First Order And Second Order Change: Understanding The Difference

Understanding the difference between First- and Second-Order Change in organisations, and how to recognise when each is appropriate

One reason that so many organisational change initiatives fail is that the leaders and consultants involved don’t recognise the difference between ‘first-order’ and ‘second-order’ change.

First-order change works within an existing structure and view of the world. You could view it as tinkering with the system – doing more or less of something, making an existing process better or more accurate, and making incremental changes.

With first-order change, the ends of the system remain the same – it’s the means of producing those results that change. What you seek, what you avoid, the way you see the world, and your values remain the same.

Second-order change is often described as ‘transformational’, ‘revolutionary’, ‘radical’, ‘disruptive’, or ‘discontinuous’. It involves seeing the world in a different way, challenging assumptions, and working from a new and different worldview.

Inevitably it involves new ways of doing things, changing values and goals, and probably structural change in the organisation as well. This can be quite scary to most people, especially where changes are imposed from above or outside, and you don’t have any input to them.

First-order changes are much easier to make, because we are always tempted to look for a single cause for a problem. Thinking is hard, and we’ll do quite a lot to avoid it. With a first-order intervention, we tend to just look at the symptoms and their immediate cause, rather than considering the system as whole.

Sometimes first-order changes work, and the efficiency of the system improves. They are most likely to be successful where the problem has a single cause – for example, a slow broadband connection that means not everyone can get online when they need to. Speed it up, and the problem goes away.

Very often, there are reasons why a first-order change won’t work. One reason could be when the problem provides a ‘secondary gain’ for one or more of the people involved.

A secondary gain is a benefit that someone gets as a side-effect of having the problem. For example, workers might be instructed to hold shorter meetings, because management has identified the amount of unproductive time spent in meetings as a problem. But if the meetings are the only time you feel listened to and valued, the temptation will be to stay in the meeting for as long as you can.

In that situation, while the ‘shorter meetings’ rule might be obeyed for a while, but if the need for being listened to is not met anywhere else, meeting times will tend to creep up again because of that ‘secondary gain’.

First-order change can also fail when the problem occurs in a complex system. Very often, the attempt to solve the problem will itself cause problems down the line or elsewhere in the system.

For example, when New Labour was elected in the UK in 1997, one of its pledges was to get hospital waiting list times down. Targets were set, and waiting times came down. However, there were unintended consequences of setting these targets, as the British Medical Journal described:

What you can’t see and measure doesn’t exist. The target is met and taken as evidence of good performance, but its true impact is concealed. After targets were introduced for inpatient and outpatient waiting times, median waits increased, waiting time was shifted to diagnostics, and bed occupancy rose to levels associated with excessive risk of infection.

A further problem with first-order change is that any given way of seeing the world (sometimes called a ‘frame’ or ‘schema’) operating in an organization usually does not serve the interests of all the organisation’s members equally.

Any first-order change, because it operates within that way of seeing the world and (hopefully) makes it work better, at least implicitly endorses and reinforces it. If it isn’t working for some members, they may feel that the change has made them worse off, and push back against it (or at least be less than enthusiastic about implementing it).

This point, and many other aspects of change, is ably described by Dr Jean M Bartunek in her paper First-Order, Second-Order, and Third-Order Change and Organization Development Interventions: A Cognitive Approach (highly recommended and very readable, by the way).

So how can we decide if a first-order change won’t suffice and a higher-order change is needed?

If the problem keeps recurring and the fixes you try don’t ‘stick’
If the intervention solves the immediate problem but causes other problems elsewhere in the system, or after a time lag
If there are multiple contributing causes to the problem
If the problem carries a ‘secondary gain’ (ask “What are the benefits of this problem, and to whom?”)

It can help to map out the ‘problem space’, using the SCORE Model (Symptoms, Causes, Outcomes, Resources, and Effects). This short e-book describes how to use the method in individual or team coaching (SCORE is a model that comes from NLP, but you don’t need any training in NLP in order to get results with it).