Understanding optima: Why organizations get stuck between local optima - and fail to see the global optimum

At almost every board meeting, the sentence "We have optimized" is uttered at some point This usually means that projects have been prioritized, budgets distributed, programs coordinated and compromises made. The result feels structured, plausible, comprehensible - and yet a residual doubt often remains. Too many initiatives, too little impact. Too much effort, too little result.

The reason for this is rarely a lack of competence. The reason lies deeper - and is mathematical: Organizations systematically move between local optima, while the global optimum remains invisible in the decision space.

This article explains what optima really mean, why classic management methods inevitably end in local maxima - and why only the calculation of the entire decision space enables access to the global optimum.

Optimum and optima: a brief conceptual framework

An optimum is the best solution within a defined framework. The plural optima describes several solutions, each of which is optimal - but often only under local conditions.

In practice, this difference is crucial:

• Local optima are solutions that do not appear to be further improvable within a limited field of vision.
• The global optimum is the best solution in the entire decision-making space - regardless of how many alternatives exist.

The problem with modern organizations is not that they do not achieve optima. The problem is that they do not know which optima they have achieved - and which better solutions lie outside their field of vision.

Why management almost inevitably ends in local optima

People decide sequentially. Committees decide iteratively. Organizations decide politically, historically and resource-driven. All of this is rational - but it creates a structural effect:

Decisions are improved step by step, not calculated across spaces.

In optimization theory, this is a classic pattern: you move along a "hill", improve the result step by step - and finally stop at a local maximum. Not because there is no better solution, but because the way there is not visible.

This is how it manifests itself in organizations:

• Existing projects are defended ("sunk costs").
• New projects are discussed additively, not combinatorially.
• Resource bottlenecks are solved locally, not optimized globally.
• Compromises replace systematic consideration.

The result is a stable, plausible, but not maximized portfolio. A local optimum.

The real bottleneck: the invisible decision space

As soon as more than a few projects are available for decision at the same time, the decision space explodes. This space is not linear, but exponential.

With N projects, there are not N decision options, but:

2ⁿ possible project combinations

Even with 10 projects, that's 1,024 combinations. With 20 projects, over a million. With 50 projects, more than 1,125 quadrillion.

This space is no longer manageable for people, Excel and traditional prioritization logic. Management only sees a tiny section - typically a few dozen variants discussed - and considers this section to be the relevant space.

This is precisely where the error in thinking arises: Local optima do not result from poor decisions, but from spatial blindness.

Why discussion is no substitute for optimization

Many organizations react to increasing complexity with more coordination: more meetings, more templates, more committees. This increases transparency and legitimacy - but not automatically the quality of the decision.

Discussions take place within the visible space. They compare variants that are already known. They can refine local optima, but they cannot discover the global optimum if it lies outside the area under consideration.

Or to put it another way:

What is not calculated is guessed.

With few options, guessing is not critical. With 50 projects, guessing is expensive.

The size comparison: Why our intuition fails

People are not made to intuitively grasp exponential spaces. Large numbers quickly lose their meaning. That's why a size comparison helps to make the dimensions of the problem tangible.

The executive conclusion is clear: If the decision space is larger than anything we can grasp intuitively, then experience alone is no longer enough. Then optimization becomes a mathematical task.

Why classic portfolio logics must fail

Scoring models, ranking lists, traffic light systems and prioritization matrices all have one thing in common: they evaluate projects individually. Even if interactions are discussed, they usually remain qualitative.

However, the best individual project is not necessarily part of the best portfolio. A project can be excellent on its own - and still be detrimental in the portfolio because it blocks resources, accumulates risks or delays other projects.

The global optimum is always a property of the combination, not of the individual decision.

From local optima to the global optimum: the methodological break

A fundamental change of perspective is needed to move away from local optima:

• Not prioritizing projects, but optimizing combinations.
• Not discussing variants, but calculating spaces.
• Not replacing gut feeling, but quantifying the consequences of decisions.

This is the point at which traditional management methods end - and computer-aided decision-making intelligence begins.

StratePlan: calculation instead of gut feeling

StratePlan addresses precisely this gap. The approach is not to provide better arguments, but to explicitly calculate the decision space.

All projects are modeled with costs, impact, resource requirements, dependencies and restrictions. This results in a formal decision space that can be searched algorithmically.

Not a few dozen variants - but the entire space.

StratePlan calculates the entire decision spaceand finds from it:

The one project combination that generates the maximum overall benefit.

What "global optimum" really means in the executive context

A global optimum is not a theoretical ideal. It is a concrete reference:

• It shows which project combination delivers maximum benefit under the defined objectives and restrictions.
• It makes visible where local optima lie and how far they are from the global optimum.
• It quantifies the costs of any deliberate deviation.

This makes governance more precise. Decisions remain political and strategic - but they are no longer made blindly.

The real added value: transparent trade-offs

In practice, the greatest benefit is not that "the one best solution" exists. The greatest benefit is that alternatives become comparable.

If a committee wants to take on a project for political or strategic reasons, StratePlan shows what it will cost: What impact is lost, what risks increase, what projects are displaced.

This fundamentally changes the discussion. Opinions become conscious decisions.

Why Optima is the right thinking model for decision-makers

The term "optima" is not an academic detail. It precisely describes the reality of modern organizations:

• There are many good solutions.
• There are few very good solutions.
• There is exactly one combination that is best within the given framework.

Without calculation, organizations are stuck between good solutions. With calculation, it becomes clear which of them are actually optimal.

Executive Takeaway

Organizations do not fail because of bad decisions. They fail because they don't see the decision space.

Local optima feel right. The global optimum is usually invisible.

StratePlan makes this space visible - and calculable.

This turns optimization from a discussion process into a strategic capability.

Do you see your local optima - or do you know the global optimum of your portfolio?