Seventy years of impact management

Impact management sounds like a field from the last ten years. The question behind it is much older. What does our work change, and for whom? For seventy years, companies, governments and organisations have looked for an answer. Each time they found new words, new figures and new rules.

This is that history, with a source at every step. First we follow the timeline, from the 1950s to now. Then we follow the practices that meet in the field, like rivers in a delta. After that we look at where all those standards come from, and in which language they were conceived. At the end we pause on a tension that belongs to the field, setting norms, and on today’s headwind.

In 1953 the American economist Howard Bowen asked what a business owes to society. For a long time that remained a conversation about good intentions. If you wanted to know about results, you mostly looked at money.

Meanwhile, two practices grew that would matter later. In factories, managers learnt to measure and improve their work. The American statistician W. Edwards Deming took that approach to Japan from 1950. There, quality became a fixed cycle: plan, do, check, act. In 1954 Peter Drucker gave managing by goals a name: management by objectives.

The criticism came almost at once. In 1956 the researcher V. F. Ridgway showed that performance figures change the very behaviour they are meant to measure. People start working for the number, not for the goal behind it. So the warning came just two years after Drucker’s book.

The other practice was development co-operation. Rich countries gave more and more money to poor countries and wanted to know what it achieved. In 1961 the OECD set up a committee for this, the Development Assistance Committee (DAC). In 1969 a consultancy working for the American government designed the logical framework. Many organisations still use that table of goals, results, indicators and assumptions.

In the same year Sherry Arnstein published her ladder of citizen participation. Manipulation sits at the bottom, citizen control at the top. So one year brought two tools with very different views. The logframe looks at a project from the funder’s side. The ladder looks at it from the side of the people it is about.

At the start of 1970 a law in the United States required an environmental impact statement for large federal projects. You had to show in advance what a project would do to its surroundings. That idea later travelled to many other countries.

In 1972 the Club of Rome used a computer model to show that growth on a finite planet has limits. In the same years, companies started to publish a social report. They wrote down what they did for their staff and their surroundings. In France such a report became compulsory for large companies in 1977.

In 1987 a United Nations commission gave sustainability a definition that is still in use. Meeting the needs of today without leaving later generations short. In the same year ISO 9001 appeared, the standard for quality management. Now an outsider could certify that an organisation had its processes in order.

That idea from quality management still shapes the field. A management system, an audit and a certificate together form the basis of many later standards. Think of ISO 14001 for the environment, and of the labels for timber, coffee and clothing. The EFQM model of 1991 and the balanced scorecard of 1992 also grew out of this tradition. They taught organisations to measure more than profit alone.

Meanwhile, evaluation became a profession of its own, with professional bodies and its own norms. In 1991 the DAC set out the criteria for judging a programme. They are relevance, effectiveness, efficiency, impact and sustainability. Many funders still ask for them.

In the 1990s sustainability became something you could count. In 1997 the Global Reporting Initiative began in Boston. That work led in 2000 to the first guidelines for a sustainability report. Alongside the annual report came a second report, with a standard of its own.

In 1994 the first King Report on corporate governance appeared in South Africa. Later versions asked companies to bring financial and social information together. That made Johannesburg one of the first places where integrated reporting took shape.

In 2000 the UN set eight Millennium Development Goals for 2015. For the first time the world had a shared list of goals. You could now also put impact into money, with social return on investment. In 2001 the GHG Protocol appeared, the standard for counting greenhouse gas emissions.

In development co-operation a counter-movement grew. In the 1990s Rick Davies developed the Most Significant Change method in Bangladesh. In it, participants themselves say which change mattered most to them. In 2001 Canada’s International Development Research Centre published outcome mapping, which looks at how the behaviour of partners changes.

Medicine brought a very different approach: the randomised trial with a control group. From 2003 a lab at MIT took that method into the fight against poverty. And in 2005 donors and partner countries agreed in Paris to manage more for results.

