Organisational Dysfunction of the Day

Budgets are bureaucracy

Context: The organisation has restructured into product teams. They have autonomy over their work, their roadmap, and how they collaborate. Then comes budget season. Each team is asked to forecast its needs twelve to eighteen months ahead, justify them against business cases, and compete with other teams for a finite pool of money. The forecasts get approved or revised by people who are not part of the teams. Mid-year, priorities shift. Some budgets get cut. Others get reallocated through a process that the teams do not control. The autonomy on paper meets the finance department in practice. Everyone knows which one wins. The team owns the product, but not the resources that decide what the product can actually become.

OST explains: Whoever controls the allocation of money controls the structure, regardless of what the org chart says. A DP2 design with DP1 budgeting is not a DP2 organisation. It is a DP1 organisation with self-managing teams operating in the spaces that money has decided in advance. DP2 requires that groups control the resources for their own work. That is what it means to own the whole task. What OST does not do is prescribe how this should look in a given organisation, because OST is a conceptual framework, not a recipe. A preset model handed down would itself be a DP1 move — exactly the kind of imposed design the theory argues against. The methods OST provides, Search Conference and Participative Design Workshop, exist to help an organisation work out its own answer, not to deliver one. Beyond Budgeting, developed in part by Norwegian practitioner Bjarte Bogsnes, is one practical attempt to separate the decisions about goals, resources, and evaluation that the annual budget collapses into one. It moves in the right direction. But the answer to how money should flow in a self-managing organisation is one that each organisation must design for itself, using its own people. The structure of money is the structure of the organisation. Both have to be designed together, by the people who will live with the result.

#OpenSystemsTheory #SocioTechnical #OrgDesign #leadership

Organisational Dysfunction of the Day

The agile PMO

Context: The teams are working well, or well enough. Sprints are running, retrospectives are happening, and there is some genuine ownership of the work. Then the question arrives from above: how do we scale this? The organisation has hundreds of developers across dozens of teams, and it needs them to be coordinated, aligned, and pulling in the same direction. A scaling framework is adopted; SAFe is the most common, but there are others. Release Trains are introduced. Programme Increment planning fills two days every quarter. A new layer of roles appears: Release Train Engineers, Solution Architects, Portfolio Managers. The teams are still doing agile. The organisation around them now looks remarkably like the bureaucracy agile was supposed to replace.

OST explains: Scaling frameworks like SAFe are not an extension of agile; they are a reassertion of DP1 over the partial DP2 that agile attempts. The coordination problem agile never solved at the team level becomes acute at scale, and in the absence of a structural answer, the bureaucracy provides one: add a layer above. The Release Train is a project in disguise, the PI planning session is a quarterly waterfall, and the portfolio layer is the PMO with new vocabulary. OST predicted this: when you introduce DP2 practices into a DP1 organisation without changing the underlying design principle, DP1 will fight back, and it usually wins, because it controls the resources, the budget, and the careers. SAFe is not a solution to the scaling problem; it is proof that the scaling problem was always an organisational design problem agile never had an answer to. The answer is not a bigger framework but a hierarchy of functions. Self-managing groups at every level, each owning its whole task and coordinating as peers. Agile needed fixing. Instead, it got scaled.

#OpenSystemsTheory #SocioTechnical #OrgDesign #agile

Organisational Dysfunction of the Day

The pair that runs everything

Context: Your team has two people who seem to understand both the work and each other. Everyone defers to them in standups. Their pull requests move faster. The real design decisions happen between those two, usually in a thread most of the team is not in. New people join and quickly learn to route questions through them. The team delivers, which is why nobody says anything. Management sees the output and calls it a high-performing team. Then one of the pair leaves. Then, within a year, the other. The team that looked healthy from the outside has no internal coordination structure of its own. It was borrowed from two people who are now gone.

