Schema evolution in organizations
~7 min read
Core Primitive
Organizations must update their schemas as the environment changes — but most fail to do so until a crisis forces the update.
The same mechanisms that make schemas useful (they simplify decision-making by filtering information) make them resistant to change (they filter out the very information that would reveal their obsolescence). Deliberate schema evolution requires practices that counteract this natural resistance.
The schema that prevents its own replacement
Thomas Kuhn's The Structure of Scientific Revolutions described how scientific paradigms — shared schemas for interpreting the natural world — resist replacement. Scientists working within a paradigm do not see the anomalies that the paradigm cannot explain, because the paradigm determines what counts as relevant data. Anomalies are dismissed as measurement errors, edge cases, or topics for future investigation. The paradigm is replaced only when the anomalies become overwhelming — when the weight of unexplained evidence makes the paradigm untenable and an alternative paradigm is available. Kuhn called this a "paradigm shift" and demonstrated that it is always resisted by the scientific community that holds the existing paradigm (Kuhn, 1962).
Organizational schemas work the same way. The organization's schema determines what information is collected, how it is interpreted, and what actions are considered. Information that fits the schema is processed efficiently. Information that contradicts the schema is filtered, reinterpreted, or dismissed. The schema is a self-reinforcing system: it produces the information environment that confirms its own validity, while filtering out the information that would reveal its obsolescence.
This is not stupidity or stubbornness. It is a fundamental property of schemas: they simplify the world by filtering information, and the filtering inevitably excludes some information that is relevant. The filtering is the schema's greatest strength (it enables efficient action in a complex world) and its greatest vulnerability (it blinds the organization to changes that require the schema to be updated).
Mechanisms of schema resistance
Several mechanisms explain why organizational schemas resist evolution even when the environment is clearly changing.
Confirmation bias at scale. Individuals exhibit confirmation bias — the tendency to seek and interpret information that confirms existing beliefs. Organizations exhibit confirmation bias at scale: the data collection systems, reporting structures, and attention patterns are all shaped by existing schemas, which means the organization systematically collects information that confirms its schemas and underweights information that challenges them. An organization whose strategy schema says "our competitive advantage is engineering quality" will build metrics around code quality, system reliability, and technical performance — and may not build metrics around customer satisfaction, market share, or competitor innovation. The metrics the organization tracks confirm the schema. The metrics it does not track contain the signals that the schema needs updating.
Sunk cost commitment. Organizations that have invested heavily in a particular schema — building capabilities, hiring talent, designing processes, establishing partnerships around the schema's assumptions — resist abandoning the schema because abandonment means writing off those investments. Barry Staw's research on "escalation of commitment" demonstrated that both individuals and organizations increase their commitment to failing courses of action precisely because they have already invested in them. The sunk costs become a reason to persist rather than a reason to re-evaluate (Staw, 1976).
Identity threat. When schemas are entangled with organizational identity — "We are a hardware company," "We are a premium brand," "We are the innovation leader" — updating the schema feels like changing who the organization is. Identity schemas are the most resistant to evolution because they do not feel like assumptions — they feel like facts about the organization's essential nature. An organization that identifies as a hardware company cannot easily update its schema to include software, because the schema is not just about what the organization does — it is about what the organization is.
Power structure preservation. Existing schemas distribute power: the people whose expertise, relationships, and capabilities align with the current schemas hold influence. Schema evolution redistributes power to people whose expertise aligns with the new schemas. Those who benefit from the current schemas have strong incentives to resist evolution — not necessarily consciously, but through the subtle mechanisms of organizational politics: controlling which information reaches decision-makers, framing new data through the lens of existing schemas, and questioning the credibility of evidence that supports schema change.
The evolution cycle
When organizational schemas do evolve, they typically follow a cycle that mirrors Kuhn's paradigm shift.
Stability. The schema is well-adapted to the environment. Information confirms the schema. The organization operates efficiently. This phase can last years or decades.
Anomaly accumulation. The environment changes. Information that does not fit the schema begins to accumulate: missed forecasts, lost customers, competitive surprises, failed initiatives. Each anomaly is individually explainable within the existing schema ("That customer left because of price, not because our product is falling behind"). But the anomalies accumulate.
Crisis. The anomalies become undeniable. A major failure, a competitive shock, or a financial crisis overwhelms the schema's ability to explain events. The organization enters a state that Andy Grove called a "strategic inflection point" — a moment when the environment has changed so fundamentally that the old schemas are no longer viable (Grove, 1996).
