Focusing on What Is Likely to Happen

Whereas budgeting primarily sets out what an organization wants to achieve, forecasting attempts to answer a different question:

Based on what we know today, what do we actually expect to achieve?

That difference is fundamental.

A budget can reflect ambition. A forecast must be as objective as possible.

That is precisely why forecasting is one of the most important tools for business control. It reveals when reality begins to deviate from the original plan, even before that deviation is fully reflected in the financial results.

A forecast is not an adjusted budget

In practice, forecasts and budgets are sometimes confused.

For example, when the first few months of the year fall short of expectations, there is pressure to inflate the figures for the remaining months so that the annual result still comes out exactly on budget.

However, this does not constitute a forecast. The original target is simply spread out over a different period.

A good forecast reflects the most likely outcome based on the information available at that time.

If that outcome is below budget, that is not a problem with the forecast. It is management information.

The relevant question then is why the expectation has changed and what actions can be taken.

This distinction between target and expectation is important for the quality of the management dialogue.

When forecasters are held accountable for predicting the desired result, there is an incentive to delay bad news. When they are evaluated on the quality of their estimates, early transparency is actually rewarded.

Actuals are just the beginning

Many forecasts start with actual results, but they shouldn’t end there.

Actuals are, by definition, historical information. For a forward-looking perspective, developments that are not yet fully reflected in the financial figures are particularly relevant.

Order intake, the commercial pipeline, personnel changes, contract renewals, purchase prices, or production planning can provide earlier indications of how revenue and costs will develop.

This is where operational data and finance operations become crucial once again.

A received sales order can signal future revenue before an invoice is recorded. An outstanding purchase order can predict future costs or cash outflows before the invoice is received. Customer payment behavior can provide clues about the future cash position.

A good forecast therefore aims to align as closely as possible with the underlying economic events.

Driver-based forecasting

Forecasting becomes much more powerful when financial outcomes are derived from operational drivers.

Consider a company whose revenue is determined by the number of customers, volume per customer, and average price.

When one of those drivers changes, the model can automatically translate the consequences into revenue and, if necessary, further into margin and cash flow.

The same principle applies to costs.

Labor costs can be based on current staffing levels, planned hiring, turnover, and salary trends. Energy costs can be linked to volumes and rates. Transportation costs can be linked to shipments and per-kilometer rates.

Driver-based forecasting has two advantages.

First, a forecast can be updated more quickly when circumstances change.

Second, it becomes clear why the outcome changes.

A revenue forecast that comes in two million euros lower is, in and of itself, merely an observation. When it becomes clear that 1.5 million of that is due to lower volume and 0.5 million to price, a basis for action emerges.

Frequency depends on volatility

Not every organization needs the same forecasting frequency.

In a very stable environment, a quarterly update may be sufficient. In a highly volatile market, monthly or even more frequent updates may be necessary.

The goal should not be to forecast as often as possible.

Every update takes time and can also introduce noise. The right frequency depends on how quickly relevant drivers change and how quickly management can respond to them.

A useful question to ask is therefore:

When could new information actually lead to a different decision?

If the answer is monthly, weekly forecasting adds little value. When cash or capacity could become critical within weeks, a higher frequency may actually be necessary.

Rolling forecasts can help with this. Instead of looking only as far as the end of the fiscal year, planning is done for a fixed period ahead—for example, twelve or eighteen months.

This eliminates the artificial effect where the forecast becomes increasingly shorter as December approaches.

Forecast accuracy is not an end in itself

A forecast should be as reliable as possible, but forecast accuracy can be misused.

A forecast that turns out to be exactly right is not automatically a good forecast. Perhaps there was simply little uncertainty.

Conversely, a forecast may deviate significantly because an unforeseen event has occurred.

It is therefore more interesting to investigate where and why forecasts deviate systematically.

Are certain business units systematically optimistic? Is revenue consistently adjusted downward too late? Are personnel costs underestimated? Are certain operational drivers insufficiently predictable?

Forecast accuracy thus becomes a tool for improving the forecasting process.

Bias is at least as relevant as absolute deviation.

A forecast that randomly lands slightly above or below reality may be acceptable. A forecast that is almost always too optimistic points to a structural problem in assumptions or governance.

Scenario planning acknowledges uncertainty

Every forecast ultimately reduces an uncertain future to a single set of assumptions.

That’s practical, but it can be dangerous. A single number easily gives the impression that one future is likely set in stone.

Scenario planning therefore adds a second dimension.

Instead of asking only what is likely, the organization examines what might happen if key assumptions turn out differently.

For example, a base scenario can be supplemented with a downside and an upside scenario.

But scenario planning doesn’t have to be limited to three variations of the same percentages.

The most valuable scenarios often describe different economic situations.

What happens if a key customer leaves? If the price of a raw material rises sharply? If an acquisition is postponed by six months? If the market suddenly grows by twenty percent and capacity becomes the limiting factor?

Scenarios become particularly useful when they are linked to management actions.

From Scenario to Course of Action

A scenario that merely shows that EBITDA is three million lower in a downside case is not yet a management tool.

The next question must be: what do we do then?

Which costs can be deferred? Which investments can be reconsidered? How much of a cash buffer is needed? When should action be taken?

This is how contingency planning takes shape.

In advance, we consider possible measures and the circumstances under which they would be implemented.

This is of great value, because decisions made under pressure are often less sound than those that have been thought through in advance.

Scenario planning thus transforms uncertainty from something that must be predicted into something for which the organization can prepare.

The Controller as Guardian of Realism

Forecasting always involves professional judgment.

Data and models can provide a great deal of support, but ultimately, assumptions must be interpreted.

This is where business control plays a key role.

Not by determining every forecast itself, but by scrutinizing the logic.

What new information justifies this change? Why has the commercial forecast increased? What capacity is needed to deliver that volume? What historical conversion rates support the pipeline?

Furthermore, the business controller must be able to distinguish between a target, an ambition, and an expectation.

All three are useful, but they must not blur into one another unnoticed.

A forecast that is politically convenient but economically unlikely is of little use to the CFO.

From Predicting to Anticipating

Ultimately, the added value of forecasting is not determined by how accurate the prediction is, but by whether it enables the organization to act sooner.

A forecast that only confirms at the end of the quarter what everyone already knew has limited value.

A forecast that reveals three months in advance that cash is coming under pressure, capacity is falling short, or margins are declining structurally can influence management decisions.

Scenario planning reinforces that effect by also preparing for alternative developments.

Thus, the main goal is not to predict the future exactly.

It is to ensure the organization is less likely to be caught off guard.

Good forecasting tells us what is likely to happen. Good scenario planning tells us what we’ll do if things turn out differently.