After 2005 investors began to move. In 2006 the UN launched the Principles for Responsible Investment. A year later a new way of investing got a name: impact investing. You deliberately seek both a financial return and a social effect.

In 2009 the Global Impact Investing Network began, with IRIS as a shared set of indicators. ISO 26000 gave guidance on social responsibility in 2010. In 2011 the UN Guiding Principles on Business and Human Rights followed. In 2014 the European Union asked large companies for non-financial information for the first time.

Other countries took a different route. As early as 2009, Mauritius required companies to spend part of their profit on social causes. From 2014, India asked the same of large companies by law. There the point was not a report but money for projects.

In evaluation, too, other voices spoke up. In 2007 African evaluators met in Niamey under the banner Made in Africa. They asked for evaluation that starts from African values and knowledge. A year later the International Aid Transparency Initiative began. Since then, many donors have published in one open standard what they spend and on what.

In 2015 the world agreed on two things. In September came the seventeen Sustainable Development Goals, in December the Paris Agreement on climate. From then on, organisations could link their work to the same goals.

After that, the question was how to compare impact. From 2016 the Impact Management Project brought hundreds of organisations together around five dimensions of impact. In 2019 the Operating Principles for Impact Management for investors followed. At the same time Europe made its own rules for sustainable finance: the SFDR in 2019 and the EU Taxonomy in 2020.

Standard-setters also sought each other out. In 2020 five large organisations promised to work together, among them the Global Reporting Initiative and CDP. In 2021 the IFRS Foundation announced a new standard-setter in Glasgow: the International Sustainability Standards Board (ISSB). In 2022 the Climate Disclosure Standards Board and the Value Reporting Foundation merged into the IFRS Foundation. That brought the SASB Standards and the framework for integrated reporting under one roof.

Yet one language is still a long way off. In 2022 researchers at MIT compared the sustainability ratings of six large agencies. The agencies often gave the same company very different verdicts. In 2024 Europe and the ISSB laid their standards side by side. The result was a guide that shows, topic by topic, what matches. It did not become one standard.

Elsewhere, too, countries wrote their own rules. In 2021 India introduced the BRSR, a report on responsible and sustainable business for its largest listed companies. In the same year the countries of South-East Asia made their own taxonomy for sustainable finance.

Until then almost everything had been voluntary. That changed quickly. In 2022 the European Union adopted the CSRD, the directive on sustainability reporting. In 2023 the European reporting standards followed, the ESRS, as well as the first global standards of the ISSB.

In 2024 came the CSDDD, on due diligence in the value chain. For small businesses there was a voluntary standard, the VSME. At the end of that year China also published a basic standard for sustainability disclosure. In thirty years the alphabet soup had grown a great deal.

Impact management is not a field with a single beginning. It is a delta where many rivers meet. Development co-operation, evaluation, participation, quality management, government policy, reporting, investing, putting a price on value and environmental policy each have a source of their own. Human rights, labels, the social economy and thinking about prosperity flow into it too. Each framework got one river: the tradition its makers worked inSource: Narrate, Our own count.

Each river brought something different. Quality management brought the cycle of improvement and the audit. Evaluation brought the question of whether the difference really comes from your work. Reporting brought the annual report, materiality and the auditor. And development co-operation brought the logframe and the DAC criteria.

Smaller, less visible rivers flowed in too. In 1947 Kurt Lewin described how groups change: unfreezing, moving and freezing. Much of change management still builds on that. In 2011 John Kania and Mark Kramer wrote about collective impact. They looked not at one project but at all the parties in a region or sector that tackle one problem together.

The Netherlands has a tributary of its own: transition thinking. Around 2000, researchers such as Jan Rotmans, René Kemp and Frank Geels described how large systems change. Such change takes decades, not the length of a project. In 2001 the Dutch government took that thinking into its fourth National Environmental Policy Plan. In 2004 the Dutch Research Institute for Transitions opened in Rotterdam.

