Skip to content

Knowledge base

Dynamic-robust organization

For engineering leaders: strategy versus plan, matching management to the problem, where responsibility sits, and whether culture can be changed directly.

Written by

Updated:

Leadership 2

Our position. We hold, as a working hypothesis, that execution is getting cheap and the scarce part is agency. Someone has agency when competence, authority and information sit together, and organizations move from process-driven to decision-driven (LinkedIn post (opens in a new tab)). The hypothesis comes from hycle (opens in a new tab), the “hyper cycle”: an open, German-language sketch of an operating model for work once AI agents take over implementation, published under CC BY-SA 4.0 by six practitioners, Benedikt Stemmildt among them.

The Denkzettel collection (opens in a new tab), 29 one-page notes in German, supplies the vocabulary for that move: plans, principles, management modes, interfaces and temporary structures. Its note on symmetric interfaces returns part of the work to the people closest to the situation.

How does a strategy differ from a plan?

A plan describes the actions for a known path. It turns rules into an ordered sequence. A strategy describes the current state, the desired state, and the principles that bound an unknown path (Strategie und Plan, Denkzettel 2 (opens in a new tab)).

That difference changes what leaders should ask for. A plan can specify what people must do. A strategy specifies what the organization must achieve or avoid, then leaves a bounded space in which people can act on their own judgment. Shared principles connect local decisions to central intent without requiring constant approval (Strategie und Plan, Denkzettel 2 (opens in a new tab)).

How should management change with the problem?

The collection separates problems with a known solution from problems that require new ideas. The first can move through the line with documented rules. The second needs people who can generate options and a temporary structure around them (Chaos und Dynamik, Denkzettel 1 (opens in a new tab), Drei Wege, Probleme zu bearbeiten, Denkzettel 16 (opens in a new tab)).

This gives leadership two different jobs. When the answer is known, management can issue instructions and preserve a repeatable process. When the answer is unknown, leadership brings the problem into contact with people who can solve it. That work depends on trust and advice that people can reject without turning it into an order. Once a solution exists, control can take over its application (Steuerung und Führung, Denkzettel 5 (opens in a new tab)).

Projects move between these modes. Even an innovation project contains routine work, while a routine project can be interrupted by an unexpected problem. The right management mode follows the current mix instead of forcing every task into one process (Dualität eines Projektes, Denkzettel 14 (opens in a new tab)). This complements the redwork and bluework rhythm in Transformational Leadership, while focusing on the kind of problem rather than the meeting pattern.

Where should responsibility sit when uncertainty rises?

When uncertainty reaches the edge of an organization, repairing central control adds more machinery to a system that has already lost the necessary overview. A better move is to make the interface more symmetric by returning part of the work to the people closest to the situation. Mutual dependence gives both sides a reason to understand each other’s constraints and can produce self-organization without constant direction (Schnittstellen symmetrisieren, Denkzettel 8 (opens in a new tab)).

The boundary should move selectively. Routine work can stay centralized when that creates useful scale. Local teams need the parts of the work that require situational knowledge. Central leadership then watches the quality of the relationship and offers advice when the two sides cannot coordinate themselves (Schnittstellen symmetrisieren, Denkzettel 8 (opens in a new tab)). That principle connects to bounded contexts and vertical systems, where ownership follows the boundary that can make sound decisions.

What separates a team from a committee?

A committee chooses among options. A team creates options by combining the judgment and persistence of people who can work through an uncertain problem. Calling every functional group a team hides the difference and gives both forms of work the wrong expectations (Team und Gremium, Denkzettel 27 (opens in a new tab)).

Leadership can help a group switch modes when the work changes. Use a committee when the organization needs a documented decision. Create the conditions for a team when the organization needs ideas, learning, and persistence before anyone can decide. The distinction complements the Guild Decision Framework, which defines who holds decision authority after the options exist.

Can leaders change culture directly?

Visible behavior and underlying values are different. Rules can shape behavior when work is routine. When people must act without a rule, their decisions draw on values built through experience. The collection therefore treats culture as evidence of how the organization’s work is designed, not as a workshop output that leaders can install directly (Kultur, Denkzettel 6 (opens in a new tab)).

That changes the leadership question from “How do we declare the right values?” to “What does our work teach people to value?” If leaders want different decisions under pressure, they need to change the situations and consequences that produce those decisions. The visible behavior is the signal. The repeated experience underneath it is the thing to learn from.

  • Leadership: the section on building the environment in which teams decide on their own.
  • Moving away from Agile: McKinsey’s evidence that smaller pods with consolidated roles outperform agile-era teams.
  • Architecture Principles: bounded contexts, where ownership follows the boundary that can decide.
  • Organizational learning: how an organization learns the new mode instead of declaring it.
  • The Five Ideals: locality and psychological safety as conditions for local judgment.

Resources

Leadership and organization