Putting AI into customer-facing revenue workflows raises questions that do not surface in a pilot but become critical at scale. This briefing sets out the governance a revenue leader should settle early — not to slow adoption, but to make sure it holds up when security, legal and customers start asking how the system works.
Security: where does the data go?
The first question is always about data. What is sent to the model, where it is processed, whether it is retained and who can see it. Revenue leaders should be able to answer these plainly, because their security teams and their customers will ask.
Privacy: what is the system allowed to use?
Not every signal is fair game. Governance means being explicit about which sources feed the system, honouring the boundaries that come with them, and making sure customer data is used in ways the customer would recognise and accept.
Explainability: can you show your work?
A recommendation that cannot be traced to a source is hard to trust and harder to defend. Every insight the system surfaces should link back to the evidence behind it, so a seller can verify it and a customer can be told where it came from.
- What data is sent to the model, and what is retained?
- Which sources are in scope, and do their terms allow this use?
- Can every recommendation be traced to a verifiable source?
- Who is accountable when the system is wrong, and how is that caught?
Settling these questions up front turns AI from a risk the organisation tolerates into a capability it can stand behind — with customers, with the board and with its own teams.