Each river also has words of its own. An evaluator says effect, an investor says return, an auditor says assurance. That explains part of the alphabet soup. That part is not a lack of co-ordination. It comes from where each word started.

One of the rivers tries to express value in money. It is an old wish. A figure in money fits into a budget and an annual report. It makes comparing easy.

In 1988 the New Zealand economist Marilyn Waring showed what falls out of view. In her book Counting for Nothing she showed that unpaid work does not count in gross domestic product. Women’s work above all disappears from the figures. What has no price seems to have no value.

Still, the pricing went on. In 1997 researchers around Robert Costanza put the value of nature at 16 to 54 trillion dollars a year, 33 trillion on average. In 2009 a report by Joseph Stiglitz, Amartya Sen and Jean-Paul Fitoussi appeared, at the request of the French government. They looked beyond gross domestic product. The Natural Capital Protocol followed in 2016 and Harvard’s impact-weighted accounts in 2019.

These methods make a lot visible. They show which costs a company passes on to people and nature. But the outcome depends heavily on the prices you choose. Two calculations for the same programme can lie far apart. And a figure in money often hides who made the choices.

One river is easy to overlook. It is about the people the work is for. In many measurements they are the source of the figures, but not their owner. They fill in questionnaires, but seldom read the report. Accountability goes up, to the funder. The data comes from below.

Resistance to that came early. In 1997 Robert Chambers asked in his book Whose Reality Counts? whose reality actually counts. He argued for methods in which people map their own situation. In 2001 Bill Cooke and Uma Kothari warned that participation, too, can become a form of power. Whoever leads the conversation decides what gets said.

In the years that followed, a movement around feedback grew. Organisations asked the people they work with for their verdict, much as a company asks its customers. Indigenous researchers went further. With the CARE principles of 2019 they asked for control over data about their own communities.

That raises an old question. Is someone a participant, a rights-holder or a data point? The answer decides who may tell the story. And it decides whether a measurement gives anything back to the people who made it possible.

When sustainability entered the annual report, the auditor came with it. The auditor had to say whether the information was right. There was already a standard for work beyond the financial statements, ISAE 3000 of 2003. In 2024 the international standard-setter for auditors also approved ISSA 5000, written for sustainability information.

There are two levels of assurance. With limited assurance, the auditor says nothing has come to their attention that suggests the information is wrong. With reasonable assurance, the auditor says the information is right in all material respects. Under the CSRD, limited assurance is enough.

In 1997 the British researcher Michael Power already spoke of an audit society. Organisations arrange their work so that it can be checked. That gives assurance, but it also has a cost. A data point that an auditor can test is not always what matters to a participant. And the large firms that give assurance also earn from advising on such reports.

So far we have followed the timeline and the rivers. Look at the map instead, and you see something else. For every framework in our lists, we looked up where it was made. That covers the alphabet soup, Fabrik’s reference data and a few names from this history: 284 in allSource: Narrate, Our own count.

For an organisation we count the city of its head office when the framework appeared. For a method we count where its makers worked, for a treaty the city where it was agreed. We place European rules in Brussels.

The picture is clear. Almost everything comes from Europe and North America. By the West we mean Europe, North America, Australia and New Zealand. The United States and the United Kingdom together supply almost half. The Netherlands ranks high, but that mostly says something about where we stand. We simply know the Dutch rules better.

A map is never neutral. Many world maps draw Europe and North America larger than they are. That is why we use the Winkel tripel projection. It keeps area, direction and distance all as close to true as it can. On that map you see how small the area is where almost everything comes from. Europe is so full that we show it enlarged beside the map.

Almost seven in ten frameworks first appeared in English. That also holds for many standards from the Netherlands, Switzerland or Sweden. If you want a standard to be used worldwide, you write it in the language of investors and international bodies.

That matters in practice. Words like accountability, outcome and materiality have no exact Dutch equivalent. A translation often only comes once a standard is finished. So if you do not speak English, it is harder to take part in a consultation on a new version. Knowledge that is not written down in English counts for less.

