Skip to main content

Achieving Decentralized Control

D1: The Second Perishability Principle: Decentralize control for problems and opportunities that age poorly.​

Certain problems and opportunities are highly perishable, requiring swift action—like putting out a fire before it spreads. On the other hand, some issues, like a bad stock investment, have self-limiting impacts.

To handle problems that worsen over time or seize fleeting opportunities, decentralize control. This means empowering individuals with the authority and resources to act quickly. Centralized planning can set tasks, but decentralized control ensures flexibility, enabling immediate responses to minor issues without delays.

D2: The Scale Principle: Centralize control for problems that are infrequent, large, or have significant economies of scale.​

Some problems require large-scale responses—big fires need fire trucks, not garden hoses. Centralized control is ideal for infrequent but significant demands, especially when economies of scale are at play. In contrast, frequent, small, and low-variability issues are better handled by decentralized resources (D1: The Second Perishability Principle).

Centralizing resources can delay response due to their size and complexity, so prioritizing these resources is crucial to ensure timely action. In product development, instead of placing an expert on every project, centralizing their expertise can be more efficient.

D3. The Principle of Layered Control: Adapt control based on emerging information.​

When it's unclear whether to centralize or decentralize control, two approaches can help: triage and escalation.

Triage, as used in hospitals, prioritizes resources by addressing solvable, potentially critical issues and avoiding those that are unsolvable or less urgent. This method is effective when enough information is available at the outset to make informed decisions.

Alternatively, when it's uncertain which task will take the most time, a round-robin scheduling approach discussed in F19: The Round-Robin Principle ensures fair resource distribution. Each task is given a time slice, and unfinished tasks return to the queue, preventing long tasks from blocking shorter ones.

In systems with layered priorities, escalation ensures low-priority tasks don't get neglected. Problems are escalated after waiting too long, striking a balance between keeping important issues from being overlooked and avoiding overloading centralized resources.

D4. The Opportunistic Principle: Adjust the plan for unplanned obstacles and opportunities.​

In line with Principle FF5: The Moving Target Principle: Know When to Pursue a Dynamic Goal, static goals call for conformance, but dynamic goals require continuous adaptation. Plans should not enforce rigid conformance; instead, they should be used to maintain alignment and synchronize complex activities, such as coordinating multiple tasks for a product launch.

Adaptive plans allow teams to seize opportunities, adding significant value in product development. For instance, if a feature turns out to be far more costly and less valuable to customers than expected, a flexible process will recognize this and adjust. Similarly, if an easier-to-implement feature is discovered to provide greater customer value, adaptive plans make it simple to pivot, something much harder in rigid, conformance-focused processes.

D5. The Principle of Virtual Centralization: Quickly reorganize decentralized resources to create centralized power.​

It's often inefficient to keep centralized resources idle for rare large demands. Instead, these resources can be used for routine tasks and mobilized when needed to tackle big problems, allowing organizations to benefit from both centralized and decentralized resources, as discussed in F28: The Principle of Preplanned Flexibility: For Fast Responses, Preplan and Invest in Flexibility.

For example, cities can train civilians in firefighting and collaborate with nearby fire departments for mutual aid, providing centralized response capacity without maintaining a large permanent fire department.

In product development, some companies use "tiger teams"—highly skilled, experienced employees who handle critical crises but perform their regular duties during normal times. This strategy enables the company to mobilize expert resources quickly in emergencies.

D6: The Inefficiency Principle: The inefficiency of decentralization can cost less than the value of faster response time.​

D7: The Principle of Alignment: More value is created with overall alignment than local excellence.​

While one barbarian might defeat a Roman soldier, 1,000 Roman soldiers would always overcome 1,000 barbarians due to superior coordination and alignment of their forces.

In war, the principle of mass refers to concentrating effort and resources on a decisive point to achieve maximum impact. Economy of force means allocating minimal resources to secondary tasks, ensuring most resources are focused on the main objective. Together, these principles demonstrate the power of alignment.

In product development, small gains across many features don't impress, but a significant advantage in a key area can sway customer preference. Aligning resources to focus on a single area can create a competitive advantage, even if it means sacrificing excellence in other areas.

D8: The Principle of Mission: Specify the end state, its purpose, and the minimal possible constraints.​

A mission order does not tell a unit what to do, by when, with what resources and by which plan. It states the intent: the end state the operation is meant to create, always a goal larger than the operation itself. It conveys the "why," not the "what" or the "how." The Marines write it as the Commander's Intent and consider an order incomplete without it, because when the intent is clear, whoever executes can choose the right course of action, above all when an obstacle or an opportunity appears that the plan did not foresee.

Intent is a robust and compact way to give direction. A project's mission works the same way: the most important part is the "why," and the fewer constraints it adds on how, the more room it leaves to adapt.

D9. The Principle of Boundaries: Establish clear roles and boundaries.​

Decentralized control works only when everyone knows what is theirs to decide and do. Unclear boundaries fail in two ways. Where roles overlap, two people do the same work or wait for each other, and every simple decision collects approvals. Where a gap lies between roles, the work falls between them and nobody does it.

