Sales agents
Read what it is allowed to do.
Every agent has a profile: its purpose, its channels, what it may do unattended, what it may never do, and the person accountable for it. This is the screen somebody signs off before the agent speaks to a customer, so it is written in commercial language rather than configuration.
Purpose in one line
“Qualify hotel prospects.” Not a list of capabilities — a job, so it is obvious when the agent is being used for something it was not built for.
Can and cannot, as two lists
Introduce the product, answer approved questions, send standard pricing, book meetings, follow up. Cannot negotiate custom terms, discount beyond 5%, or commit to a service level.
A named human owner, required
An agent cannot be deployed without one. The owner receives its approvals, its escalations and anything it gets wrong.
Every action records which agent and which version
So a conversation from March can be explained by the rules in force in March rather than the ones in force now.
AI sales agents
The rest of this area.
Consistent follow-up, with authority that stops where it should.
Human handoff
Everything discussed transfers with the conversation.
Commercial authority
Standard pricing, a discount ceiling, and a hard stop.
Approvals
A discount request with the reasoning attached.
Communication policy
Hours, disclosure, attempts, opt-outs, consent — enforced.
Commercial promises
“I'll send that tomorrow”, captured and tracked.
Make sure every opportunity gets a next action.
Let AI handle the repetitive engagement, keep follow-up moving, and bring your team in when a customer needs a real conversation.
Agents identify themselves on every call · Nothing commits without you · [email protected]