Last week, in Are you the rarest of people: a system builder?, I described the person who can see how a system’s parts fit together and help people change it without breaking what they still need. I also asked whether we create the conditions for such a person to help us. This week I want to turn the question around: how do we become that person?

Thank you to everyone who wrote privately with stories of systems that worked and systems that went spectacularly wrong. It is remarkable what curiosity, goodwill and thoughtful analysis can achieve, even when an organisation is blowing more smoke than a 1980s Mitsubishi Sigma.

The route to becoming a system builder is indirect. You do not get there by collecting a systems vocabulary and drawing a more impressive diagram. You start by noticing what the system does to the people inside it, and what those people do in response.

For me, that means beginning with social psychology. In The Lucifer Effect, Philip Zimbardo argued that a “bad barrel” can shape behaviour more powerfully than a few “bad apples” explain it. Roles, permission, authority, group expectations and the apparent costs of speaking up can make harmful conduct seem ordinary. If you want a different outcome, you have to ask who made the barrel and how it works.

A necessary qualification: Zimbardo’s famous Stanford prison experiment is not secure proof of that broad claim. An archival investigation found serious problems with its methods, including instructions given to the guards. The useful lesson is a question, not a licence to declare that character never matters. What behaviour does this particular setting reward, discourage or make easier to ignore? Zimbardo himself also emphasised that people can resist a bad situation, and that institutions can help them do so.

Angela Duckworth takes that question in a more hopeful direction in Situated. Where Grit concentrated on the individual’s passion and perseverance, Situated asks how space, peers, mentors and culture can bring those qualities out or suppress them. She encourages readers to improve the conditions around their own efforts. A system builder takes a further step: whose efforts does our organisation support, and whose does it make unnecessarily difficult?

That step leads us from social psychology to strategic culture. In his Joint Special Operations University monograph Strategic Culture, Russ Howard examines how shared beliefs, experience and assumptions help shape what a group regards as an appropriate response to a threat. He is writing about states and security groups, not a neighbourhood business. My extension of the idea is that any cohesive organisation develops expectations about who it is, what it is for, which risks matter and what its people are allowed to do.

You can learn a great deal by asking what a group praises, excuses and repeats. Does a manager celebrate someone who prevents a mistake, or only someone who rescues the organisation after the mistake? Does the person who knows the most about a process get heard, or does the person with the highest title? The answers tell you why an apparently sensible reform may be welcomed, quietly worked around or treated as an attack on the group’s identity.

Once you see those habits, you have to discuss them without sounding like an agent of chaos. Marilee Adams’s Change Your Questions, Change Your Life offers a useful discipline. Her Learner mindset asks what is happening, what we might be missing and what else is possible. Her Judger mindset jumps towards blame and certainty. We can all do both. The first skill is noticing which mindset we have brought into the room.

“Why on earth do you still do it this way?” may be a reasonable thought, but it is rarely a useful opening question. “What does this step protect us from?” gives a maintainer room to explain. “What happens to the work after it leaves your desk?” invites a user to show you the handover that the process map missed. If a gatekeeper responds defensively, that, too, is information; it may tell you what they fear losing. Curiosity does not mean agreeing with every answer. It gives you a better basis on which to challenge one.

Only then would I reach for Russell Ackoff again. You might be getting tired of my references to a brilliant thinker from the 1970s and 1980s, but Ackoff is like broccoli: a bit too much at the beginning, then you develop a taste for it.

In The Art of Problem Solving, his puzzles show how an assumption we impose can block a solution. A practical problem demands choices and judgment. A mess consists of connected problems that change each other. Those distinctions are useful because a system often presents a neat little puzzle precisely when we are least able to afford a neat little answer. “How do we make staff answer enquiries faster?” might hide a mess involving unclear promises, overloaded staff, poor handovers and customers who must ask the same question twice.

There are genuine puzzles in the world. I simply want us to become suspicious when a persistent human difficulty can be made to look like one. Before solving the visible problem, stand back and ask what its apparent solution will do to the rest of the system. That was the heart of last week’s argument. It is also the moment when a curious observer begins to work like a builder.

To see how far this practice can develop, I look to high reliability organisations. Karl Weick and Kathleen Sutcliffe’s Managing the Unexpected describes organisations that must keep functioning when mistakes can have grave consequences. Their habits include looking for small signs of failure, resisting simplistic explanations, paying close attention to operations, preparing to recover, and letting relevant expertise matter more than rank.

The point is not that a small business should behave like an aircraft carrier. It is that “nothing has gone wrong yet” is weak evidence of safety, and a near miss can teach us more than a reassuring report. High reliability organisations do simplify tasks through useful routines. They resist simplifying the explanation of why the work succeeds or fails. They know an unexpected problem will arrive eventually, and they practise noticing, responding and learning before a small fault becomes a cascade.

All that learning still leaves one practical challenge. As a system builder in training, you may be the most junior person in the room, or simply the person least entitled to alter the system on paper. How do you tell someone more powerful than you that their favoured arrangement is causing harm?

Some people recommend Machiavelli when the subject turns to influence. I would start with Baldassare Castiglione’s The Book of the Courtier. Castiglione wrote about serving powerful Renaissance rulers. His ideal adviser earns trust through competence, restraint and good conduct, then uses that trust to tell the ruler a difficult truth without either flattering them or trying to humiliate them. I would not transplant a sixteenth-century court into a modern workplace, but the ethical challenge is familiar: influence should make candour possible.

You probably will not plan your next process improvement while wearing a velvet cape in a spectacular villa. You can still become the person others trust to notice what is wrong, explain it fairly, and help them act without losing face. That trust is part of the system you are trying to build.

Taken together, this is a demanding education. Notice how roles and incentives shape behaviour; examine the space, relationships and culture that help people flourish; learn what a group believes about itself; ask Learner questions before offering Judger answers; distinguish a puzzle from a problem within a mess; watch for weak signals and listen to the people closest to the work; and earn enough trust to offer honest, tactful advice. None of those skills works as well on its own as it does in combination.

You can begin without waiting for a grand appointment. Choose one small system near you: the way an enquiry reaches a colleague, the way a new person learns the job, or the way a recurring error gets reported. Ask the people involved what the process helps them do, where it makes their work harder, and what must not be lost. Change one thing together. See what actually happens, including somewhere else in the system, and adjust.

Do this well with the little stuff and people may eventually invite you into much larger systems. More importantly, the people using your first small improvement will already have a better day. System builders need not be as rare as they seem if we give more people the chance to learn by building.

Thank you for reading to the end. If you got value from this, please share it with your network and sign up below for updates. It really helps, and it makes me smile.

Sources and further reading