How Topics Are Created
As messages arrive from your connected sources, Modem analyzes them and groups related conversations into topics. When a new message fits an existing topic, it’s added. When it doesn’t, a new topic is created. Within channels, Modem also identifies where one conversation ends and another begins, creating distinct discussions within topics even when multiple conversations happen in parallel.Topics are generated automatically. You don’t need to tag, label, or categorize messages yourself.
What a Topic Contains
Each topic has an AI-generated title and summary, along with a priority level and the people involved.Feedback Classification
Each topic is classified by feedback type:| Category | What it captures |
|---|---|
| Bug Report | Error descriptions, broken functionality, unexpected behavior |
| Feature Request | Suggestions, wishlist items, enhancement ideas |
| Complaint | Frustration, dissatisfaction, negative experiences |
| Praise | Positive feedback, compliments, appreciation |
Summaries
Modem generates summaries to help you scan topics quickly:- Topic summaries. A high-level overview of what a topic is about and its key themes.
- Attributed summary points. Bullet-point takeaways linked back to the specific messages and authors they came from, so you can trace any insight to its source.
Discussions
A topic can contain multiple discussions: distinct conversations that share the same theme but happened at different times or in different channels. For example, a topic about “API rate limiting” might have:- Initial discussion about rate limit errors (January, Slack #support)
- Feature request for rate limit headers (March, Discord #suggestions)
- Bug report about incorrect 429 responses (May, GitHub Issues)
Working with Topics
Browsing
Navigate to Topics in your dashboard to see all discovered topics, sorted by recent activity. Filter by:- Search. Find topics by keyword or phrase.
- Category. Filter by topic keywords or tags.
- Feedback type. Bug reports, feature requests, complaints, or praise.
- Priority. Very low through very high.
- Company. Topics where a specific company’s users have participated.
Searching
Search uses semantic matching, so you’ll find related topics even if they don’t contain your exact search terms.Topic Details
Click on a topic to see:- All messages organized by discussion
- Attributed summary points linked back to source messages
- People involved and their contributions
- Source channels and platforms
- Change history and any linked Linear issues
