How do I justify an ERP budget to the management board?

An ERP budget gets approved when three things are on the table: what the current state costs every year, which business goals are unreachable without the project, and which risk grows with every year of delay. A list of features convinces no management board. A comparison between the cost of staying and the cost of acting does.

Where an ERP genuinely saves money: three items that appear on no invoice

What you have to reckon with: the four cost blocks of an implementation

Making the cost of the current state visible

The most effective part of a budget case is the counter-calculation. It begins with effort that arises today and appears to nobody as cost, because it belongs to daily routine: the days that go into manually preparing figures every month, the corrections from faulty master data, the double entry between systems, the rounds of reconciliation because two reports show different numbers.

These items can be quantified without a study. Ask in three areas how many hours per month go into activities that would not be needed with a sound data situation. Extrapolate that to a year. In most mid-sized companies the figure is larger than those involved expect, and it is more solid than any savings forecast.

The second part are the costs of delayed decisions. If a contribution margin report is only available three weeks after month-end, management decides for three weeks on the basis of assumptions. What that costs cannot be quantified exactly, but it can be shown against concrete decisions from the past year.

Linking to business goals

A management board approves budgets for goals, not for systems. The question is therefore not what the ERP can do, but which of the agreed company goals are unreachable with the current state. Anyone planning to expand into two further countries needs a system base that carries that. Anyone wanting to improve delivery performance needs reliable stock data.

That link turns an IT initiative into a business initiative. It also changes the counterpart: instead of negotiating about investment sums, you talk about the preconditions for something that has already been decided.

It helps to connect the goals with dates. A rollout into a further country needs lead time. If market entry is planned for the year after next, the project start follows from that, and the discussion moves from whether to when.

Presenting the risk realistically

Honesty includes naming the risks of the project and not only those of doing nothing. Large technology programmes exceed budget or timeline in 65 to 80 percent of cases. Putting that figure on the table yourself does not look weak but prepared, and lets you explain in the next sentence what is being done about it.

Source: McKinsey: The end of ERP as we know it. Five ways AI is disrupting ERP, May 2026

The most effective safeguard is the clarification work before the start and steering that sits on your side. Both can be shown as their own budget line, and that is exactly what makes the case credible: you are not only asking for money for software, but for the measures that reduce the known risk.

Also name a stop condition. A case that describes at which point and against which criteria the initiative is reviewed takes much of the weight out of the decision. Management boards approve more easily when the decision is not final.

How the case should be structured

A sound case fits on two pages. It starts with the current situation in figures, names the goals for which the initiative is a precondition, and sets the cost over five years against the cost of staying. Then follow the risks with countermeasures, and at the end the decision path with dates.

What does not belong in it are feature lists, product comparisons and technical detail. They belong in an appendix or a separate document, because they obscure the core question.

One practical note at the end: calculate in ranges, not in point values. A range with named assumptions is more credible than an exact sum that everyone recognises as a negotiating position, and it spares you the awkward situation of having to defend the first number later.