AI in Finance
How AI Is Changing What Finance Directors Actually Do
13 April 2026
The finance director role has always been described as the strategic partner to the business. In most organisations, the reality has been something different.
A senior accountant who spends a disproportionate proportion of their time producing and reviewing information rather than using it. Overseeing the month-end close. Reviewing management accounts before they go to the board. Managing the audit relationship. Maintaining compliance. Supervising a team that spends most of its time on transaction processing.
Important work. Not inherently strategic.
The “strategic partner” framing has been aspirational for a long time. In most finance functions I have worked with, the aspiration and the daily reality are some distance apart.
AI is closing that gap. Not by making finance directors redundant: the argument that AI replaces the FD is based on a fundamental misunderstanding of what the FD is actually supposed to do. But by removing a substantial layer of work that finance directors have traditionally done, or supervised others doing, the technology is forcing a genuine reckoning with what the role is actually for.
That reckoning is overdue.
What the FD Role Has Actually Been
In a typical SME or mid-market organisation, the finance director’s week looks something like this. A significant portion of the month is oriented around the close: chasing numbers, reviewing journals, checking that the management accounts reconcile, preparing or reviewing the board pack. A further portion is spent on compliance: VAT, payroll, corporation tax, statutory accounts, audit preparation. Another portion on team management: answering questions, reviewing the work of junior staff, handling the issues that escalate up.
What remains, the time actually spent on commercial questions, investment analysis, strategic planning, is often the smallest slice of the week.
This is not a criticism of finance directors. It reflects how finance functions have been structured. The tools have required it. The audit trail has demanded it. The compliance environment has enforced it. Finance has been organised around the production of information, and the FD has sat at the top of that production process.
The “strategic partner” title reflects what the organisation needs the FD to be. The actual job description has often reflected something narrower: the person responsible for making sure the numbers are right and the compliance obligations are met.
Both of those things matter. But they are not the same as strategic partnership. The gap between them has been a persistent source of frustration for finance directors who want to do more and boards who wonder why they are not getting more.
What AI Removes From That Picture
The capabilities that exist today are not theoretical. I want to be specific about what they are and what they replace.
Reconciliation oversight, when AI handles matching at 85 to 95% automation rates, means the FD’s time spent checking that the bank reconciliation has been done correctly is largely gone. The exception queue is smaller. The human review is focused on genuine anomalies, not routine confirmation that the process ran.
Invoice processing review, when intelligent document processing handles structured AP end to end, means the FD is no longer reviewing a stack of invoices before payment runs. The controls are built into the tool. The review is of exceptions the tool has flagged, not of the full population.
Variance analysis, when AI flags significant variances and provides first-pass explanations, means the FD is starting from “here is what moved and here is the likely reason” rather than “here is the data, now work out what happened.” The analytical layer is faster and the FD’s input is focused on the interpretation, not the identification.
Board pack assembly, when AI is pulling and formatting the data, means the FD’s time on board preparation shifts from hours of production to a much shorter review of the output.
None of this is speculative. These capabilities exist in deployable form today. The organisations that have built the right foundations, clean data, documented processes, appropriate governance, are using them. See the agentic AI post for the fuller picture of where this is heading.
The time that is freed up is real. In a finance function with mature AI adoption, the estimate I work with is 30 to 40% of the FD’s current time commitment to production and oversight tasks. That is a material shift.
What That Leaves
This is the important part.
What remains when you remove the production and oversight layer is judgment. Commercial judgment. Strategic judgment. The decisions that cannot be automated because they require context, relationship knowledge, organisational understanding, and the kind of pattern recognition that comes from years in the role.
Should we invest in this market? Is this customer worth the credit risk? Is the management accounts picture telling us the true story or a flattering one? Should we raise the question with the board now or wait for more data? Is this acquisition price reasonable given what we know about the target’s actual financial position?
These are FD questions. They always were. AI does not answer them. It frees up the time and attention to ask them properly.
The FD who spends less time on information production and more time on information use is a more valuable FD. This is what the role has always been supposed to be. The full description of what an FD actually does covers the judgment layer in detail. The point here is that AI is making the gap between the aspiration and the reality smaller, in practice, for the first time.
