Change the system to change the outcomes

The exhortation fallacy

The most common response to unsatisfactory organizational outcomes is exhortation: ask people to try harder, care more, pay closer attention, or be more committed. The exhortation assumes that the gap between current outcomes and desired outcomes is a motivation gap — that the people operating the system could produce better outcomes if they chose to.

This assumption is almost always wrong. In most organizations, the people operating the system are already doing their best within the constraints the system provides. They are making rational choices given the incentives, information, and options available to them. The gap between current and desired outcomes is not a motivation gap — it is a design gap. The system is designed (often unintentionally) to produce the outcomes it produces.

Russell Ackoff, the systems theorist, captured this insight precisely: "The righter we do the wrong thing, the wronger we become. When we make a mistake doing the wrong thing and correct it, we become wronger. When we make a mistake doing the right thing and correct it, we become righter. Therefore, it is better to do the right thing wrong than the wrong thing right." The "wrong thing" in organizational change is trying to improve outcomes without changing the system that produces them. No amount of doing the wrong thing — exhorting, training, coaching, replacing — can produce the right outcome when the system itself is misaligned (Ackoff, 1999).

Three failed approaches

Before exploring what works, it is worth understanding in detail why the three most common non-systemic approaches fail.

Training without system change

Training teaches people new skills, knowledge, or behaviors. But training operates on the individual, not the system. A trained individual returns to the same system — the same incentives, the same information flows, the same constraints, the same processes — and the system reasserts its influence. Research consistently shows that training transfer rates (the percentage of trained behavior that actually appears in the workplace) are between 10% and 20% when the organizational system does not change to support the new behavior (Baldwin & Ford, 1988).

The math is unforgiving. If you train 100 people in a new practice but the system does not support the practice, 80-90 of them will revert to the old behavior within weeks. The training investment is largely wasted — not because the training was poor but because the system was not changed to accommodate the trained behavior.

Training works when it is paired with system change: when the structures, incentives, and processes are redesigned to support the trained behavior. Without system change, training is a cost center that produces temporary skill acquisition and permanent frustration.

Motivation without system change

Motivational interventions — inspiring speeches, team-building exercises, recognition programs, engagement initiatives — attempt to increase the emotional energy people invest in their work. But emotional energy operates within the system, not on the system. A highly motivated person operating within a poorly designed system will produce marginally better outcomes than an unmotivated person in the same system — but both will produce outcomes bounded by the system's design parameters.

Frederick Herzberg's two-factor theory distinguished between "hygiene factors" (system properties like working conditions, compensation, and policies) and "motivators" (individual experiences like achievement, recognition, and growth). Herzberg found that motivators could increase satisfaction and performance only when the hygiene factors were adequate. Motivation cannot compensate for system dysfunction — it can only amplify what a functional system enables (Herzberg, 1968).

Personnel changes without system change

Replacing people is the most dramatic non-systemic intervention — and the most clearly futile when the system is the cause. If the system produces a 5% defect rate, replacing the workers produces a 5% defect rate with different workers. If the management system produces burned-out managers, replacing the burned-out manager produces another burned-out manager.

Personnel changes work when the system is sound but the individual is genuinely mismatched — wrong skills, wrong temperament, wrong values for the role. But when the outcome is systemic (it recurs across multiple individuals in the same role), personnel change is misdiagnosis in action.

What system change looks like

Effective system change modifies the structural elements that produce the current outcome and replaces them with elements that produce the desired outcome. There are four categories of system change, each operating on a different system element.

Structural change

Structural change modifies the organization's formal architecture: reporting lines, team composition, role definitions, resource allocation. Structural change works when the current structure creates misalignment between the people doing the work and the outcomes the organization needs.

A company that reorganizes from functional silos (engineering, design, product) to cross-functional product teams is making a structural change that eliminates coordination overhead and aligns team incentives with product outcomes rather than functional metrics. The change does not require people to try harder — it makes the desired behavior (cross-functional collaboration) the structural default.

Incentive change

Incentive change modifies what the organization measures, rewards, and punishes. Incentive change works when the current incentives drive behavior that contradicts the desired outcome — when people are literally being rewarded for producing the wrong result.

