Group Dynamics: No Authority, Only Approvals

The guest comes to the topic from practice: more than fifteen years in IT, starting in technical support, then heading that function, running project teams in custom software development and working as a portfolio manager. In a foreign, more product-oriented company she ran into the problem that she could not get an idea across to top management, even though everything was laid out in metrics and indicators — so she went to study for the master's degree she is now finishing. The hardest term she met there is identity: a set of habits, emotions, values, talents and current goals belonging both to a person and to the system that person is part of. Her example: a team does a lot of technical work, but someone needs to see the result on the outside, in a product feature, and satisfaction never arrives even though the hard skills fit. Identity is fluid and changes over time; a company has one too — strategy, mission, values, how open it is to the outside.

Alexander admits he has never used the word identity: for him it is culture and norms a person either fits or does not, the thing a culture fit interview probes. He recalls a book subtitled "what an anthropologist can teach a top manager": instead of applying stencils such as Amazon's customer obsession principle, you first immerse yourself in the system — listen to the language people speak, their rituals, what earns a reward and what earns a punishment — and only then rise to a helicopter view and offer an interpretation. Otherwise you get the familiar story where a consultancy imports values from outside, everyone agrees, a significant sum is paid and nothing changes; or where principles hang on the wall but an employee standing with their back to it cannot list them. Alexander asks whether anyone will want to listen at all, and answers himself: usually in a crisis, when the earlier tools stop solving the problem. Nastya adds that the psychoanalytic method surfaces what goes unsaid — for instance a task everybody discusses and nobody does. The method, she says, is also gentler: people hear that what is happening is not directly their fault, that there are circumstances such as the economic context they cannot influence.

The slide shows the individual, the context and the system: the role is born where the three cross, and it is not reducible to a job description. Context can change even a chief executive's role, moving them to a different level and style of management. The fourth element is passion, which directly sets how much responsibility a person actually takes on: when the overlap between personal identity and the system is thin, the responsibility turns out just as shaky. Alexander asks whether this is engagement by another name: engineering models, he says, measure satisfaction rather than engagement, and SPACE opens with satisfaction. His counter-example is the work-to-rule strike: in at nine, out at six, exactly an hour for lunch, the work formally done. The person turns into a biorobot, while the company sees burnt-out staff moving like a car on flat tyres — it rolls, but will not get far.

Nastya insists that passion is wider than engagement: people work with shining eyes, get feedback and development, and still burn out or leave. What is missing is the belief that you are not alone, that the company is behind you and you are behind the company, plus the sense of belonging usually called ownership. Alexander adds the mechanics: long-term motivation and share grants give a top manager the feeling that the company's success is their own. A separate thread is Amazon's principles as a shared language: when making a decision a person names the principle it follows, and the others check the fit and accept it. The host points to his own two-hour episode with Dima, who spent five years at Amazon and four at Microsoft: the principles genuinely work, while Microsoft has no common set and differs a lot depending on where you sit. A culture uniform across a whole large organisation is rare; each wing has its own cultural code.

Nastya draws responsibility as a boundary between the individual and the system, and a dotted one at that: when the C-level is not aligned and the board is storming, responsibility further down goes full of holes as well, and even very conscientious people stop understanding what to do and when. Her case: after a merger the leadership changed completely, the staff stayed, production could not be stopped — and defects grew several times over; consultants rewrote the processes on paper, which did not help, because people still did not understand what they were meant to do. This is where BART comes in — a model from the Tavistock Institute, one of the first to study dynamics inside and between groups; later, by the guest's account, it was extended into CBART. Alexander jokes that this is not the Bart drawn in the corner of the slide. The four parameters are the group's boundary, its authority, roles and the primary task, and it is the task to bring the group back to first.

A group can usually name its tasks but rarely its authority; hence the mantra that it is easier to ask forgiveness than permission, and the self-regulation Alexander describes: while you carry the work, your zone of responsibility and credit of trust grow, and when you keep making mistakes they shrink. Nastya's case: two leaders of a 150-person unit spent a year and a half agreeing everything with each other, down to printer paper, because neither boundaries nor authority had been set. Alexander answers with an analogy from Raft, where nodes that cannot see the leader run endless election rounds and paralyse the group. The mirror image is top management that never uses its formal authority because results would have to be answered for, leaving hard tasks hanging in the middle for years. They then recall Tuckman's model — forming, storming and the adjourning that Nastya compares to grief work — and The Mythical Man-Month with its point about a late project. BART, she says, helps move through the early stages faster, and she signs off by recommending a group relations conference: three or four days with strangers, whose only brief is to study your own group dynamics.