Schema search. The organization searches for new schemas — new mental models that can explain the anomalies and guide action in the changed environment. This search is chaotic and contested. Multiple candidate schemas compete for adoption. Power struggles intensify as different factions advocate for schemas that align with their interests and expertise.
Schema adoption. A new schema is adopted — often through leadership change, because existing leaders are too invested in the old schemas to champion the new ones. The new schema reframes the organization's purpose, strategy, and operations. The organization begins adapting its processes, capabilities, and culture to the new schema.
Restabilization. The new schema becomes the operating default. Information collection, reporting structures, and decision processes are redesigned around the new schema. The cycle begins again.
Deliberate schema evolution
The crisis-driven evolution cycle is costly. By the time the crisis forces schema change, the organization has often lost significant competitive position, talent, and financial resources. The alternative is deliberate schema evolution — practices that enable the organization to update its schemas before a crisis forces the change.
Environmental scanning. Systematically collect information from outside the schema's filter. Talk to customers who left, competitors who are growing, and industries that are adjacent. The information that contradicts the schema is more valuable than the information that confirms it, because contradicting information reveals where the schema is misaligned with reality.
Schema stress testing. Periodically challenge the organization's core schemas with structured exercises. "What would have to be true for our strategy schema to be wrong?" "What signals would we expect to see if our process schemas were outdated?" "What would a competitor who held different schemas be doing that we are not?" The stress test reveals the schema's vulnerabilities before the environment exploits them.
Designated challengers. Assign specific people or groups the role of challenging existing schemas. Intel's "constructive confrontation" culture gave every employee the obligation to challenge the thinking of colleagues, regardless of hierarchy. Red teams, devil's advocates, and "loyal opposition" structures serve the same function: they create institutional pressure against schema calcification (Grove, 1996).
Schema versioning. Treat schemas like software: version them, document changes, and maintain a change log. "Our strategy schema was updated in Q3 2025 from 'win through product superiority' to 'win through ecosystem integration.' The change was triggered by the following evidence: [data]." Versioning makes schema evolution visible and deliberate rather than invisible and accidental.
Small experiments. Test alternative schemas through bounded experiments before committing to full schema change. "What if we operated from the schema that 'speed matters more than polish' for this one product line for this one quarter?" The experiment produces evidence about whether the alternative schema is more adaptive, without requiring the organization to abandon its existing schema organization-wide.
The Third Brain
Your AI system can serve as a schema evolution partner — helping the organization identify schemas that may be outdated, test alternative schemas, and manage the transition from old schemas to new ones. Share the organization's core schemas and ask: "Given current market conditions, technological trends, and competitive dynamics, which of these schemas are most likely to be outdated? For each, what would the updated schema look like? What evidence would we need to collect to determine whether the existing or updated schema is more accurate?"
The AI can also run schema stress tests: "Assume our strategy schema is wrong. What would a competitor who held the opposite schema be doing? Are any competitors doing that? If so, how are they performing?" The AI's analysis can reveal competitive threats that the organization's existing schemas are filtering out.
For ongoing schema health monitoring, use the AI to track the anomaly accumulation that precedes schema crises: "Here are our key metrics for the last four quarters. Here are our schema-based predictions for what those metrics should show. Where are the largest gaps between prediction and reality? Are the gaps growing or shrinking? What do the gaps suggest about which schemas need updating?" This monitoring catches schema obsolescence during the anomaly accumulation phase, before the crisis phase.
From evolution to infrastructure
Understanding how schemas form, propagate, and evolve provides the conceptual foundation for managing the organization's schema landscape. But schemas do not operate in a vacuum — they operate on a substrate of organizational knowledge: the facts, expertise, relationships, and institutional memory that the schemas interpret and organize.
The next lesson, The organization's knowledge graph, examines the organization's knowledge graph — the network of knowledge and expertise that organizational schemas depend on — and what happens when that knowledge network degrades.
Sources:
- Kuhn, T. S. (1962). The Structure of Scientific Revolutions. University of Chicago Press.
- Staw, B. M. (1976). "Knee-Deep in the Big Muddy: A Study of Escalating Commitment to a Chosen Course of Action." Organizational Behavior and Human Performance, 16(1), 27-44.
- Grove, A. S. (1996). Only the Paranoid Survive. Doubleday.
Put it into practice
Exercise
Identify one organizational schema that has been stable for more than five years. Ask: (1) What environmental conditions was this schema adapted to when it formed? (2) Have those conditions changed? (3) If the conditions have changed, has the schema been updated to reflect the new conditions? (4) If the schema has not been updated, what information is the schema filtering out that might signal the need for change? Write a one-paragraph description of what the updated schema might look like — the mental model the organization would hold if it were adapting to current conditions from scratch rather than continuing a historical schema.