Steven Kerr's classic paper "On the Folly of Rewarding A, While Hoping for B" cataloged the organizational tendency to create incentive systems that reward the exact opposite of the desired behavior: rewarding individual performance while hoping for teamwork, rewarding short-term results while hoping for long-term investment, rewarding conformity while hoping for innovation (Kerr, 1975). Fixing the incentive misalignment is often the highest-leverage system change available.

Information change

Information change modifies who knows what, when they know it, and how they receive it. Information change works when decision-makers lack the information they need to make good decisions — or receive it too late to act on it.

Making previously hidden information visible can transform behavior without any other intervention. When a hospital made hand-washing compliance data visible in real time (displayed on screens in each ward), compliance rates rose from 30% to 90% — not because people were punished for non-compliance but because the information created social accountability that the previous system lacked.

Process change

Process change modifies the sequence, timing, and method of work activities. Process change works when the current workflow creates bottlenecks, errors, or delays that produce the undesired outcome regardless of individual effort.

Toyota's production system is the canonical example: by redesigning the production process to enable workers to stop the line when they detect a defect (the "andon cord"), Toyota changed the system so that quality was built into the process rather than inspected after the fact. The process change eliminated the systemic source of quality problems — defects propagating downstream — without requiring workers to be more careful or more skilled.

The system always wins

There is a law-like regularity in organizational life: when individual behavior conflicts with system design, the system wins. Not immediately — motivated individuals can fight the system for weeks, sometimes months. But system design operates continuously, automatically, and without fatigue. Individual motivation is episodic, effortful, and exhaustible. Over time, behavior converges toward what the system rewards, enables, and makes easy — regardless of what individuals intend or desire.

This is not cynicism. It is engineering. A bridge designed to bear ten tons will fail under twenty tons regardless of how much the engineer wants it to hold. An organizational system designed to reward individual competition will produce competitive behavior regardless of how much the leader wants collaboration. The system is not defiant — it is deterministic within its design parameters.

The practical implication is liberating: if you want to change the outcome, stop fighting the system and redesign it. Stop asking people to overcome the system through willpower and instead make the system produce the outcome you want. This is harder than exhortation — it requires understanding the system, identifying the high-leverage elements, and investing in structural change. But it produces durable results, because the redesigned system continues to produce the desired outcome long after the change agent has moved on.

The Third Brain

Your AI system can help you design system changes by modeling the interactions between system elements. Describe the current system (structures, incentives, information flows, processes) and the desired outcome change, and ask: "Design three system interventions that would shift the outcome from [current] to [desired]. For each intervention, specify: (1) which system element is being changed, (2) what the change is, (3) why this change would shift the outcome, (4) what unintended consequences might emerge, and (5) how the change would interact with other system elements. Rank the interventions by leverage — which one produces the largest outcome shift for the smallest implementation cost?"

The AI can also help you anticipate resistance: "We are planning to change [system element]. Who benefits from the current system design? Who would benefit from the proposed change? What resistance should we anticipate, and how can the change be designed to address the legitimate concerns of those who benefit from the current system?"

From principle to practice

The principle is clear: change the system to change the outcomes. But applying this principle requires a specific skill — the ability to identify the system before attempting to change it. Many system change efforts fail not because the change was wrong but because the change agent misidentified the system — addressing a subsystem while the real driver sits in a different part of the organization.

The next lesson, Identify the system before trying to change it, establishes this prerequisite: identify the system before trying to change it. Without accurate system identification, even well-intentioned system changes can produce no improvement — or, worse, unintended harm.

---

Sources:

1. Ackoff, R. L. (1999). Ackoff's Best: His Classic Writings on Management. Wiley.
2. Baldwin, T. T., & Ford, J. K. (1988). "Transfer of Training: A Review and Directions for Future Research." Personnel Psychology, 41(1), 63-105.
3. Herzberg, F. (1968). "One More Time: How Do You Motivate Employees?" Harvard Business Review, 46(1), 53-62.
4. Kerr, S. (1975). "On the Folly of Rewarding A, While Hoping for B." Academy of Management Journal, 18(4), 769-783.