CFOs and heads of FP&A who signed off on AI initiatives over the past two years, watched the projects stall, and then faced awkward board conversations are now deeply sceptical about any role that involves touching AI again. We hear it in candidate calls across the DACH region: they will take the meeting, they will listen to the pitch, but the moment a hiring manager cannot explain exactly how the technology will be used and measured, the conversation ends.
Companies trying to build AI-enabled finance functions face a specific challenge here. The candidates with budget authority and operational experience to make these projects work are often the same people who have been disappointed before. A client in Zurich recently described their ideal hire as someone who had ‘survived a failed AI rollout and learned from it’. They want someone who will ask hard questions before committing resources, someone who has seen what happens when the vendor promises outrun the implementation.
In practical terms, ‘AI confidence’ for a finance leader now means knowing which use cases deliver measurable value, which ones remain experimental, and how to tell the difference before the money is spent. We have started asking candidates directly about projects that did not work. The quality of those answers often matters more than the success stories, because it reveals whether the person learned to scope properly or simply moved on to the next role.
Job specifications that lean heavily on ‘AI transformation’ language may repel the exact candidates companies need. The strongest finance leaders we place want AI that solves a defined problem with a clear metric attached. Companies that can articulate that level of specificity in the hiring conversation tend to close candidates faster, because they sound like organisations that have learned the same lessons the candidate has.
Prompted by reporting from CFO Dive.