy
- Remove personal data and avoid profiling individual customers.
Sample input
180 anonymised café reviews from six months.
Expected output
Seven themes with counts, evidence, confidence and a priority matrix highlighting wait time and order accuracy.
What people do next
Three recipes chosen for the task that usually follows this one, not just for sharing a category.
Business Automations
Route customer feedback into consistent, useful themes
30 mins · Zapier + OpenAI
Start this workflow →
Business Automations
Turn spreadsheet data into a plain-English weekly report
25 mins · ChatGPT
Start this workflow →
Business Automations
Create a sourced competitor comparison for a small business
45 mins · Perplexity
Start this workflow →