How measuring shapes decisions

Why measure what your work changes? The answer you hear most is: to make better decisions. A board, a funder or a leadership team wants to know what works, and to steer by it. Yet many reports end up in a drawer. The decision is made somewhere else, for other reasons.

Does that mean measuring does nothing to decisions? No. It does a lot, but rarely in the way you expect. In this piece we look at how impact management shapes decisions. By that we also mean the wider practice of monitoring, evaluation, accountability and learning. Sometimes it helps. Sometimes it pushes decisions in a direction nobody chose on purpose.

The promise

The simple picture goes like this. You collect data, you analyse it and you choose the best option. Knowledge goes in, a decision comes out. Many handbooks on monitoring and evaluation start from this picture. So does the cycle of plan, do, check, act.

As early as 1947 the American economist Herbert Simon argued that it does not work like that. People have limited time, limited information and limited attention. They do not look for the best solution, but for one that is good enough. Later he gave this a name: bounded rationality. In 1978 he received the Nobel Prize in economics for his work on decision-making.

How decisions are really made

In 1959 the political scientist Charles Lindblom described how governments decide in practice. They take small steps from what is already there. They compare a few options, not all of them. He called it muddling through. That sounds careless, but it is often wise. Small steps can be undone.

In 1972 three researchers went further: Michael Cohen, James March and Johan Olsen. They compared decision-making in some organisations, such as universities, to a garbage can. Problems, solutions, people and moments land in it separately. A decision happens when they happen to meet. Sometimes a solution goes looking for a problem, rather than the other way round.

What does that mean for measuring? An evaluation rarely reaches exactly the right person at exactly the right moment. If you want knowledge to count, you need to know when decisions are made, and by whom.

Three ways knowledge finds its way in

The American researcher Carol Weiss spent decades studying how policymakers use research. Her work gave evaluators three forms they still use.

The first is direct use. An evaluation shows that a programme does not work, and the board stops it. This is what you usually expect. It happens less often than you think.

The second is enlightenment. Knowledge slowly changes how people think about a problem. Nobody can point to the moment, but after a few years a sector talks differently about poverty, loneliness or climate. Weiss found this is the most common form.

The third is symbolic use. A report serves to justify a decision that had already been made. Or to show that you work carefully. That is not always wrong, but you need to recognise it.

The evaluator Michael Quinn Patton added a fourth form: the use of the process itself. People who build a theory of change together, or make sense of results together, start to work differently. Even if the report never reaches anyone.

What you measure gets attention

Measuring also shapes decisions in a less visible way. What you measure gets attention. What you do not measure fades into the background. A board that sees the number of participants every month starts talking about participants. What changed for those participants comes up less often.

The American researcher Donald Campbell wrote in 1979 that a figure comes under pressure once it is used for decisionsSource: Donald T. Campbell, Evaluation and Program Planning 2(1), Checked. It starts to shape the behaviour it was meant to measure. An organisation judged on the number of support plans starts more support plans. A short, easy form of help then weighs as much as a long and difficult one.

Time shifts too. What shows quickly wins over what is slow. An outcome that only shows after five years fits badly in a yearly report. So a funder or a council leans towards what can be counted.

In 1997 the British researcher Michael Power called this the audit society. Organisations start making decisions that are easy to account for, rather than decisions that are good. A decision you can back with a figure feels safe. A decision based on experience or a conversation feels exposed. Even when it is better.

Who sits at the table

Measuring also changes who has a say. Whoever holds the data has a strong voice. That is often the funder, the analyst or head office. The participant or resident the work is about rarely sits at the table when the figures are discussed.

It can be different. In participatory methods, participants help interpret what the results mean. A funder can ask a grantee what they want to learn, instead of only setting indicators. Then the report becomes a conversation, not an inspection.

The same plays out in companies. With a materiality assessment, a company works out which topics matter: to the company itself, and to the people and the environment it affects. Whoever is asked to take part helps decide what the company steers by. Since the 2026 Omnibus package, the EU’s Corporate Sustainability Reporting Directive (CSRD) applies to fewer companiesSource: Europees Parlement en Raad van de Europese Unie, Not yet checked. The question of who has a say in what matters remains.

On time, or too late

A common complaint is that the evaluation comes too late. The new programme has already been designed, the grant already applied for. The report then confirms what the team already thought, or it disappears into the drawer.

Government is no different. A policy review sometimes only shows that a measure was not effective after the minister has announced a new one.

That is why some organisations choose forms that run alongside the work. Patton called this developmental evaluation: the evaluator works with the team while the work is still changing. You learn along the way, at the moment when there is still something to choose.

It does require knowing when decisions are made. When does the board meet to discuss what comes next? When does the council set the budget? When does the funder decide on a new grant? If you know that, you can bring the knowledge to that moment.

What helps

Finally, a few lessons you can use tomorrow.

Start from the decision, not the measurement. First ask which decisions will be made in the coming year, and what knowledge would help with them.

Ask what would change your mind. If no result would change the decision, you do not need to measure for it.

Measure little, but well. A few indicators that are really discussed say more than a dashboard nobody opens.

Combine figures with stories. A figure shows how much. A story shows why, and what it means for someone.

Keep track of what you did with a finding. A simple log from finding to decision shows whether knowledge carries through. Even when the answer is that you chose to do something else.

Leave room for judgement. A figure informs a decision. It does not make it. People decide, for reasons they can explain.

What remains

Measuring does not make decisions better by itself. It does change what you look at, who has a say and what feels safe. That influence is often bigger than the report’s own. Once you know that, you can set up your measuring so that it helps you choose well. So start with the question of which decision you want to make, and what you need to know for it.