OST explains: What you are seeing is Bion's pairing assumption operating at the team level. When a group has no real coordination structure of its own, it creates a surrogate: two people whose competence and rapport become the informal hub the rest of the team orbits around. It works, until it doesn't. Merrelyn Emery showed that pairing is not always destructive. In a well-functioning DP2 group, two or more people can come together in an animated exchange around a new idea, energising the whole room and pulling everyone into more creative work — a genuine prelude to the group firing on all cylinders. That is the regenerative form. What appears in most agile teams is the other one: a pair whose rapport becomes the de facto coordination mechanism, creating asymmetrical relations with the rest, parallel monologues rather than group conversation, a brittle stability that depends entirely on those two people being present. The structure determines which form pairing takes. In a DP2 group with genuine shared coordination, pairing is a spark. In a laissez-faire team without a coordination structure — which is most agile teams — it hardens into something that looks like high performance but is actually borrowed from two people the team was never asked to share accountability with. The pair did not fail the team. The structure failed the team, and the pair was the price.

#OpenSystemsTheory #SocioTechnical #OrgDesign #teamdynamics

Organisational Dysfunction of the Day

It's just a job

Context: The workplace is tolerable. Not great, but manageable. You have learned when to speak and when to stay quiet. You have learned not to care too much about decisions made above you. You protect your energy where you can. Then one evening, you notice you are short with your partner for no reason. You are tired in a way that sleep does not fix. You have stopped talking about work at home, not because things are fine, but because there is nothing useful to say. Friends ask how work is going. "Fine," you say. "It's just a job." You mean it as a boundary. It has quietly become something else.

OST explains: Merrelyn Emery described the spiralling maladaptive effects of bureaucracy, and the word spiral matters. It does not stabilise. A structure that generates contempt, humiliation, guilt, shame, and anger at work produces defence mechanisms: withdrawal, cliques, playing politics, and passing the buck. These make the workplace worse, which deepens the distress, which reinforces the defences. The spiral turns. And it does not stop at the door. Apathy and fatigue at work feed family and community disruption at home. Physical and mental health erode. Economic anxiety compounds it. This is where burnout ends up when it is not named or addressed — not recovery, but a permanent lowering of what life is expected to offer. Emery was clear that this is not a collection of individual failures. It is the predictable, structural output of DP1 applied at scale and over time. Organisations pay for it in productivity. Families pay for it in absence and conflict. Communities pay for it in disconnection. "It's just a job" sounds like pragmatism. It is the sound of the spiral completing.

#OpenSystemsTheory #SocioTechnical #OrgDesign #leadership

Organisational Dysfunction of the Day

Out of sight, out of sync

Context: The team is fully remote. Async communication is the norm, the tools are good, and the documentation is better than it ever was in the office. Standups happen on video, retros happen on Miro, and the team ships regularly. Once or twice a year the whole team gets together for a few days. The energy is noticeably different in person: things get resolved that had been stuck for weeks, new people finally feel like they belong, and everyone goes home re-energised. And then, slowly, things drift back. On paper, everything still works. In practice, something is harder to name. Decisions that should be simple take longer. Small misunderstandings accumulate. Someone does something that puzzles the rest, and nobody quite knows how to raise it. The team is coordinating. It is not quite cohering.

OST explains: DP2 is structurally possible in a remote setting. The requirements for self-management, that the group controls its own coordination and work design, do not depend on physical presence. But the social fabric that makes those requirements liveable does not. Solomon Asch identified four conditions that must be present for communication to actually work: openness, basic psychological similarity, a mutually shared field, and trust. Remote work weakens all four. The cues we use to sense whether someone is struggling, whether a decision landed well, whether a conflict is brewing, most of them are non-verbal, ambient, and invisible over video. A co-located DP2 group builds and repairs its social fabric continuously and mostly unconsciously. A remote DP2 group has to do that work deliberately. Team gatherings help, and people feel the difference when they happen. But they are injections of social capital into a structure that depletes it continuously. They are not a substitute for the daily, informal repair work that physical proximity enables. The result is a group that is formally self-managing but socially thin. Coordination holds. Cohesion frays. Remote does not break DP2. It makes the invisible work of maintaining it visible and expensive.

