Mining customer comments for insight
Listen to this lesson
Nigerian voice narration
Fictional training case. At 'Bloom Skincare' (fictional), customer care lead Nneka has 300 unread WhatsApp and Instagram comments after a product relaunch. Comments range from praise to complaints about a fictional 'ClearGlow Serum' packaging. Work through the example, identify the evidence needed and practise the method with cleared or synthetic inputs. Any additional figures in the explanation are separate practice examples, not claimed rows in a downloadable file.
Worked example
Nneka removes all names and phone numbers, then prompts: 'Group these customer comments into themes with counts and example quotes. Then suggest 3 quick fixes and 1 bigger improvement per theme, ranked by effort.' The AI groups them: 'Packaging leaks' (62 comments), 'Price too high' (48), 'Loved the scent' (90), 'Delivery delays' (35). It suggests quick fixes for packaging (tighter cap, bubble wrap) and a bigger fix (redesigned bottle). Nneka verifies the counts roughly match a manual sample check of 50 comments before trusting the theme breakdown.
The method
Always strip names and phone numbers from customer comments before pasting them into any AI tool — this keeps customer identities out of prompts and any logs the tool might retain. Theme counts tell you where to focus first; always sanity-check the AI's counts against a manual sample, since miscounted themes could misdirect your whole response plan.
Worked practice and common mistakes
The four named themes total 235 of 300 comments; 65 remain outside those themes. Packaging at 62 of 300 is about 20.7%, not 26% of all comments. If themes overlap, state the coding rule and denominator. A positive scent theme with 90 comments is larger than the packaging complaint; distinguish largest overall theme from largest complaint. Severity and effort can matter alongside count, so do not automatically rank every action by frequency alone.
Review before using the result
Nneka reports packaging leaks in 62 of 300 comments, about 20.7%. The listed themes cover 235 comments and leave 65 unclassified. She distinguishes the largest complaint from the larger positive scent theme, validates coding and considers severity and effort when prioritising action.
Your workplace practice
Use the recommended free assistant with this lesson's fictional example. Group these customer comments into themes with counts and example quotes. Then suggest 3 quick fixes and 1 bigger improvement per theme, ranked by effort. Keep a copy of the input and draft. Compare the result with the verification steps above, record any correction, and retain placeholders where evidence is missing. The expected result is a checked draft for human review, not an automatically approved decision. Course examples and practice responses are fictional; amounts, dates, rules and thresholds are not official guidance or supplier quotations.
Open Google Gemini (free tier) and paste this prompt there. Can propose themes from manageable batches of fictional comments; verify grouping, counts and quoted passages.
Group these customer comments into themes with counts and example quotes. Then suggest 3 quick fixes and 1 bigger improvement per theme, ranked by effort.
Why this tool: Can propose themes from manageable batches of fictional comments; verify grouping, counts and quoted passages.
Hit a free limit? Switch to ChatGPT or Gemini and continue — the prompt works in any assistant.