The Physics of Organizational Trust and Rapid Transformation
Complex transformations do not move through strategy decks but through trust, safety, and the people an organization already believes
At a Glance: Traditional business transformations often fail because leaders mistakenly treat behavioral change as a simple matter of communication rather than a complex social shift. While companies like Tesla succeed through integrated, product-centric structures, legacy organizations struggle to replicate this agility because employees prioritize social safety and peer validation over top-down mandates. By utilizing Organizational Network Analysis, firms can identify informal influencers who hold the most trust within the workforce. The case illustrates that centering a transformation around these trusted peers can accelerate adoption significantly. Ultimately, successful execution requires building a social architecture that makes abandoning old habits feel safe for the individuals involved.
If you’d rather take this piece on a walk or drive, I used Google’s NotebookLM to create a short podcast-style conversation that explores the same ideas from a slightly different angle.
In the late 2000s, the traditional automaker looked like a monument to managerial logic. Thousands of brilliant minds were organized into distinct, optimized groups: a braking department, an interior trim division, a powertrain team. Each unit was excellent at its specific function. The weakness was not inside the departments. It was in the seams between them.
When a vehicle rolled off the assembly line, its capabilities were locked in time. If a braking system required an optimization post-launch, it meant a multi-year development cycle for the next model year, or a costly, logistically painful recall.
Then came Tesla. In 2018, Consumer Reports noted that the Model 3 had a braking distance flaw — its stopping distance exceeded that of a full-sized pickup truck. Tesla didn’t initiate a recall or wait for the next model cycle. Within days, its engineers pushed a remote firmware update over the air. The update altered the software code controlling the anti-lock braking algorithm, shortening the car’s stopping distance by nearly twenty feet.
It is tempting to see this as a story about software. It was partly that. Over-the-air updates gave Tesla a delivery mechanism most automakers were still learning to match. But the deeper advantage was organizational: the workforce architecture allowed the right people to find each other, work across boundaries, and solve an interdependent problem quickly. The software update was the visible event. The collaboration system underneath it was the real capability.
This is the part executives admire. What they often underestimate is what it takes for a traditional project-centric organization to become product-centric. Tesla’s update looked technical from the outside. Inside a legacy company, the same capability would require people to surrender familiar roles, decision rights, approval paths, and sources of influence.
Tesla’s agility is possible because of their organizational structure: software and hardware engineers sit at the same table, treating the car not as a series of projects with a fixed end date but as a living system that learns continuously. In traditional automakers, those engineers lived in different buildings, operated on different calendars, and handed work to each other across organizational walls. By the time a problem surfaced in the market, the people who could fix it were already three model years deep into the next project.
Over time, this difference compounds. One model improves through heavy, start-stop coordination cycles; the other learns while already in the market. Boards and CEOs everywhere see this latter capability as an advantage and want to replicate it. They announce product-centric transformations and hire consultants to restructure reporting lines and rename departments.
And then they discover that the organizational physics are working against them in a way they did not anticipate. McKinsey has estimated that roughly 70% of large-scale transformations fail to meet their objectives. When your odds of failure are more likely than not, it’s no wonder many organizations hesitate.
But anyone who has lived inside a large transformation knows the strategy is rarely the whole problem. The employees being asked to change their behavior have spent ten or twenty years building professional competence inside the existing system. They know the rules, they understand how to succeed, and they have accumulated influence by mastering a model that is now being discarded by outsiders. A corporate directive can tell them the destination. It cannot tell them whether the new way of working leaves their roles safe or if they’ll have relevance on the other side of change.
That distinction — between knowing the strategic aims and feeling safe enough to adopt — is where most transformations quietly die. Strategy defines the target. Execution requires something far more specific: understanding the social architecture through which behavior actually changes inside an organization.
The failure to treat these as distinct problems is not a minor oversight. It is one of the quiet reasons transformation programs fail: organizations announce change before they have designed the conditions under which people can safely adopt it.
A large financial services company I’ll call “Northbank” faced exactly this challenge.1 The strategy was to transition more than 10,000 employees out of decades-old, project-based routines and into a modern, product-centric operating model. The change touched every legacy system and business unit across the institution.
The consulting firm brought in to design the roadmap studied the transformation’s scope, modeled the organizational weight — the accumulated routines, dependencies, and institutional habits of a company built on project-based work — and delivered their verdict: ten years.
Their logic was sound by conventional standards. Traditional transformation playbooks assume adoption must move through the formal hierarchy. You secure executive sponsorship, brief the middle managers, cascade the message down the reporting lines, run the training modules, and track compliance on an adoption dashboard.
Communication creates awareness. Awareness, the model assumes, produces behavior change.
We ran the playbook. The town halls happened. The decks were cascaded. And then we watched what actually followed: employees nodded in the meetings, returned to their desks, and continued working exactly as they had before. Compliance was surface-level. The language had moved; the operating model had not.