European rules are an exception. They appear in every language of the Union at the same time. Partly because of that, the share of English has fallen since 2010. The centre of gravity is moving from London and Washington to Brussels. But that is still the West.

Of the 284 standards, 17 come from outside the West. That is six per cent. Five of them are only here because we added them ourselves for this article. They were missing from the alphabet soup and from Fabrik’s reference data. We also added Kaupapa Māori evaluation from New Zealand, but it counts as Western here, because we count by country.

Look beyond the standards, and the map changes. For Fabrik, its makers also gathered sources from practice, in many languages. In them we found another 72 frameworks, laws and methodsSource: Narrate, Our own count. Almost half of them, 35, come from outside the West. Two in three first appeared in a language other than English.

Think of Buen Vivir in Ecuador’s constitution, Gross National Happiness in Bhutan and BRAC’s graduation model in Bangladesh. Or of the social economy laws in South Korea, Ecuador and Uruguay. With the King Reports, South Africa led the way on integrated reporting. India has made its own rules for sustainability disclosure since 2021, China since 2024 and Japan since 2025.

So outside the West, plenty of norms are being set. But those norms rarely become a global standard. A list of frameworks shows who is heard, not who has something to say. We will take that into account when we update our lists.

Impact management has a tension built in. Impact is about change for people and their surroundings, and that differs from place to place. A standard, by contrast, wants you to measure the same way everywhere. Without a norm you cannot compare anything. With a norm you risk measuring what fits, and missing what matters.

The tension is not new. After Ridgway in 1956, the British economist Charles Goodhart wrote something similar in 1975. A statistical regularity tends to collapse once you start to steer by it. The American researcher Donald Campbell wrote about it in 1979Source: Donald T. Campbell, Evaluation and Program Planning 2(1), Checked. The more a figure is used for decisions, the more it comes under pressure. The figure then starts to shape the behaviour it was meant to measure.

In 1995 the historian Theodore Porter showed that figures win trust above all where people do not know each other. A standard then replaces the judgement of those who know the situation. In 1998 Wendy Espeland and Mitchell Stevens described what happens when you put unlike things on one scale. They showed that something that does not fit the measure gets lost.

Setting norms is therefore also a question of power. Whoever writes the standard decides which change counts and how you prove it. As long as those writers mostly come from a few countries, the norm fits their practice best. Methods such as Most Significant Change and Made in Africa are also a response to that. They ask who may say what a change is worth.

Power has another side too: who pays for putting a standard into practice? Writing a standard costs less than using it well. If you have no money for training, data and checks, you can often only tick the norm off. There is no room left to learn from it. Then a framework becomes a form.

That does not mean you should work without standards. A shared language makes working together easier. It helps a funder, a board or a policymaker to compare. It does mean using a standard as a tool, without losing sight of your own account of your work.

Around 2022 the wind turned. In the United States, politicians in dozens of states put forward bills against sustainable investing. The American securities regulator adopted a rule on climate disclosure in 2024. A year later it stopped defending that rule in court.

Companies responded in two ways. Some said less about their goals. They feared criticism from both sides: that they did too little, or that they were playing politics. That silence got a name: green hushing. Others carried on, because customers, funders and staff kept asking for it.

In 2026 Europe, too, rolled back part of its rules. With the Omnibus package, the CSRD now only applies to the largest companies. They have more than 1,000 employees and more than €450 million in turnoverSource: Europees Parlement en Raad van de Europese Unie, Not yet checked. Many organisations that were getting ready no longer have to report.

The question has stayed. A funder, a customer or a board still wants to know what your work changes. What this history teaches is that rules come and go. Where they are made shifts too. If you understand for yourself what your work does, you do not need to wait for the next rule.

We do not know what comes next either. Perhaps another wave of rules is coming, perhaps not. That is why we would rather start from that one question than from the newest framework.