A budget is a decision, not a forecast

At the end of the budget period

Almost everywhere the same question then hangs in the room: did we land over, under or exactly on budget? It sounds so obvious that hardly anyone notices what it quietly assumes. It treats the budget as a forecast you can hit or miss, and it makes the deviation from that forecast the measure of whether the money was managed well.

Yet the budget was never a forecast. A budget is an estimate made under uncertainty, and it is one more so today than ever, because markets, prices and technical foundations shift faster than any yearly plan can capture. Whoever treats a budget like a weather report ends up measuring mostly how uncertain the world was, not how good their own decisions were.

What a budget really is

It pays to think about the budgeting process in a radically different way, namely as one more tool in a company’s risk management. A budget does not describe what will happen; it sets how much risk you are willing to take at a given point. It is the deliberately drawn frame within which a team can act without checking back on every expense, and beyond which a new decision is due.

Read this way, a budget is a decision, not a prediction. The number does not say “this is what we will spend”, it says “this is what we are willing to invest, because we expect something in return”. That shifts attention away from the after-the-fact check of whether the number was hit, towards the decision made up front about why this number and not another one is the right one.

How much risk, and where

If a budget sets how much risk you take, then not every euro is equally risky. The money that keeps a running product alive, meaning infrastructure, licences and operations, is fairly predictable and moves mainly with usage. The money for a new initiative, by contrast, is a bet on an assumption, for instance that a feature will bring new customers or that a market will open up, and that bet can be lost entirely. Whoever folds both into a single number loses sight of where the real risk lies.

For every initiative it is therefore worth writing down three things before it starts. First, the assumption it rests on, stated concretely enough that it can later turn out right or wrong. Second, the stake, meaning the amount of money and effort you are prepared to lose if the assumption does not hold. Third, the point at which you decide afresh whether to carry on, change course or stop. An expense thereby becomes a bounded stake whose worst outcome is known and bearable from the outset.

For a small team in particular this limit matters, because a single initiative that runs on without a boundary can easily tie up half a year. A defined stake does not protect you from mistakes, but it makes sure a mistake stays small and is recognised as one in time.

Putting it into words makes it tangible

The real value appears while you are still putting it into words, long before any money moves. The moment you have to translate a plan into the question of how much it will cost and what it will bring in, it loses its vagueness. A “we will grow” turns into the less comfortable question of what price that growth comes at, what return makes it add up, and with which resources and people it can be done at all.

An example makes this concrete. A small product team sets out to grow from one thousand to five thousand active users over the next year. As an intention, that is a slide. As a budget it becomes the statement that this step costs roughly twenty thousand euros of additional infrastructure per year and one more developer, that it only pays off above a certain price per user, and that it means not tackling another plan during that time. Added to that is the agreement to decide afresh after six months, based on the actual numbers. The number itself may turn out wrong, but the questions it forced were right.

Money was never an enabler

This is the core that strips the budget of its bad reputation. Money was never an enabler, it has always been a disabler, because the available frame works precisely by ruling something out. You cannot do everything at once, and only the limit forces the decision about what you focus on and what is deliberately left undone.

Focus does not come from good intentions, it comes from a frame that makes the alternatives visible and expensive. That is exactly why a tight budget is not an obstacle for a team but often its greatest help, because it turns the question “what do we leave out” from an awkward exception into a normal question asked all the time. A frame that permits everything steers nothing.

Effective, not merely met

From this follows a different measure of a good budget. What counts is not whether a target was met, but whether it shaped decisions. A budget hit to the euro at year end that never changed a single decision was ineffective, while a budget clearly overshot can have been valuable, if the overshoot came from a conscious, well-reasoned decision.

This is the same stance with which we approach rules in general: a rule is worth only as much as it actually does in daily work, not as much as it is cleanly observed on paper. In that sense a budget is a form of governance that works when it triggers conversations, and turns into decoration when it is merely ticked off.

The budget as a corridor

For a budget to trigger decisions rather than merely count deviations, it needs a form that can do more than a single number. What has proven itself is a corridor, meaning a planned value with a lower and an upper edge around it. Inside the corridor the responsible team decides on its own, without checking back and without having to justify itself. When the actual course leaves the corridor, that is not a breach but the moment, agreed in advance, at which you decide afresh together.

Each edge carries its own meaning. Staying below the lower edge means either the team worked more economically than expected or an initiative has stalled, and in both cases money frees up that can do more elsewhere. Going above the upper edge raises the question of whether to deliberately add more, because the initiative is doing better than expected, or to stop, because costs are growing without anything in return.

