AI in Finance: How to Reduce Decision Latency in Finance and Controlling
CCO Managing Partner
At Chicago Booth we spent a lot of time on a simple idea: the value of information is closely tied to how quickly it is incorporated into decisions. In efficient markets, prices adjust fast. Inside many finance and controlling functions, the opposite still
happens.
I noticed this during nearly 25 years in banking. Important signals often arrived late, fragmented, or already diluted by the time they reached the people who could act. Then I moved into data and AI services in 2019 and began seeing the same problem from the
other side. The technology has improved dramatically. The organisational habit of slow, sequential decision-making has not always kept pace.
That gap is where the real opportunity sits.
Why AI in Finance Is More Than a Cost-Cutting Tool
Many discussions about AI in finance still start with efficiency: faster closes, fewer manual entries, lower headcount. Those gains exist and they matter. But the stronger results I see are elsewhere.
Decision quality is improving. Forecast accuracy is improving. The time between a material signal and a usable response is shrinking. In short, AI is beginning to function as a decision engine rather than just a cost lever.
This matches what recent global surveys are showing. Productivity effects are already visible, especially in technology, data and operations roles. Yet a large share of institutions still find it difficult to measure the enterprise value of what they have deployed.
The organisations reporting clearer benefits tend to be those that directed AI toward judgment-heavy work – planning, forecasting, risk assessment – rather than treating it mainly as process automation.
What Is Slowing AI Adoption in Finance and Controlling?
Data quality and fragmented systems remain the biggest bottlenecks. This is not new. The same constraints appeared in earlier research waves and they are still dominant. Legacy architecture, siloed data and incomplete lineage continue to limit how far even
good models can go.
Talent and internal capability form the second constraint. Many organisations are upskilling existing teams, which is necessary. Fewer are redesigning roles and decision rights so that people and AI systems actually work together at the required speed.
Governance is often framed as a brake. In practice the opposite is visible. Institutions with clearer controls, explicit human oversight and defined accountability are reporting better outcomes. Trust in the output determines whether it is used. Without that
trust, even accurate signals sit unused.
How Finance and Controlling Leaders Can Reduce Decision Latency
If the goal is to reduce decision latency while preserving accountability, a few priorities stand out:
- Direct AI first toward areas where delayed or incomplete information currently weakens material decisions.
- Treat data quality, lineage and access as core infrastructure, not a side project that can be fixed later.
- Keep human accountability explicit on forecasts, risk assessments and capital allocation discussions.
- Measure progress by the reduction in time between signal and decision, not only by the number of use cases live.
- Build enough internal capability that the organisation owns the change rather than permanently renting it.
None of this requires science fiction. It requires treating AI as a change in the operating model rather than a collection of pilots.
How Interim Controllers Can Accelerate AI Transformation
Sometimes the fastest way to move is to add people who already see the opportunity through both lenses – deep controlling or finance experience and real technology expertise. From our competence centre for planning we deploy interim controllers who work inside the existing finance or controlling unit, identify where decision latency is still expensive, and help implement changes that create impact quickly. They do not replace the team. They accelerate it.
Why Faster Decision-Making Is Becoming a Competitive Advantage
Finance theory assumes that valuable information gets incorporated reasonably quickly. Many internal processes still assume the opposite: that important decisions can wait for the next committee cycle, the next reconciled extract, or the next version of the
slide deck.
AI is making that assumption more expensive. The institutions that reduce decision latency while keeping experienced judgment in the loop will create a real advantage. Those that continue to optimise the old sequence will keep receiving last week’s information
– only faster.
That is not a technology problem. It is a design choice.
Where is decision latency costing your finance function most?
Discuss your current Finance & Controlling setup with our experts and identify where AI, better data processes, or interim support could create faster impact.
AI in Finance: How to Reduce Decision Latency in Finance and Controlling
CCO Managing Partner
At Chicago Booth we spent a lot of time on a simple idea: the value of information is closely tied to how quickly it is incorporated into decisions. In efficient markets, prices adjust fast. Inside many finance and controlling functions, the opposite still happens.
I noticed this during nearly 25 years in banking. Important signals often arrived late, fragmented, or already diluted by the time they reached the people who could act. Then I moved into data and AI services in 2019 and began seeing the same problem from the other side. The technology has improved dramatically. The organisational habit of slow, sequential decision-making has not always kept pace.
That gap is where the real opportunity sits.
Why AI in Finance Is More Than a Cost-Cutting Tool
Many discussions about AI in finance still start with efficiency: faster closes, fewer manual entries, lower headcount. Those gains exist and they matter. But the stronger results I see are elsewhere.
Decision quality is improving. Forecast accuracy is improving. The time between a material signal and a usable response is shrinking. In short, AI is beginning to function as a decision engine rather than just a cost lever.
This matches what recent global surveys are showing. Productivity effects are already visible, especially in technology, data and operations roles. Yet a large share of institutions still find it difficult to measure the enterprise value of what they have deployed. The organisations reporting clearer benefits tend to be those that directed AI toward judgment-heavy work – planning, forecasting, risk assessment – rather than treating it mainly as process automation.
What Is Slowing AI Adoption in Finance and Controlling?
Data quality and fragmented systems remain the biggest bottlenecks. This is not new. The same constraints appeared in earlier research waves and they are still dominant. Legacy architecture, siloed data and incomplete lineage continue to limit how far even good models can go.
Talent and internal capability form the second constraint. Many organisations are upskilling existing teams, which is necessary. Fewer are redesigning roles and decision rights so that people and AI systems actually work together at the required speed.
Governance is often framed as a brake. In practice the opposite is visible. Institutions with clearer controls, explicit human oversight and defined accountability are reporting better outcomes. Trust in the output determines whether it is used. Without that trust, even accurate signals sit unused.
How Finance and Controlling Leaders Can Reduce Decision Latency
If the goal is to reduce decision latency while preserving accountability, a few priorities stand out:
- Direct AI first toward areas where delayed or incomplete information currently weakens material decisions.
- Treat data quality, lineage and access as core infrastructure, not a side project that can be fixed later.
- Keep human accountability explicit on forecasts, risk assessments and capital allocation discussions.
- Measure progress by the reduction in time between signal and decision, not only by the number of use cases live.
- Build enough internal capability that the organisation owns the change rather than permanently renting it.
None of this requires science fiction. It requires treating AI as a change in the operating model rather than a collection of pilots.
How Interim Controllers Can Accelerate AI Transformation
Sometimes the fastest way to move is to add people who already see the opportunity through both lenses – deep controlling or finance experience and real technology expertise. From our competence centre for planning we deploy interim controllers who work inside the existing finance or controlling unit, identify where decision latency is still expensive, and help implement changes that create impact quickly. They do not replace the team. They accelerate it.
Why Faster Decision-Making Is Becoming a Competitive Advantage
Finance theory assumes that valuable information gets incorporated reasonably quickly. Many internal processes still assume the opposite: that important decisions can wait for the next committee cycle, the next reconciled extract, or the next version of the slide deck.
AI is making that assumption more expensive. The institutions that reduce decision latency while keeping experienced judgment in the loop will create a real advantage. Those that continue to optimise the old sequence will keep receiving last week’s information – only faster.
That is not a technology problem. It is a design choice.
Where is decision latency costing your finance function most?
Discuss your current Finance & Controlling setup with our experts and identify where AI, better data processes, or interim support could create faster impact.