Use cases / Customer and revenue
AI Sales
More selling time, with outbound messages under control.
01
Problem
Reps lose hours to research and CRM upkeep. Unsupervised outbound AI risks off-brand claims and consent violations.
02
AI capability
Research accounts, summarize call notes, draft personalized outreach and keep CRM records current.
03
Data
The agent works from the context you connect, not from the open internet by default.
- CRM records
- Call and meeting notes
- Approved messaging and pricing
- Public company information
04
Agent
Sales Agent
Prepares account briefs and draft outreach for the rep to send.
05
Workflow
- 1. Account selected
- 2. Research and brief
- 3. Draft outreach
- 4. Claims and consent check
- 5. Rep approval
- 6. Send
- 7. CRM update
06
Governance
Governance runs inside the agent at runtime: policy changes what it can actually do.
Can
- Read CRM accounts in the rep territory
- Draft messages from approved messaging
- Update CRM notes
- Summarize calls
Cannot
- Quote unapproved pricing
- Contact opted-out contacts
- Send external email unapproved
- Edit others accounts
Requires approval
- External messages
- Discounts
- Use of customer references
Records
Agent identity, Data accessed, Model used, Output, Tools called, Policy applied, Decision, Approval, Action, Outcome.
Suggested starting autonomy: L2 Approve
Start at L2: reps approve outbound messages; CRM note updates can move to L3.
07
Outcome
What you measure, so the agent earns more autonomy on evidence:
- Rep time on research and admin
- Reply rate on approved outreach
- CRM completeness
- Messages blocked by policy
Questions
Will it email prospects on its own?
Not unless you raise its level for that action. Outbound messages default to rep approval.
How are opt-outs respected?
Contact status is checked by policy before any draft is released.