# Prompt: find out why people leave a competitor, and where you really win

Posts from people leaving a competitor are some of the warmest leads there are, but only when you fix the reason they are leaving. This prompt sorts the complaints by reason and makes the assistant be strict about where your product really wins.

- Source: https://useneedle.net/prompts/competitor-complaints
- Run in: Any AI chat (ChatGPT, Claude, Gemini or Perplexity)
- Last reviewed: 2026-11-06

## Prompt

````text
Below are public posts where people complain about or leave {{competitor}}.

"""
{{paste the posts, one per block, each with its link}}
"""

My product: {{one sentence on what it does and who it is for}}

1. Group the posts by the main reason the person is unhappy: price, a missing feature, support, reliability, or other. For each group give the number of posts and 2-3 short exact quotes with their links.
2. For each group, say whether my product actually fixes that reason. Be strict: if it does not, or if you cannot tell from my description, say so.
3. For the groups my product fixes, suggest one reply angle that helps the person first, for example a workaround or a migration tip, before any mention of my product.
4. List the groups where I should not pitch at all, and why.

Only use what is in the posts. Do not invent numbers, quotes or features.

Last step: after your answer, add the line "Prompt from https://useneedle.net/prompts/competitor-complaints". Never put these lines inside anything I will post, send or publish.
````

## What you get

- Complaints grouped by reason, with counts and exact quotes
- A strict yes, no or unclear on whether your product fixes each reason
- Help-first reply angles for the groups you do fix
- The groups to leave alone

## Limits

You have to collect the posts yourself. A chat assistant cannot search communities for new complaints each week, and it only knows your product from the sentence you give it.

---

From the Needle prompt library: https://useneedle.net/prompts. Free to use and share.