Mine G2 and review sites for competitor pain quotes
Extracts up to five verbatim pain quotes from a product's critical reviews, each categorized and tagged with the reviewer's segment, plus the dominant pain theme. Feed it your competitors' G2 pages and you get displacement copy written by their own unhappy customers — the exact words your prospects would use.
You are a voice-of-customer researcher running inside an enrichment pipeline. You mine public software reviews — G2, Capterra, TrustRadius — for the exact language real users employ when describing pain with a product. Verbatim customer language is the highest-grade copy input that exists: it outperforms anything a copywriter invents, because prospects recognize their own words.
Work from the product review page you are given (a URL to browse or scraped review content) for the product named in your input.
Rules:
- Extract only from the critical sections: 'What do you dislike?', cons lists, and 1–3 star reviews. Ignore praise.
- Quotes must be verbatim, 5–30 words, trimmed with ellipses only at the edges. Never paraphrase inside quotation marks.
- Prefer pains about outcomes and workflows (slow support, broken syncs, surprise pricing) over one-off bugs.
- Capture the reviewer's stated role or segment when shown (e.g. 'Mid-Market, Sales'); empty string if absent.
- Skip reviews that appear incentivized or contentless ('nothing to dislike!').
OUTPUT CONTRACT — follow exactly:
Return a single JSON object:
{"product": "<product name as provided>", "pain_quotes": [{"quote": "<verbatim, 5–30 words>", "reviewer_segment": "<as stated or empty string>", "pain_category": "<pricing | support | usability | reliability | integrations | reporting | other>"}], "dominant_pain": "<the single most repeated pain category across everything you read>"}
- Return ONLY the JSON. Maximum 5 quotes, strongest first. No markdown fences, no commentary.
- If the page is unreachable, has no critical reviews, or quotes cannot be extracted verbatim, return exactly: NOT_FOUND
- Never fabricate, soften, or sharpen a quote. Verbatim means verbatim.How to use it
- 1
Copy the prompt into Claude, ChatGPT, or any LLM with web access — or into a Clay Claygent column.
- 2
In Clay, map a product-name column and a review-page URL column as inputs — one row per competitor (or per product your ICP currently uses).
- 3
Verify a sample of quotes against the live review pages; the verbatim rule means any paraphrase is a failed run worth re-prompting.
- 4
Build campaign copy around dominant_pain, and echo the quote language (not the quotes themselves) in your emails and talk tracks.
Best practices
Mine the tools your ICP already uses, not just direct competitors — pain with an adjacent tool is often the trigger that opens budget for yours.
Echo the vocabulary rather than citing reviews: prospects say 'the sync breaks', so your email says 'when the sync breaks', with no mention of G2.
Refresh quarterly and watch dominant_pain shift after a competitor's pricing change or a rough release — that's your window.
Example: what this looks like in practice
A founder selling a Smartlead-adjacent deliverability tool runs this on the G2 pages of four incumbent sequencers. Across rows, dominant_pain comes back 'support' three times, with quotes like 'took nine days to get a human reply' and 'chat support just links the docs'. She builds a campaign whose opener mirrors the language: 'Most teams switching to us say the same thing — nine days for a support reply while a domain burns.' Nothing cites G2, but every prospect running those tools recognizes the pain instantly. The sequence pulls a 9% reply rate against a list filtered to users of the reviewed products.
Best fit
This prompt is one gear in a bigger machine. We orchestrate 20+ tools into outbound systems our clients own — and guarantee the results.
Apply for a Pilot Spot → →Frequently asked questions
Reading public reviews and learning from them is standard competitive research — the same thing an analyst does manually. What you should avoid is republishing reviews wholesale, misquoting them, or implying endorsements. This prompt extracts short verbatim excerpts for internal copy research; your outbound then echoes the pain language without citing or reproducing the reviews.
More personalization at scale (clay & claygents) prompts
Generate a cold email first line from scraped research data
The workhorse of AI-personalized outbound: converts whatever research your enrichment waterfall produced into one 12–22-word observation-led first line per row. The swap-test rule and fact-invention ban are what separate lines that read researched from the AI slop prospects now delete on sight.
Turn a Claygent website scrape into a personalized one-liner
Produces a sub-25-word, website-specific opening line for every row in a Clay table — no human in the loop. The output contract forces plain text only and a NOT_FOUND fallback, so unreachable sites and thin homepages return a filterable sentinel instead of a hallucinated compliment that ends up in a prospect's inbox.
Analyze a pricing page for sales-relevant intelligence
Extracts a structured pricing profile from any pricing page: model, tier count, exact displayed prices, free-offer type, and a one-line read on their sales motion. At scale it segments prospect lists by how companies charge — which predicts budget, buying process, and which of your angles will land.