Client Path documentation
Everything you need to research prospects, run the agent, and close deals faster.
Client Path lets you research B2B prospects against your offering, score their propensity to purchase, and draft outreach. An agent handles this in the background while you focus on relationships.
Where to start
If you are new, follow this path:
- Overview: understand the mental model in five minutes.
- Sign in: set up your passkey and access your workspace.
- First research run: import a client, start a run, and read your first findings.
Main surfaces
| Surface | What it does |
|---|---|
| Research runs | Batch research of prospects against an offering |
| Automations | Declarative trigger-step automation |
| Task inbox | Review and approve agent-proposed actions |
| Signals | External event feeds that trigger automations |
| Outreach | AI-drafted messages anchored to findings |
Key concepts
Understanding a few terms will make the rest of the docs click:
- A client is a company you are prospecting.
- A research run is a batch job that researches many clients against one offering.
- A finding is a discrete piece of intelligence about a client, such as budget signals, buying team references, and technology fit gaps.
- A propensity score is a 0-100 rating of how likely a client is to buy your offering right now.
- An automation is a trigger-step pair that lets the agent propose or automatically take pipeline actions when something changes.