# Analyze customer feedback

Group raw customer comments into themes, priorities or feature requests, with counts and quotes that show where each finding comes from.

Paste into: **a new Claude conversation**

## Prompt template

How the builder on the page fills in this template, so you can do the same:

- `{{key}}` becomes the text entered for that field, exactly as typed. Empty text, or text that is only whitespace (spaces, tabs or line breaks), counts as empty: a required field then becomes `[fill in: <label>]` with its label, and an optional field becomes nothing.
- For a choice field, `{{key}}` becomes the text of the chosen option, or of the default when none is chosen.
- Text between `{{#key}}` and `{{/key}}` stays only when that field is not empty; a choice field always counts as filled. The markers themselves are removed, and a marker that stands alone on its line takes its line break with it.
- Each value is inserted once, as it is: `{{…}}` inside a value stays as typed.

```text
<feedback>
{{feedback}}
</feedback>

Analyze the customer feedback above.

What I want to learn: {{goal}}
{{#product}}
Product: {{product}}
{{/product}}

Please:
1. Group the comments by the underlying issue or request, not by wording. Give each group a short, plain name.
2. For each group, count how many comments belong to it and quote one or two short, representative comments word for word.
3. Order the groups for my goal: by how common they are for themes, and by how often they are asked for in the case of feature requests. For priorities, order groups by the customer impact the comments support and state that evidence; if impact is unclear, label it unknown and say what you would need to learn before ranking it.
4. End with three to five sentences summarizing what stands out.

Count only what the comments actually say, and never paraphrase inside quotes. Note comments you could not place in any group, and say if the sample looks too small or one-sided to support a conclusion.
```

## Fields

- `{{feedback}}` (required): Customer feedback. One comment per line works best. Remove names, emails and other personal details before pasting.
- `{{goal}}`: What you want to learn. Choose one:
  - The main themes and how common each is (default)
  - Which problems to fix first
  - A list of feature requests
- `{{product}}`: Product or service. Optional. A short description helps Claude interpret product-specific terms.

This site doesn't run Claude or show model output. Results depend on your input and the model you use.

## When to use this template

- You have a pile of open-ended comments from a survey, reviews or support tickets and need to see the patterns quickly.
- You want findings you can trace back to real comments, with counts and quotes, before sharing them with a team.
- You need a first pass at priorities or a feature-request list to discuss, not a final report.

## When not to use it

- You need statistically sound results from thousands of responses. Use a proper survey tool or analysis; a prompt can miscount at that scale.
- The comments contain personal data you are not allowed to share. Remove it before pasting, or do not paste at all.
- The feedback is already structured, such as numeric survey answers. A spreadsheet will summarize those more reliably.

## Why this structure

- All comments go first inside `<feedback>` tags, so the full set comes before the instructions, and the tags help Claude keep customer words apart from yours.
- The prompt asks for groups by underlying issue rather than wording, so comments that say the same thing differently are counted together. That is the main value of the exercise.
- The request for counts and verbatim quotes for every group gives you a way to trace each finding back to the comments.
- The ordering rule changes with your goal, which keeps one template useful for themes, priorities and feature requests.
- The last paragraph asks for unplaced comments and a warning about thin samples, so weak evidence is not presented as a trend.

## Example input (fictional)

Customer feedback:

```
Delivery was late again this week.
Love the recipes but the portions are small.
The app logged me out twice during checkout.
Please add a vegetarian plan.
Driver left the box in the rain.
Would like more vegetarian options.
```

What you want to learn:

```
Which problems to fix first
```

Product or service:

```
Fernbox (fictional), a weekly meal-kit delivery service
```

## Follow-ups to send Claude

- Show every comment you placed in the top group, so I can check the grouping.
- Draft a short reply we could send to customers about the most common issue.
- Which groups overlap? Would merging any of them change the order?

## Common mistakes

- Pasting more comments than Claude can count carefully. For large sets, analyze in batches and combine the group counts yourself.
- Leaving personal details in the comments. Strip names, emails and order numbers first.
- Treating the counts as exact. Spot-check one or two groups against the raw comments before quoting numbers to others.

## Related templates

- [Extract data to JSON](https://awesomeclaudeprompts.com/prompts/extract-data-to-json/): Pull specific fields out of messy text into JSON that matches your field list, with clear rules for missing or ambiguous values.
- [Summarize a document](https://awesomeclaudeprompts.com/prompts/summarize-document/): Turn a long document into a faithful summary: main point first, key points by importance, then decisions and open questions.
- [Compare options and recommend one](https://awesomeclaudeprompts.com/prompts/compare-options/): Weigh two or more options against your criteria and hard limits, see the trade-offs side by side, and get a recommendation when the facts support one.

## Sources

- [Prompting best practices: Long context prompting](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices#long-context-prompting) (Anthropic documentation)
- [Prompting best practices: Structure prompts with XML tags](https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/claude-prompting-best-practices#structure-prompts-with-xml-tags) (Anthropic documentation)
- [Reduce hallucinations](https://platform.claude.com/docs/en/test-and-evaluate/strengthen-guardrails/reduce-hallucinations) (Anthropic documentation)

---

Awesome Claude Prompts is an independent project. It is not affiliated with, endorsed by, or sponsored by Anthropic. Claude is a trademark of Anthropic, PBC.