A common pattern in organizations that are struggling: no single element is obviously broken. The strategy is defensible. The org chart is reasonable. The people are competent. The incentive structure isn't crazy. And the organization still isn't working well, because the pieces don't actually fit each other - a genuinely well-designed structure paired with an incentive system that rewards the opposite behavior produces dysfunction that no single-element diagnosis will find, because every element passes inspection on its own.
The four elements
The Congruence Model, developed by David Nadler and Michael Tushman, names four elements that have to fit together, not just individually make sense:
Work - the actual tasks that need to get done, and what doing them well requires.
People - the individuals doing the work, their skills, and what they need to perform.
Formal structure - the official organization: reporting lines, processes, systems, incentive structures, how decisions are officially made.
Informal structure (culture) - the unwritten norms, actual behavior patterns, and real (not stated) values that shape how people actually act, which can diverge sharply from the formal structure on paper.
Congruence, not correctness, is the diagnostic question
The model's central claim isn't that each element needs to be individually excellent - it's that misalignment between any two elements produces friction, regardless of how good each one looks in isolation. A formal structure that rewards individual output paired with work that genuinely requires collaboration is a congruence problem, not a "people" problem or a "structure" problem in isolation - fixing either element alone without addressing the mismatch between them tends to just relocate the friction rather than resolve it.
A concrete example of the mismatch
A team's formal structure states clearly that decisions are made collaboratively, with input from everyone before a call is made. The informal structure - what people have actually learned happens in practice - is that one person's preference reliably wins regardless of the stated process. The formal structure and the informal structure are misaligned, and the practical effect is that people stop offering real input (since it doesn't change outcomes) while continuing to go through the motions of the collaborative process, producing exactly the appearance of collaboration with none of its actual value. Neither element is "wrong" on its own - the stated process is reasonable, and the person whose preference wins might even be making good calls - the problem is specifically the gap between what's declared and what's real.
Using this as a diagnostic, not just a description
The practical use of the model is as a checklist when something isn't working and the cause isn't obvious: look at each pair of elements - work and people, work and formal structure, work and informal structure, people and formal structure, people and informal structure, formal structure and informal structure - and ask whether they actually fit, not whether each one is individually well-designed. Most organizational friction that survives a first round of individual fixes (better people, clearer structure, more explicit process) is a congruence problem between elements that were each addressed separately without checking whether they now fit each other.
This is a different, complementary lens to symptom versus root cause - that page is about pushing past the first, most visible explanation for a single recurring problem; this one is about checking whether several individually-reasonable elements are actually consistent with each other, which a single-cause root-cause analysis can miss entirely.