This is the same mechanism we describe for metrics in the article Getting KPIs right, where a target corridor with risk and opportunity thresholds turns a bare number into a tool. With a budget, one more property makes it especially useful: the width of the corridor is itself a decision, because it sets how much room the team gets without checking back. A narrow corridor means a lot of coordination and little risk, while a wide corridor means trust and speed.

A deviation is a signal

If the budget is a decision under uncertainty, then a deviation is not a breach but a piece of information. Landing over or under the planned frame means, at first, only that the world behaved differently than assumed, and the genuinely valuable question is what you learn from it about your own assumptions. In practice most deviations come down to three patterns, and each calls for a different response.

The first pattern is growth that arrives faster than planned. Costs are above plan, but revenue grows with them, and the cost per user stays stable or even falls. Here the assumption was too cautious, and the right response is usually to deliberately add more rather than to brake the growth with a rigid budget.

The second pattern is costs running away without anything in return, for instance through forgotten resources, an architecture that scales inefficiently under load, or a contract nobody keeps an eye on any more. Raising the budget does not help here; only finding and fixing the cause does. This is exactly what the view of cost per product is for, which we describe in the article FinOps is not a tool problem.

The third pattern is the most uncomfortable, because it often does not show up as a deviation at all. The initiative is exactly on plan but does not deliver what was expected of it, because the underlying assumption does not hold. A budget met to the euro then hides the fact that a new decision is long overdue, and this is precisely where it shows why the measure is the assumption and not the number.

Only this interpretation turns the number into an insight, and this is exactly where the budgeting process becomes a learning system rather than a yearly reckoning. An alert that merely fires does not replace that interpretation.

The same team, six months later

How this fits together shows when you think the example from the beginning half a year further. The team wanted to grow from one thousand to five thousand active users, planned roughly twenty thousand euros of additional infrastructure per year for it, so just under 1,700 euros a month, and agreed on a decision point after six months. With an even course, about three thousand users would be expected at the halfway mark.

In the first course, four thousand users are already in the product after six months, and infrastructure costs about a third more than planned. On paper the budget is overshot, yet the cost per user has stayed stable and revenue is above plan. The agreed decision here is not to cut costs but rather to hire the second developer earlier, because the initiative is carrying better than assumed.

In the second course, costs are exactly on plan, even though only eighteen hundred users have arrived. On paper everything is fine, and that is exactly where the danger lies. The assumption that this initiative carries the growth has not been confirmed, and the real question is whether to change course, for instance on price or target group, or to steer the remaining money into a different initiative. Whoever looks only at whether the budget was met would be praised in this case and would still have invested half a year in the wrong direction.

Both courses show the same thing, namely that “over budget” or “on budget” says little on its own. Only the comparison with the assumption and the decision point agreed in advance turn it into a basis you can act on.

Rhythm and ownership

A yearly budget alone is usually too sluggish for a small product company, because assumptions are overtaken several times within twelve months. A plan renegotiated all the time would, conversely, be too much effort for a team that actually wants to build a product. What has proven itself is therefore a combination of a yearly frame that sets the broad lines and separate decision points for each initiative, timed by its risk rather than by the calendar.

Just as important is that every frame has exactly one responsible person. They may decide within the corridor without asking, and they call the decision point when the corridor is left or the agreed date arrives. A budget that everyone is responsible for is, in practice, steered by nobody, so a deviation only surfaces at the end of the period.

Which initiatives get a frame at all is not a question of the budget but of the strategy. It sets what a company does and what it deliberately does not do, as we describe in the article A strategy you can decide by, and the budget translates that decision into money and effort. When the two drift apart, the budget is usually the first place where it shows that the strategy does not hold up in daily work.

The first step

Getting started is smaller than the reframing sounds. It is enough, with the next plan, not to stop at “we will do this” but to finish the sentence: it costs roughly this much, we expect that in return, in exchange we deliberately give up something else during this time, and on this date we decide afresh. That one fully stated sentence does more than any reporting tool, because it makes the decision visible before money and effort move.

Once you have taken this stance, the budgeting process no longer has to be endured as a tiresome duty but can be used as what it can be: a tool that names risks, forces focus and moves decisions to where they are made with good reason. If you want someone alongside you who brings this view of cost, risk and operations together, we are glad to work through it with you.