#OpenSystemsTheory #SocioTechnical #OrgDesign #distributedWork

Organisational Dysfunction of the Day

Them and us

Context: There is a clear divide in the organisation. It might be between IT and the business, between development and operations, between project teams competing for the same resources, between headquarters and the regions, or simply between management and everyone else. People talk about the other group with mild contempt or resigned frustration. Collaboration across the boundary is effortful, slow, and often fraught. Everyone acknowledges it is a problem. Workshops are held, values are restated, and leadership talks about being one team. The divide persists.

OST explains: This is one of the most reliable products of DP1, understood since Kurt Lewin's group dynamics research in the 1940s. Three characteristic features of communication in DP1 drive this divide: asymmetry, where communication flows up and down a status chain with no real conversation between equals; egocentrism, where individual accountability means people's interests are best served by looking out for themselves rather than sharing information; and "them and us" as the natural end state of both. When people are organised into separate functions with different managers, different goals, and different metrics, the in-group/out-group dynamic is not a cultural failure; it is the structural output. In DP2, teams are organised around whole tasks with shared goals, and all relations are peer negotiations. The divide disappears not because people are nicer, but because neither asymmetry nor egocentrism has anything to feed on.


#OpenSystemsTheory #SocioTechnical #OrgDesign #GroupDynamics

Organisational Dysfunction of the Day

Fixing the process

Context: Your team runs retrospectives. The board is visible. The ceremonies happen on time. You adopted lean thinking years ago, worked on flow, reduced batch sizes, and improved feedback loops. For a while, it helped. Then the gains plateaued. The same patterns returned, now with better labels. Scrum is in place, but the standups feel like status reports. The process is being followed, and the problems are still there. Something underneath the process is not changing.

OST explains: J. Robins & Sons, a hundred-year-old Australian fashion shoe manufacturer, tried the manufacturing-improvement playbook and ran into the same wall. They were heading the way every other Australian shoe maker had gone, towards offshoring or extinction. So they did something different: they used Participative Design Workshops to legally change the design principle from DP1 to DP2, the self-managing-group structure. Over five years, total stock dropped 50%, lead time went from 15 days to 2 hours, customer returns dropped 45%, downtime dropped 65%, pairs produced per person rose 30%, and absenteeism fell from 4% to 1.5%. They became the sole remaining large footwear manufacturer in Australia, competing with India and China without shifting jobs offshore. The lean and scrum tradition was built on real insights about flow and teamwork, but it stopped at process. OST goes one step further and changes the structure that the process runs on top of. Process improvements without structural change can take you a long way, until they cannot. J. Robins shows what is on the other side.

#OpenSystemsTheory #SocioTechnical #OrgDesign #lean

Organisational Dysfunction of the Day

Change agents of the status quo

Context: The organisation is struggling with its agile transformation. Teams are frustrated, delivery is slow, and the gap between the agile principles on the wall and the daily reality is hard to ignore. Leadership responds by bringing in agile coaches. Sometimes it is an external consultant. Sometimes it is a team lead who rebrands as a servant leader or agile enabler, sensing that the old role needs a new name to survive. The coaches run workshops, facilitate retros, introduce new ceremonies, and work hard to make things better. The teams appreciate the attention. Things improve slightly at the edges. The fundamental structure does not change. A year later, the same problems are present in slightly different forms. New coaches are brought in.

