When a finance organization starts with AI, the first business case is almost always the same: automate the manual work. Reconciliations, report prep, invoice matching, the close checklist. It's the right place to start — the savings are real, the math is easy, and the board understands it.
It's also the smaller prize.
Automation pays you back the hours you were already spending. Its ceiling is the cost of the work you automate. Necessary, measurable — and bounded. If your entire AI story is "we do the same things with fewer hours," you've bought efficiency, not advantage. Your competitors are buying the same thing from the same vendors.
The bigger prize is what I call force multiplication: using AI to let the people you already have do work that used to be out of reach. The senior FP&A analyst who understands the business cold but doesn't write code — with the right tools and a few weeks of deliberate enablement, that person builds the driver-based forecast model you were about to hire a consultant to build. The controller who's been wanting a real flux-analysis process for years builds one, because the distance between "I know what this should look like" and "it exists" collapsed.
The pattern that matters: automation removes work, force multiplication removes constraints. And in every finance team I've worked with, the binding constraint was never the volume of manual tasks. It was the list of things the team knew they should be doing and couldn't get to — the analysis that never got built, the question that never got answered, the specialist hire that never got approved. That list is where the enterprise value is, and automation alone never touches it.
Force multiplication is also harder, which is why most organizations stop at automation. It doesn't show up as a line item you can cut. It requires picking the right people, not just the right tools — the ones with domain knowledge, curiosity, and enough organizational credibility that what they build gets used. It requires treating their enablement as an investment with a number attached: the avoided hire, the displaced consulting spend, the software you didn't have to buy, the decision made a quarter earlier.
And it changes what the CFO's AI conversation with the board sounds like. The automation story is a cost story: same output, fewer hours. The force multiplication story is a capacity story: same team, bigger mandate. Finance stops being the function that reports what happened and starts being the function that tells the business what to do next — without the headcount request that usually accompanies that ambition.
The sequencing matters, so I'll be precise about it. Start with automation — it builds the muscle, funds the journey, and earns the organization's trust in small, verifiable wins. But start with the end in mind: every automation win should be creating the data foundation and the confidence for the multiplication play behind it. Teams that treat automation as the destination plateau in a year. Teams that treat it as the on-ramp compound.
The hours you save are table stakes. The capabilities you unlock are the game.