The metric traditional budgeting does not measure

Every year, planning restarts from the same question: how accurate will the budget be? It is the wrong question. In a context where volumes, energy prices, rates, and supply chains shift several times within the same quarter, a budget that is accurate at kick-off says little about how well the company will stay aligned with reality by June.

The variable that makes the difference is not forecast error, but the speed of the decision cycle: how quickly the organization turns new information into a new allocation of resources. We keep optimizing the wrong metric, refining the sharpness of a photograph in a world that moves like a film.

What Decision Latency is

We call Decision Latency the time an organization takes to travel the full loop: a shock or a shift in a driver, detection, simulation, decision, reallocation. It is the distance, measured in days, between the moment information becomes available and the moment it turns into a concrete move.

It is a quantity a CFO can estimate right away, with a simple question: how many days passed, last time, between a relevant event and the resulting decision? That number, not forecast error, is the real indicator of how ready the company is for 2027.

The numbers tell a story about time, not error

Industry data, usually cited to argue that “the budget is obsolete,” actually describes a problem of time. According to the Association for Financial Professionals, more than half of finance leaders (53.9%) report a budget already exceeded by mid-year. 46% of teams’ time goes into collecting and reconciling data, and 96% of companies still plan on spreadsheets.

The budget cycle length, which according to APQC ranges from 25 days for top performers to 56 for the slowest, is a direct proxy for rigidity: it shows how slowly, and how rarely, the organization can revisit the plan. Read together, these numbers do not measure forecast quality: they measure components of time.

The four drivers that can break the 2027 plan

2027 is not “uncertain” in the abstract: it is exposed to four drivers that can make a plan built on a single assumption fragile. Demand, with weak growth (Italy around +0.5/0.6%, the euro area near 1%). Energy, with volatile prices and possible shocks. Rates and inflation, with the cost of money and financial charges weighing on cash and covenants. Trade, with exceptionally high uncertainty affecting supply chains and exports.

The useful question is not what the average value of each driver will be, but which decisions break if the driver moves, and at what threshold. A number becomes actionable only when it is tied to a threshold and to a decision that is already prepared: not “energy rises 6%,” but “if the price crosses X, hedge Y triggers.”

From budget-as-document to budget-as-continuous-cycle

Reducing Decision Latency means changing the very nature of the budget. No longer a document approved once a year, but a continuous decision cycle: a set of thresholds and triggers that says, in advance, what to do when a driver moves.

The measure of success shifts from accuracy to adaptability; the cadence from the static annual budget to a rolling forecast with on-demand simulation; the selection criterion from the theoretical optimum to the best executable plan. The budget stops describing the world and starts to pre-decide the moves.

What changes for CFOs, controllers, and CEOs

For the CFO and management control, it means measuring the speed of the decision cycle as a KPI, moving to a driver-based budget with rolling forecasts, and automating data collection to return hours to analysis. The controller becomes the owner of the thresholds: the person who ties each driver to a trigger and keeps the plan aligned with real data.

For the CEO and the board, it means asking for the budget as a range with triggers and breaking points, not as a fixed promise: governance shifts from after-the-fact control to managing thresholds. A company that states its thresholds, triggers, and risk profile is also more legible to investors and, all else equal, more defensible.

The role of AI in the decision process

Regenerating scenarios, exploring combinations of levers, updating drivers continuously: on a spreadsheet this is economically prohibitive. Here, artificial intelligence is not the machine that guesses the future, but the infrastructure that makes it sustainable to decide often instead of once a year, lowering the marginal cost of every new simulation.

Caution remains part of the thesis: data quality is the first obstacle, and a decision infrastructure is only as good as the data feeding it. AI structures the decision and shortens its timing; it does not replace judgment, which stays with leadership.

KEY TAKEAWAYS

• In 2027 competitive advantage goes not to the best forecaster, but to whoever shortens the time between a change and the resulting decision.
• Decision Latency is the time a company takes to travel the loop: shock, detection, simulation, decision, reallocation.
• More than half of finance leaders have a budget already exceeded by mid-year: a problem of time, not of forecast error.
• A driver becomes actionable only when tied to a threshold and a decision that is already prepared.
• The budget stops being an annual document and becomes a continuous decision cycle of thresholds and triggers.

FREQUENTLY ASKED QUESTIONS (FAQ)

What is Decision Latency?

It is the time an organization takes to turn a change in its business drivers into a decision and a reallocation of resources. The lower it is, the more the company governs uncertainty instead of enduring it.

Why is budget accuracy no longer enough in 2027?

Because in a context that changes several times per quarter, a budget that is accurate at kick-off says little about how aligned the company will stay in the following months. What matters more is how fast the plan is updated.

How do you reduce Decision Latency?

By automating data collection, adopting a driver-based rolling forecast, and turning the budget into a system of thresholds and triggers that pre-decide the action.

‍