More on putting it into practice
See it in action
A media company built its business around print advertising. Its core schema was: 'We connect advertisers with readers through high-quality editorial content.' For decades, this schema was perfectly adapted. Revenue came from advertisers, value came from editorial quality, and the medium was print. When digital media emerged, the schema filtered the new information in a predictable way. Online advertising was dismissed as low-quality ('Our advertisers want premium placement, not banner ads'). Digital content was dismissed as inferior ('Our readers expect editorial standards that blogs cannot match'). The schema correctly identified the differences between print and digital — and incorrectly concluded that those differences made digital irrelevant. The schema was not wrong about the present. It was wrong about the future. By the time the schema's obsolescence was undeniable — when print advertising revenue had declined 60% — the organization was too far behind to catch up. The schema had preserved itself by filtering out the very information that would have revealed the need for change. The CEO, Natasha, later reflected: 'We did not fail because we lacked information. We failed because our mental model of the business prevented us from interpreting the information correctly. Every signal that digital was the future was processed through a schema that said print was the present. The schema was the problem, and the schema was the last thing we were willing to change.' Natasha's reflection captures the fundamental challenge of schema evolution: the schema is both the lens through which the organization sees the world and the barrier that prevents it from seeing what the world is becoming.
Watch for the failure mode
Two failures that are mirror images. The first is schema rigidity — refusing to update schemas until a crisis forces the change. This produces organizations that are perfectly adapted to the past and catastrophically maladapted to the present, which is the pattern described in the example above. The second is schema volatility — updating schemas so frequently that the organization never develops the deep competence that comes from sustained schema commitment. An organization that changes its strategy schema every quarter, its process schemas every month, and its values schemas with every new leadership hire never develops the operational depth that stable schemas provide. The discipline is to distinguish between schemas that need to evolve (because the environment has changed) and schemas that need to persist (because they represent hard-won wisdom that remains valid).
Make it stick
This lesson completes the first half of Organizational Schemas. Return to Organizations run on shared schemas and trace the arc: organizations run on shared schemas, these schemas are often implicit (Organizational schemas are often implicit), making them explicit enables examination (Making organizational schemas explicit), and the major schema types — strategy (Strategy is an organizational schema), process (Process is an organizational schema), values (Values are organizational schemas) — combine to produce culture (Culture is the sum of organizational schemas). Schemas can conflict (Schema conflicts within organizations) and propagate (Schema propagation in organizations), and now we understand how they evolve over time. The second half of the phase (The organization's knowledge graph through Healthy organizational schemas produce healthy organizational behavior) shifts from understanding schemas to managing them: the organization's knowledge graph, institutional knowledge loss, documentation, organizational learning, schema debt, and schema design as leadership work. Look forward to The organization's knowledge graph, which examines the organization's knowledge graph — the network of knowledge, expertise, and institutional memory that the organization's schemas operate on.
Practice
Map a Legacy Schema's Evolution Gap in Miro
Create a visual map showing how an organizational schema formed under past conditions and identify what's being filtered out in current conditions. This makes invisible resistance to change visible and concrete.
- 1Open Miro and create a new board titled 'Schema Evolution Analysis.' Create three columns: 'Original Context (5+ years ago)', 'Current Schema', and 'Current Reality.' In the first column, use sticky notes to document the environmental conditions when a stable organizational schema formed (e.g., 'Waterfall development was standard when our approval process was created').
- 2In the 'Current Schema' column, map the mental model your organization still operates from using shapes and text. Include decision filters (what gets attention vs. ignored), key assumptions (e.g., 'customers prefer stability'), and standard responses (e.g., 'always escalate to VP before launch'). Connect these elements with arrows to show how the schema reinforces itself.
- 3In the 'Current Reality' column, document environmental changes that have occurred using contrasting colored sticky notes. For each change, draw a dotted line to the schema element it contradicts (e.g., 'Competitors ship daily' contradicts 'customers prefer stability').
- 4Below your three columns, create a 'Filtered Information' section in Miro. List specific signals, data, or feedback that the current schema causes people to dismiss or not notice. Use red sticky notes for high-impact filtered information that would most strongly signal the need for schema change.
- 5At the bottom of your Miro board, create a text box titled 'Updated Schema' and write a one-paragraph description of what mental model the organization would adopt if designing from scratch today. Reference specific elements from your 'Current Reality' column and explain how this new schema would process information differently than the legacy one.