The annual budget remains an established practice in companies… but more and more finance departments are recognising its limitations: it is rigid, slow to draw up and quickly becomes obsolete in the face of a changing environment. The solution? The rolling forecast: a rolling forecasting method that can be continuously updated, enabling performance to be managed at a more realistic pace. Find out how to move from a static budget to truly agile management.
Inflation, energy prices, geopolitics, shifts in demand, new technologies… The financial calendar is still organised in twelve-month financial years. The economy, however, is far less so.
In March 2026, the Banque de France’s uncertainty indicator – based on feedback from 8,500 business leaders – surged to the levels seen in the early months of the war in Ukraine (Bank of France). In the spring, 76 % of the 1,136 European finance directors surveyed by Deloitte considered external uncertainty to be high or very high, and 51 % were using scenario analysis (Deloitte).
In this context, one question springs to mind: Is it still possible to manage a business effectively with a budget overview drawn up just once a year?
This is precisely where the rolling forecast comes in.
The annual budget is reaching its limits
The budget isn’t the problem; that’s all we ask of it: to forecast, allocate resources, set targets and assess managers. These tasks have different timeframes, and a forecast drawn up in October won’t hold up for fifteen months when energy prices, interest rates or demand fluctuate every quarter.
Researchers in management accounting have long been aware of this paradox, which was first articulated by Anthony Hopwood as far back as 1974: It is when the economic climate becomes uncertain that businesses feel the greatest need for a budget, even though the accuracy of its forecasts depends on a stable economic climate (CREG, Versailles Education Authority).
Despite everything, French companies remain very budget-conscious: At the end of 2023, 91 % of the 40 major groups surveyed by Eight Advisory were still using a traditional budgeting method, whilst only 20 % were using a rolling forecast (Daf Mag).
However, the planning horizon has shortened: 61 % of organisations are only planning six months ahead (OneStream).
In other words, many companies still think in terms of a 12-month financial year, even though their actual time horizon has become half-yearly, or even quarterly.
Rolling forecast: always forecast over the same time horizon
The principle is simple. An annual budget generally runs until 31 December of that year. As the year progresses, its timeframe gets shorter: twelve months in January, six months in July, and just a few weeks in December.
The rolling forecast works differently. Each time the data is updated, a new period is added to the forecast horizon.
A company that uses a rolling 12-month forecast and revises its forecasts every quarter will therefore add three new months to each financial year. It always maintains a 12-month outlook. Some organisations work on an 18-month, 24-month or eight-quarter basis.
It is worth clarifying the terminology, as these three roles often coexist within the same teams:
| Annual budget | Revised (reforecast) | Rolling forecast | |
|---|---|---|---|
| Horizon | Exercise N | End of financial year N | Constant (12 to 24 rolling months) |
| Frequency | Once a year | 2 to 4 times a year | Monthly or quarterly |
| Question asked | What are we aiming for? | Where will the exercise end up? | Where are we really heading, beyond this exercise? |
| Usage | Objectives, budgets, evaluation | Landing control | Forecasting and decision-making |
The change appears to be technical. It is, above all, a managerial one.
A forecast is no longer simply an estimate intended to match the original budget. It becomes a regularly updated picture of what is actually likely to happen.
A target indicates where the company wants to go. A rolling forecast indicates where it actually seems to be heading.
Confusing the two sometimes leads managers to «negotiate» forecasts, to delay announcing bad news or, conversely, to maintain a safety margin. Distinguishing between forecasts and targets allows the rolling forecast to fulfil its role to the full: to provide the most realistic view possible of the future.
The Commercial Code already requires this to some extent
Companies with at least 300 employees or a turnover of €18 million must draw up a projected profit and loss account within four months of the financial year beginning, and then revise it in the second half of the year (Articles R232-2 and R232-3). This account «may include one or more variants»: the legislator had anticipated these scenarios. A well-managed rolling forecast makes this obligation a by-product of management.
Rolling forecasts do not spell the end of the budget
This is one of the main misunderstandings surrounding the method.
Switching to a rolling forecast does not mean doing away with the budget. This is the main misconception surrounding the method, and French practitioners often point this out. At Eight Advisory, it is presented as a management tool that complements the budgeting approach; Pierre Gauthier, a partner, compares it to a toolbox that can be used on an ad hoc basis, applying it only to certain aggregates.
The same caution should be exercised in the debate on the beyond budgeting. Nicolas Berland, a professor at Paris Dauphine University, points out that going «beyond the budget» does not mean managing without a budget, let alone without any oversight. Emmanuel Millard, then chairman of the DFCG, explains that the company retains a few key indicators: turnover, EBITDA, profit margin and net profit (Daf Mag, February 2023).
The division of roles then becomes clear:
- the budget sets out the objectives, allocates certain budgets and serves as a basis for evaluation; ;
- the rolling forecast provides the likely trajectory and allows time to react.
The main change is the question asked in committee. Faced with a slowdown in sales or a rise in the cost of an input, we no longer simply ask, «Why are we off budget?». Instead, we ask: «What does this information mean for the next twelve or eighteen months, and what decisions do we need to make now?»
Finance departments have made this a priority. In Gartner’s survey on CFOs’ priorities for 2026, 51 % ranked improving the quality and accuracy of forecasts amongst their top five priorities (CFO.com, December 2025).
Example: a medium-sized enterprise facing the energy crisis
Illustrative case study; figures are fictitious. Let’s take a medium-sized plastics manufacturing company as an example: €60 million in turnover, a budgeted EBITDA of €6.0 million and an annual energy bill of €3.6 million, or €0.3 million per month. Its net debt of €15 million is subject to a bank covenant: a net debt-to-EBITDA ratio of 3.0x or less as at 31 December.