Clear roles make communication cheaper, because each person knows whom a question belongs to without a meeting to find out. Defining roles is not enough on its own: look for the gaps between them too, since work there goes undone without anyone deciding to drop it.

D10. The Main Effort Principle: Designate a main effort and subordinate other activities.​

In product development, it's essential to focus on the few key features that drive success. While customers may request hundreds of features, only a handful—typically four to six—truly influence their decision. These are the “main effort” of the project, where the team should concentrate the most resources. Other features should take a back seat, allowing the team to excel in the areas that really matter. By prioritizing these core features, you maximize the impact and effectiveness of the product, much like focusing on a key point in a mission.

D11. The Principle of Dynamic Alignment: The main effort may shift quickly when conditions change.​

In product development, priorities can change as new information arises. Initially, one feature or aspect of the project may be the main focus, with all resources aligned to support it. However, as the project progresses, new insights may reveal better opportunities or shifts in market needs. When this happens, it's essential to be flexible and quickly redirect focus to areas that offer the greatest value. Adapting swiftly ensures that resources are always aligned with the most impactful objectives, optimizing the project's success.

D12. The Second Agility Principle: Develop the Ability to Quickly Shift Focus.​

It's not enough to decide to change direction quickly—you must be able to execute that change rapidly. In product development, you can pivot faster by:

  • Having a small, skilled team instead of a large, slower one.
  • Focusing on a streamlined feature set rather than managing unnecessary features.
  • Maintaining some reserve capacity, as overworked teams struggle to adapt quickly.

Additionally, designing flexible product architecture helps absorb changes. For example, segregating uncertainty into specific parts of the design allows for adjustments without affecting the entire system. This enables swift changes without disrupting overall progress.

This ties directly into Principle FF10: The First Agility Principle, which emphasizes that you don't need long-term planning when you can quickly adjust course. A short "turning radius" allows teams to adapt to changing conditions and maintain control, even at high speeds.

D13: The Principle of Peer-Level Coordination: Tactical coordination should be local.​

D14: The Principle of Flexible Plans: Use simple modular plans.​

Some things are planned in detail before the action starts, such as how units will communicate and the order supplies are loaded so they come off in the order needed. But plans are expected to change, so they are kept simple, modular and flexible. Simplicity makes adaptation fast; modules can be recombined in many ways; and flexibility comes from preplanning branches, points where a different path can be taken depending on conditions, and sequels, follow-on actions prepared in advance. Such plans keep the effort aligned when conditions change.

D15: The Principle of Tactical Reserves: Decentralize a portion of reserves.​

If response time did not matter, reserves would be consolidated, since a single central pool covers a large area with less total capacity. In maneuver warfare, reserves are held at every level instead, so each level can deal at once with a certain range of contingency: one of three companies held back can double another's strength at a critical moment, and reserves at each higher level bring ever larger forces, with longer delays. Force arrives at the right time and place instead of being committed on guesses before the battle; one-third to one-half of a force may be held this way.

In product development, the equivalent is capacity margin pre-positioned at several levels of the organization, as people or budget, rather than a single program-level reserve under the project manager, and safety margin in key processes. It lets each group absorb variation locally instead of escalating it (D5: The Principle of Virtual Centralization). A computer's memory hierarchy is layered for the same reason: a small, fast, expensive cache, larger and cheaper DRAM, and the cheapest and slowest disk together perform close to the fastest layer at close to the cost of the cheapest.

D16: The Principle of Early Contact: Make early and meaningful contact with the problem.​

The Marines make early contact with opposing forces to reduce uncertainty about them, and that contact is substantial action, not light observation: it resolves uncertainty and keeps more from arising.

In product development, the opposing forces are the market and technical risks a project must overcome. Assessing them is not enough; a project must spend meaningful effort to learn how severe they are and whether they can be overcome, and there is no substitute for a quick proof of concept or early market feedback. Aim that early effort at the zones of highest technical or market risk. A systematic top-down design of the whole system looks logical but rarely resolves risk fast enough.

D17: The Principle of Decentralized Information: For decentralized decisions, disseminate key information widely.​

Decentralizing control means decentralizing both the authority to decide and the information needed to decide well. The Marine Corps' minimum standard is to understand the intent of commanders two levels up. Companies that hold important information closely undermine the people they expect to act for the mission: where the decision is made, its information must be available.

That does not mean every piece of information should live in isolated local pools. The Boeing 777's cost-versus-weight decision rule was calculated at the level of the whole system and then handed to each engineer, which let thousands of small decisions be made quickly and correctly (E13: The First Decision Rule Principle).

D18: The Frequency Response Principle: We can’t respond faster than our frequency response.​

D19: The Quality of Service Principle: When response time is important, measure response time.​

D20: The Second Market Principle: Use internal and external markets to decentralize control.​

D21: The Principle of Regenerative Initiative: Cultivating initiative enables us to use initiative.​

D22: The Principle of Face-to-Face Communication: Exploit the speed and bandwidth of face-to-face communications.​

Scope: human-to-human communication. What people decide this way reaches agents only once written; see The Corollary of Contextual Readiness.

D23: The Trust Principle: Trust is built through experience.​