Speed in decision-making is also part of decision quality. MIT Sloan documents this in a study of more than 260 companies: organisations that reduce the gap between data and decision achieve 62% higher revenue growth and 97% higher margins than those that do not. What distinguishes these organisations is their ability to turn information into action before the context changes.
Most organisations have invested heavily in improving the quality of analysis. Far fewer have invested in redesigning the speed at which that analysis becomes a firm position. And that gap carries a cost that rarely appears in any report.
A good decision executed in a market that has already changed is not simply a good decision made late. In practice, it is a non-decision with good documentation.
What usually causes the delay
Gartner estimates that around 55% of decisions in complex organisations take longer than the context allows. And in most cases, the cause is not a lack of information.
What slows things down is the architecture of the process: layers of validation that add friction without adding real clarity, unclear decision rights that force everything upwards before anyone can close a decision, and approval frameworks designed for annual cycles trying to operate in environments that change every week.
McKinsey has documented this across multiple sectors: when it is unclear who decides what, the result is multiple rounds of review that do not improve decision quality — they simply delay the moment when someone takes responsibility for making the decision.
In a real-time data environment, that delay has a specific cost. Forrester describes the shift clearly: the bottleneck has moved from a lack of data to an inability to turn it into action in time.
Organisations that are addressing this are not doing so by investing in more analytics. They are redesigning how decisions flow — who has the authority to close them, at what point, and with what minimum information — and, in most cases, that has more to do with organisational design than with technology.
What this means in practice
The pattern repeats across very different sectors and contexts. What changes is the name of the window that closes.
In financial services and insurance, the risk is not simply getting the modelling wrong. It is getting the modelling right after the customer has already chosen another provider. Forrester documents how sector leaders prioritise reducing the latency between customer signals and action, because arriving late with a technically well-segmented offer means losing the opportunity.
In telecoms, a well-designed pricing plan or network deployment that comes into effect 12 or 18 months late enters a different market — customers are more price-sensitive, competition is stronger and regulation has changed — and its actual impact is significantly lower than expected. Specialist consultancies in the sector point out that the main problem in 2026 is not technology but decision cycles: delays and cost overruns caused by sequential approval processes, even when the strategic direction is correct.
In industry and energy, data arrives in real time from sensors and SCADA systems. The intervention is approved days or weeks later. Predictive maintenance market studies report that unexpected downtime costs high-volume manufacturers between $50,000 and $200,000 per hour. Plants that implemented predictive maintenance in 2024–2025 report reductions of up to 90% in unplanned downtime. The prediction is correct. The delay in acting turns that accuracy into a cost.
In supply chains, data arrives in milliseconds. The decision arrives in days. Recent analyses indicate that organisations still relying on manual processes and fragmented tools are at a disadvantage compared with those able to turn information into action quickly, with direct effects on profitability and resilience.
In the public sector, the window for impact is defined by social cycles — an emergency, a crisis, a moment of demand — that do not wait for validation processes to catch up. Recent reports on economic policy highlight that, despite greater analytical capacity, a lack of institutional adaptation means decisions can arrive after the problem itself has already changed.
Across all these contexts, the conclusion is the same: the technical quality of a decision does not compensate for the cost of arriving too late.
How organisations are reducing the gap
Organisations that are reducing this gap share several patterns. They know who can close each type of decision before urgency arises. They have defined criteria for action so those criteria do not need to be created under pressure. And they have designed how information reaches the people with genuine authority to act.
The difference often lies less in technology than in how authority is distributed and how analysis is connected to execution.
Three questions to identify where the real work lies
At Cognodata, we frequently observe this pattern in data-driven transformation projects. The hardest part is rarely generating more information. It is usually ensuring that the right information reaches the person who can act within the window in which action can still change something.
- How much time passes in your organisation between data becoming available and someone taking and closing a position?
- Have any recent decisions been correct but arrived outside the window in which they could have had a real impact?
- Who has the authority to close a decision without waiting for the next committee meeting?
If any of these questions takes time to answer, the most urgent work probably does not lie in improving the analysis.
Frequently asked questions about decision-making speed
Why can a correct decision arrive too late?
Because the context in which it was meant to have an impact may already have changed. A decision can be technically correct and still lose value if the customer, market, risk or opportunity has evolved by the time it is executed.
What slows down decision-making in an organisation?
A lack of information is not always the cause. Delays also arise from excessive validation layers, unclear decision rights and approval processes that force decisions upwards when they could have been closed earlier.
How can organisations reduce the time between data and decision?
By defining in advance who has the authority to decide, what minimum information they need and the criteria under which they can act without opening additional rounds of validation.
What is decision latency?
Decision latency is the time between having the information required and turning it into an executable decision. When that latency exceeds the window of opportunity, even a correct decision can lose much of its value.
