Getting started

ThreadSonar is social listening with judgment: you describe the conversations that matter to your product in plain English, and it watches X, Reddit, LinkedIn, Threads, Bluesky, and Hacker News for posts that match, scores each one for relevance and urgency, tags who is talking, and delivers the qualified ones as leads. This page covers the human workflow; if you are wiring up an agent, jump to the MCP server docs.

1 · Onboard with your website

Point ThreadSonar at your website during onboarding. It reads what you sell and who buys it, then drafts four listening signals for you: buyers shopping around, competitor complaints, brand mentions, and the pain you solve. You can edit or discard any of them.

2 · Signals

A signal is a sentence, not a keyword list. Write it the way you would brief a colleague: “companies frustrated with their current CRM's pricing who are actively looking to switch.” The compiler turns that into platform-native queries plus a scoring rubric, and shows you a calibration panel of what will and will not match before you save. Pain alone is not intent: someone venting is scored differently from someone asking for recommendations.

Platform polling cadence is decided for you; your plan determines how quickly qualified leads are delivered.

3 · The inbox

Qualified leads land in the inbox ranked by intent score, each with a hot/warm/review band, the original post, a one-line rationale for why it matched, and a reply draft when available. Triage in one tap: 👍 tunes the signal sharper, 👎 credits the lead back to your quota and asks why. Statuses (saved, contacted, dismissed) keep your pipeline view honest.

4 · Quotas and refunds

You pay for qualified leads, not scrapes. Each plan includes a monthly lead quota; a 👎 refunds that lead automatically. Refunded leads are redacted and the 👎 is final, so you cannot keep both the credit and the lead. Usage lives in Settings, with top-ups available any time.

5 · Delivery

Two ways to get leads out of the inbox and into work: connect a Slack webhook in Settings to drop each qualified lead into a channel for a human to action, or create an API key and let an agent pull leads programmatically over MCP or the REST API. Each workspace has its own Slack destination and its own keys. Agents can also self-serve the docs: /llms.txt is the machine-readable index of everything here.

6 · Workspaces

Running listening for several products? The workspace switcher (top left) keeps each one fully separate: its own signals, leads, Slack channel, API keys, and billing. Create a new workspace from the dropdown and onboarding runs again for that product.