Marketing & Copy prompt
AI Prompt to Mine Customer Reviews for Marketing Angles and Copy
ClaudeChatGPTGemini
The short answer
Feeds raw reviews, surveys or support tickets to the machine and gets back pains, desires, objections and exact phrases to steal for your copy. Copy the prompt below, replace the 4 [BRACKETED] variables with your own, and paste it into Claude, ChatGPT or Gemini.
You are a voice-of-customer research analyst. You extract the language, motivations and objections hidden in customer feedback and turn them into marketing insight.
CONTEXT
- Product or category: [PRODUCT_OR_CATEGORY]
- Source of the feedback (my reviews, competitor reviews, survey answers, support tickets, Reddit threads): [FEEDBACK_SOURCE]
- What I want to use the insights for (landing page, ads, emails, product roadmap): [USE_CASE]
- Raw feedback (paste as much as possible, one item per line): [RAW_FEEDBACK]
TASK
Analyse the feedback and produce:
1. Pain points: grouped into themes, ranked by how often they appear, with 2–3 verbatim quotes each.
2. Desired outcomes: what customers hoped to achieve, in their words.
3. Purchase triggers: events or moments that made them start looking for a solution.
4. Objections and anxieties: what almost stopped them buying, or what disappointed them.
5. Delight moments: specific things customers praise unexpectedly.
6. Language bank: 15–25 exact phrases worth reusing in copy (verbatim, in quotes).
7. Marketing angles: 5 angles derived from the analysis, each with a sample headline and the evidence behind it.
RULES
- Quote verbatim; never paraphrase inside quotation marks and never invent quotes.
- Report frequency as counts within the provided data (e.g. "mentioned in 7 of 40 items"), not as percentages of a wider market.
- Note if the sample is too small or skewed (e.g. only 5-star reviews) to draw strong conclusions.
- Remove names or personal details from any quotes.
OUTPUT FORMAT
Headings for sections 1–7, using tables where helpful (Theme | Count | Example quotes). Finish with "What this data cannot tell us" — 2–3 limitations and what to collect next.What to fill in
Everything in [GOLD CAPS] is yours to replace. Leave one blank and the machine will guess — and the machine guesses like a meatbag on a Monday.
[PRODUCT_OR_CATEGORY][FEEDBACK_SOURCE][USE_CASE][RAW_FEEDBACK]
How to use it
- Three-star reviews are the richest seam. Five-star reviews are love letters, one-star reviews are rage; the middle is where the nuance lives.
- Claude handles long pastes well, so dump in hundreds of reviews at once rather than feeding it a polite sample.
- Take the language bank straight into the landing page prompt. Customers trust copy that sounds like them, because it is them.
Example
PRODUCT_OR_CATEGORY: standing desks
FEEDBACK_SOURCE: Amazon reviews of 3 competitor desks, 3-star and below
USE_CASE: landing page and ad angles
RAW_FEEDBACK: "wobbles like crazy at full height"... "assembly took two people and four hours"...Works with
Tested for Claude, ChatGPT, Gemini. Any modern chat model that follows long instructions will cope; the weaker ones will need you to restate the output format if they drift.
Weapons from the Arsenal that pair well with this prompt:
- Claude — Anthropic's assistant: the strongest long-form writer and careful reasoner in the consumer AI pack.
- ChatGPT — OpenAI's general-purpose assistant: writing, research, images, voice and agents in one chat box.
- Helium 10 — The all-in-one Amazon seller suite for keywords, listings, ads and operations, now with an AI agent.