At the end of March, the management accounting department draws up a forecast covering the period from April 2026 to March 2027, based on four key factors: energy prices, volumes, rates and the date on which these are passed on to customers.
| Script | Assumptions (April–December) | EBITDA 2026 | Lever | Decision |
|---|---|---|---|---|
| Budget | Energy at November prices | €6.0 million | 2,5x | - |
| Central | Energy +20 %, half of which will be passed on in July | €5.6 million | 2,7x | Renegotiating the indexation of contracts |
| Unfavourable | Energy +35 %, volumes −5 %, impact in October | €4.4 million | 3,4x | Defer €1 million in capex, negotiate a waiver from June |
Calculation of the worst-case scenario: additional energy costs −€0.95 million; loss of margin on variable costs (5 1Q–3Q of 45 million in turnover at a 35 1Q–3Q margin rate) −€0.79 million; impact from October to December +€0.16 million; resulting in a reduction in EBITDA of €1.6 million.
Three lessons for finance departments:
- The risk is apparent in April, not in the following January. The annual budget would only have revealed the breach of the covenant at the end of the financial year; the forecast allows eight months to take action.
- Every scenario has its trigger. Example: if energy costs exceed +30 % for two consecutive months, the contingency plan is triggered without the need for a new committee meeting.
- The legal requirement becomes a by-product. For a company subject to Article R232-3, the revised forecast profit and loss account due in the autumn is already ready, including alternative scenarios.
From management accounting to decision-making
This is undoubtedly the most far-reaching change.
In a traditional organisation, financial meetings devote a great deal of time to reviewing the past: actual figures versus budget, analysis of variances, and explanations for variations.
The rolling forecast gradually shifts the focus of the discussion towards the future.
If volumes fall by 5 %, what will be the impact on the margin? If a supplier raises its prices, when will profitability be affected? If the order book grows rapidly, will we need to recruit, invest or increase production capacity? If a product launch is delayed by three months, which expenditure needs to be rescheduled?
The forecast then becomes less of a financial document and more of a simulation and decision-making tool.
This development is in line with the current expectations of finance departments. In a Gartner survey of more than 200 CFOs in 2025 regarding their priorities for 2026, 51 % organisations ranked improving the quality and accuracy of forecasts amongst their top five priorities.
But speeding up the forecast is not enough.
The new AI paradox
In August 2026, Gartner warned of a new paradox: artificial intelligence now makes it possible to produce and update forecasts ever more quickly, but This pace may exceed the ability of managers to understand the changes and make decisions.
Technology is therefore only of value if the assumptions remain explainable and responsibilities are clearly defined.
A few key variables rather than thousands of lines
A rolling forecast comprising several thousand lines, updated every month, quickly becomes an annual budget repeated twelve times. The method works when it focuses on the drivers (business drivers), or key variables, which are the real drivers of performance. Ten to twenty key drivers per activity are generally sufficient.
| Business model | Typical key variables |
|---|---|
| Industry | Volumes, raw material prices, energy costs, hourly productivity, exchange rates |
| Services, digital services, consultancy | Billable staff, occupancy rate, average daily rate, recruitment lead time |
| Distribution | Footfall, conversion rate, average basket value, markdowns |
| Subscription, SaaS | New customers, monthly recurring revenue, churn rate, customer acquisition cost |
The improvement in quality is measurable. In the OneStream 2026 study, 77 % of teams working on dynamic or driver-driven models rate their forecasts as good or excellent, compared with 27 % for basic models. However, only 15 % have a model built entirely on key variables.
For accountants, this represents a real shift in thinking. The forecast is no longer drawn up account by account, in the order of the general ledger, but is based on operational variables, before being presented in financial terms.
Interim management figures (sales margin, value added, EBITDA) provide a good overview : aggregated enough to remain readable, standardised enough to be reconciled with the accounts.
Getting started without all the red tape
The temptation is to overhaul everything at once: ERP, planning tools, reporting and organisation.
It’s better to stick to five simple principles:
- Start with the decision that needs to be informed: cash flow, profit margin, recruitment, investments.
- Select between ten and twenty inducers.
- Adjust the investment horizon and frequency to suit the sector’s volatility.
- Assign a threshold and an action to each scenario.
- Start with a pilot project: payroll, cash flow or a business unit.
What about the accountants? Their role is growing rather than diminishing. They ensure a swift and reliable monthly close, without which the forecast would be based on out-of-date data. Each month, they reconcile the forecast with the actual figures to assess the quality of the assumptions rather than simply tracking variances. And they translate the forecast result into cash flow, via working capital: this is often where the forecast reveals what the profit and loss account conceals.
Budget or rolling forecast? That’s not the real issue
The debate is sometimes portrayed as a head-to-head: annual budget versus rolling forecast. This is probably a false dichotomy.
A budget can still be useful for providing a framework, setting targets and committing certain resources. The rolling forecast meets a different need: to maintain an up-to-date view of the trajectory and give the company time to react.
In the current climate, this need is becoming hard to ignore. The Deloitte study from spring 2026 shows that European CFOs are simultaneously strengthening cost discipline, resilience and scenario analysis.
Ultimately, the change in philosophy can be summed up in a single sentence: We are no longer simply asking the finance department to explain what has happened. We are asking it to help the company decide what it will do if the situation changes. Because, in any case, it will change.
