Canonical: https://botbento.com/usecases/extract-product-opinions-from-forum-threads/
Format: Markdown representation of the public HTML page.

# Extract product opinions from forum threads

[Research](https://botbento.com/usecases/research/) · Intermediate

Turn scattered community discussions into a structured opinion table

A bot periodically reads public discussion threads, identifies product mentions, and pulls out concise opinions with sentiment and cited context. It writes each finding to a structured table so you can compare products without manually skimming forums.

## Quick answers

**What does it do?** Turn scattered community discussions into a structured opinion table. Result: Structured opinion table.

**What does it need?** Tools: Browser, Database, Spreadsheet. Intermediate: some setup. The prompt includes an approval step.

**How do I run it?** On a schedule: Weekly forum scan. Copy the prompt below into a bot in BotBento.

## How it works

1. On a schedule: Weekly forum scan
2. Then: Extract opinions and citations
3. Result: Structured opinion table

## Prompt to start with

```text
On each scheduled run, scan the listed public forum pages for product mentions. Extract the product name, the opinion stated, sentiment, and a short quote with thread link. Deduplicate entries and write rows to the database. Do not post, reply, contact users, make payments, or delete any existing rows without the user's explicit approval. If a page layout changes or access fails, stop and report the issue instead of guessing.
```

## Tools it uses

Browser, Database, Spreadsheet

## Works with

BotBento

## Tags

research, data-extraction, forums, sentiment

## Source

[Show HN: 13M opinions on 48k products extracted from Reddit and HN](https://news.ycombinator.com/item?id=50364483) — Hacker News · checked 2026-11-13

Summarized in our own words. Any results described there are the authors’ own, not measured by BotBento.

Verified 2026-11-13: passed every automated gate and a second-model review (kimi-code/k3.1-pro). Automated, not human review. [How verification works](https://botbento.com/usecases/verification/)

By BotBento Editorial · Added 2026-11-13 · Updated 2026-11-13 · [Suggest a correction](https://botbento.com/contact/)

## Related use cases

- [Extract structured web data for research](https://botbento.com/usecases/extract-structured-web-data-for-research/) — Collect and organize public web search listings and directory data into structured datasets
- [Research forum digest bot for open problem threads](https://botbento.com/usecases/research-forum-digest-bot-for-open-problem/) — Summarize agent discussions and draft reviewable contributions
- [Turn plain-language place requests into map search plans](https://botbento.com/usecases/turn-plain-language-place-requests-into-map/) — Convert location questions into structured map filters and candidate places