OST explains: Agile coaches are hired by management, which means the coaches who survive in the role are those comfortable operating within DP1. Those who challenge the structure too directly lose the engagement. Over time, the role selects for people skilled at making DP1 more bearable, not at replacing it. The coach becomes a buffer: absorbing frustration from the teams, translating it into something leadership can accept, and returning something that resembles a response. The energy that might have driven structural change is consumed in the process. For team leads and managers who sense that DP2 threatens their position, coaching offers a way to remain relevant, embracing the language of empowerment and self-management while preserving the hierarchy that sustains their career. When they cannot do it themselves, they hire consultants who can do it for them. The result in both cases is the same: a DP1 organisation with a DP2 vocabulary, which is laissez-faire. The coaches are not the villains here. They are the rational product of a system that needs the appearance of change more than it needs change itself. As Gerald Weinberg observed, organisations are perfectly designed to get the results they get.

#OpenSystemsTheory #SocioTechnical #OrgDesign #agile

Organisational Dysfunction of the Day

The AI we cannot talk about

Context: The organisation has adopted AI. There are productivity metrics, usage dashboards, and a strategy document that references "responsible AI." The head of sustainability has a slide about carbon neutrality by 2035. At a recent industry panel on AI, none of the participants would answer questions about the environmental cost of their deployments when the topic came up from the audience. Not because they did not know. Because their agreements with technology vendors included non-disclosure clauses that prevented them from saying. The audience noticed. Nobody said anything about that either.

OST explains: Data centre electricity demand is growing faster than any other sector. AI-focused facilities are the primary driver. The scale is documented, the trends are public, and in many organisations the internal numbers are known. The NDA is not a knowledge problem. It is a structural choice to prevent knowledge from circulating in the field. Active adaptation requires organisations to act on the field they operate in, not just absorb signals from it. At its fullest, that means contributing to conditions under which the whole system, including the broader environment, can develop sustainably. Fred Emery identified shared values as the only viable coordination mechanism for a field in motion. An industry that signs away its ability to speak about the environmental cost of its own technology has not just failed to share values. It has actively prevented the field from forming them. The oil industry spent decades funding doubt about climate science while its own engineers documented the damage. The pattern is the same: internal knowledge, external silence, and a field that cannot self-correct because the people with the relevant information have contracted out of saying it. This is not a communication problem. It is a field that has chosen not to adapt, at the cost of everyone in it.

#OpenSystemsTheory #SocioTechnical #OrgDesign #ai #climate

Organisational Dysfunction of the Day

The pilot trap

Context: The organisation wants to test a new way of working before committing to it fully. A pilot team is selected. They get more autonomy, a flatter structure, and direct access to the people they serve. The experiment is watched closely. Leadership visits. Consultants evaluate. Progress is tracked and reported. The team delivers well. They are energised and proud of what they have built together. Then the question of scaling comes up. Resistance appears, not from people far away, but from those closest to the experiment. The managers whose teams were not selected. The HR department asked to create exceptions for the pilot. The architects whose standards the pilot bypassed. The steering committee that was promised a clear verdict. The experiment ends, or quietly returns to normal. The organisation concludes that the approach works in theory but is not right for them.

OST explains: Merrelyn Emery identified this dynamic as paradoxical inhibition, drawing on Pavlov's research on conditioning: the people and areas closest to structural change feel most threatened by it and develop the strongest resistance, while those at a safer distance are more likely to adopt it. Norway lived this in the 1960s. The Industrial Democracy Programme experiments worked. The experiments did not spread. Sweden picked up the ideas and ran with them instead. Treating change as a pilot produces exactly this effect. Those around the pilot are confronted daily with a visible alternative to their own way of working. That is not a neutral situation. It is a structural provocation. Partial DP2 in a DP1 organisation is inherently unstable, not because the experiment fails on its own terms, but because the surrounding system has every structural reason to absorb or eliminate it. This is not cynicism or politics. It is the immune system of DP1 doing exactly what it was designed to do. The only way to avoid the pilot trap is not to pilot. You cannot sneak DP2 past DP1 one team at a time. The fence will hold.

#OpenSystemsTheory #SocioTechnical #OrgDesign #changemanagement