There is a related effect worth naming. When the FD is less occupied with production oversight, they are more available. More available for conversations with commercial and operations teams. More present in strategic discussions. More useful to the CEO as a thinking partner. The AI adoption is not just changing what the FD does in isolation. It is changing how the FD connects to the rest of the organisation.
That connection is where the strategic partnership actually lives.
The Capability Gap This Exposes
Some finance directors have built their value primarily in the transaction layer. Deep knowledge of how the reconciliations work. Strong control over the close process. Expert management of the audit relationship. The ability to navigate complex compliance requirements with precision.
These are real skills. The organisations that have them have benefited from them. But AI is making those skills less scarce. The reconciliation expertise that took years to develop in a specific system context is less differentiating when the system handles the reconciliation. The close process mastery is less valuable when the close has been restructured around automation.
The FDs who have built their value primarily in the judgment layer are going to find their position strengthened by AI adoption. Commercial acumen, strategic thinking, board-level communication, the ability to translate financial complexity into clear decisions that non-finance leaders can act on: these do not become less valuable when AI handles the transaction layer. They become more visible, because the transaction layer is no longer occupying the time the judgment layer should be using.
The FDs who have built their value primarily in the transaction layer need to develop the judgment layer capabilities. This is not a dramatic statement. It is a practical one. The window to develop those capabilities is not unlimited. Development does not happen by waiting. It happens by taking on more commercial work, by being more present in strategic conversations, by building the relationships and the pattern recognition that strategic judgment requires.
The question of when to hire a finance director is partly a question about what kind of FD an organisation needs. The answer to that question is shifting. Boards are going to want less of the transaction oversight and more of the judgment. FDs who are ahead of that shift will be in demand. The ones who are not will find the market getting harder.
What Good Looks Like in 2026 to 2028
The finance director role in a finance function with mature AI adoption will look like this.
More time with commercial and operations teams. The FD is in the room when pricing decisions are being made, when new customer contracts are being negotiated, when operational investments are being evaluated. Not reviewing the financial model after the decision has been made. Shaping the decision as it happens.
Deeper involvement in investment and strategic planning. The FD’s input is earlier in the process and more substantive. The analytical work that used to take days takes hours. The FD’s capacity for scenario analysis and sensitivity testing is higher.
Closer connection to the board on strategy. The board pack is produced faster and with less FD time. The FD uses the time saved to have better conversations with board members between meetings, to understand what questions are coming, to prepare the kind of insight that goes beyond the numbers.
Oversight of AI-generated outputs rather than production of manual ones. The FD is reviewing and approving what the AI has produced, understanding its limitations, knowing when to push back and when to rely on it. This is a different skill set from manual production. It is the skill set that matters going forward.
Fewer hours on administration. More hours on judgment. That is the direction of travel. For finance directors who have wanted the role to be more strategic, it represents a genuine opportunity.
The condition is readiness. The AI readiness assessment is the right starting point for understanding where a specific finance function sits and what the path forward looks like.
Where This Leaves the Role
The finance director role is not disappearing. The argument that AI makes the FD redundant misunderstands what the FD is for. The judgment layer, commercial, strategic, relational, is not automatable. It is exactly what boards and CEOs need from senior finance leadership.
What AI is removing is the transaction and production layer that has occupied too much FD time for too long. That removal is a clarification. It makes the role more honest. The “strategic partner” aspiration becomes the actual job description.
That is an improvement. The finance functions that move there first will have more effective FD relationships, better strategic input, and faster decision-making. The ones that do not will find the gap between what they need from their FD and what the FD has capacity for getting wider.
If you want to understand what AI readiness looks like for your specific finance function, the AI in Finance Strategy page is a good starting point. The assessment is concrete. It starts with where you are, not where the vendors say you should be.
Maebh Collins is a Fellow Chartered Accountant (FCA, ICAEW) with Big 4 training and twenty years of operational experience as a founder and senior finance leader. She writes about AI in finance transformation from the inside out.