Fund learning, not faith.

Every board of a growth-stage SaaS business is having a version of the same conversation this quarter. The CEO has an AI-enabled revenue idea. The CFO cannot fund it because the business case does not compute. The chair asks whether the alternative to funding is to do nothing, and everyone in the room knows the answer to that is worse.

The pattern is not scepticism. It is category confusion. The business case template every CFO has spent twenty years building is designed to value efficiency investments. It cannot value a revenue play whose evidence base does not yet exist. The board approves what the business case shows. The business case shows efficiency AI. The revenue AI dies in committee before it is ever built.

This paper gives CFOs and boards a way to fund AI-enabled revenue without abandoning the discipline that made them CFOs and board directors in the first place. Stage the investment into four decision points. Attach kill criteria and evidence gates to each. Approve the next stage on evidence produced by the previous one. Fund learning, not faith.

The four patterns

• Productised Expertise. Codify what the senior humans do and sell it as software.
• New Product Surface. A capability the previous stack could not deliver at all.
• New Pricing Model. Price on outcome, not on seat or usage.
• New Reachable Segment. Serve customers the unit economics previously ruled out.

What is inside

• The four revenue patterns in depth, with worked examples of each and how to spot the one your idea sits in
• The six-axis qualifying lens: Pull, Adjacency, Defensibility, Capability, Speed to Signal, Downside Bounded — scored one to five
• The four-stage funding ladder: Frame, Probe, Pilot, Scale. What each stage funds, what evidence it must produce, and the kill criteria that end it
• The CFO memo template: three paragraphs, one budget number, one kill criterion, one evidence gate. Approvable in a single sitting
• Cross-industry worked examples: a SaaS platform funding a Productised Expertise play through Probe and Pilot, and a healthtech that killed a Scale bet at the Pilot gate
• Where the framework does not apply: pre-Series-A, distress situations, and the hobby-project exception

Who it is for

CFOs of growth-stage B2B SaaS businesses being asked to approve AI-enabled revenue investments. Chairs and boards adjudicating between AI efficiency bets and AI new-revenue bets. Private equity operating partners setting portfolio-wide funding discipline for AI-adjacent moves. CEOs who suspect the reason their last AI revenue idea died was the template, not the idea.

Written from an operating perspective

Four exits over three decades (Lotus to IBM, Paragon to Phone.com, Apertio to Nokia $240M, Clearswift to Lyceum) plus COO and CGO seats at Lumeon and Sapio Sciences. Every stage of the funding ladder is drawn from a real investment decision that either produced evidence or produced sunk cost.