As of 13 August 2026, AI can categorise your customer quality complaints.

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. Open your approved quality complaint taxonomy and copy the category names, definitions and overlap rules into a working document.
   2. Gather a batch of complaints with their IDs, removing customer names, contact details and other information not needed for categorisation.
   3. Add several previously approved examples showing how your team handled straightforward and ambiguous complaints.
   4. Paste the taxonomy, rules, examples and complaint batch into the prompt, then run it in a chatbot or document-processing tool.
   5. Export the returned table and sort it by low confidence, human review and categories with the largest complaint counts.
   6. Compare every sampled label and quoted evidence with the original complaint and your approved definitions, then ask a quality colleague to resolve exceptions before updating the quality log or routing work.

   Prompt

   Categorise the customer quality complaints below using only the taxonomy and rules provided. Do not invent categories, causes, product defects, customer intentions or corrective actions. If a complaint fits more than one category, choose the primary category using the stated priority rules and list the secondary category separately. If the evidence is insufficient or the case is ambiguous, use "Needs human review" rather than guessing.
   
   Return a table with these columns: complaint ID, primary category, secondary category if applicable, evidence quoted from the complaint, confidence as high/medium/low, and reason for any human review. Keep the complaint wording unchanged. Do not infer facts that are not in the text.
   
   Taxonomy and category definitions:
   [PASTE YOUR APPROVED CATEGORY LIST AND DEFINITIONS]
   
   Priority rules for overlapping categories:
   [PASTE YOUR APPROVED PRIORITY RULES]
   
   Examples of previously approved classifications:
   [PASTE 5 TO 20 REDACTED EXAMPLES]
   
   Complaints to categorise:
   [PASTE COMPLAINT IDs AND TEXT HERE]

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