Damon Centola, a sociologist at the University of Pennsylvania, studied how behaviors propagate through social networks by examining online health communities — groups where people were adopting new and often unfamiliar health practices. To differentiate how we think change happens and how it actually unfolds, Centola distinguished between two fundamentally different phenomena: simple contagions and complex contagions.
A simple contagion, representing how we think change occurs, behaves like a virus or a piece of news. It requires only a single point of contact to spread. Exposure to a new idea, it is believed, is sufficient.
Shifting how a human being does their job, however, is a complex contagion.
In a corporate setting, people rarely resist change because they failed to understand the memo. They hesitate because changing their behavior carries personal and social risk. When the operating model shifts, employees ask themselves pragmatic questions: Will my manager reward this new way of working? Will my peers respect this, or will I look incompetent? Will I lose the influence I spent a decade earning?
To cross the threshold of a complex contagion, an individual needs more than a clear explanation. They need multiple points of social confirmation proving that it’s safe for them to adopt.
Centola’s research formalized this as a behavioral threshold. Different people have different thresholds of risk tolerance; they need to see two, three, or more trusted peers adopt a behavior before they feel safe enough to cross their own threshold. A town hall can broadcast awareness to 15,000 people simultaneously. It cannot manufacture safety. Safety only occurs when an employee looks horizontally and sees someone like themselves — not an executive with a mandate, but a peer they actually respect — already moving.
This is why the cascade had stalled adoption at Northbank. We had been running a simple contagion playbook against a problem that was far more complex in nature.
The breakthrough was recognizing that you cannot force people across behavioral thresholds, but you can deliberately engineer the pathways through which social safety travels.
We shifted our focus from looking at where formal authority sat and started identifying where relational trust resided. To do this, we leveraged Organizational Network Analysis — a method for making the informal social structure of an organization visible. Instead of asking leaders to guess who the “change agents” should be, we allowed peers to identify the people they already trusted.
When we laid out the results, the formal org chart effectively dissolved. The network revealed a different, hidden anatomy of influence. We did not simply give those people talking points or ask them to become ambassadors. We designed the transformation around them. This was the shift from strategy development to strategy execution.
Leveraging Centola’s behavioral insights, we designed an in-house university and made it the single point of entry through which all employees would encounter the new operating model. Faculty were not consultants parachuted in from outside the organization, nor executives issuing directives from a safe distance. Instead, they were the peer influencers the ONA had identified — mid-level engineers, veteran project managers, analytical anchors who had earned the organization’s trust through years of work inside it.
This design mattered more than it might appear. Teams came through the Academy together — not as individuals scattered across a training catalog, but as intact working units, sitting alongside their direct managers. They were taught by someone they already recognized as credible. They watched their teammates try the new patterns of work in real time. They watched their managers respond to those attempts not with skepticism but with visible reinforcement. The room changed the risk calculation. The social confirmations that Centola’s research identified as necessary to cross a behavioral threshold were not left to chance. They were built into the room.
The thresholds did not erode one employee at a time. They shifted together, inside a shared experience. The consequence we hadn’t fully anticipated was what happened on the other side of that collapse. Demand outran our capacity to meet it. Teams that had been through the Academy wanted to return to improve their skills. Teams that hadn’t been through it were asking when they could come.
We went back to the senior leadership and made the case for additional headcount — not for the Academy’s program broadly, but to accelerate the transformation’s progress. They approved the investment, and we expanded the curriculum, added elective coursework beyond the core model, and gamified learning to reinforce new behaviors.
What made the difference was not a new philosophy about how transformations should unfold. It was a more honest accounting of how change moves through organizations. Instead of guessing at influence, or asking managers to volunteer the names of people they trusted, we calculated it. ONA told us where trust actually lived. The Academy gave us a way to activate it — by putting the organization’s real carriers in front of their peers, under conditions specifically designed to manufacture the social safety that adoption requires.
The final result: we compressed what had been estimated as a nearly decade-long transformation into roughly two and a half years. The difference was not a better communication strategy. It was a different theory of how organizations execute strategy.
The high failure rate of corporate transformation programs is not an indictment of executive ambition or strategic clarity. It is a predictable consequence of using structural tools — reporting lines, compliance dashboards, communication cascades — to solve what is fundamentally a social adoption problem.
Tesla proved that a product-centric model could make a century-old industry’s assumptions look obsolete. What the Northbank case demonstrates is something equally important: legacy organizations are not structurally condemned to stay where they are. The gap between where they may be and where the future dictates is not closed by announcing a transformation. It is closed by understanding how change actually propagates through a large human system — and designing the execution around that reality rather than around the org chart.
The ten-year estimate was not wrong because the consultants were incompetent. It was wrong because it was built on an incomplete theory of how organizations move. When you replace that theory with one grounded in network science — when you stop broadcasting change and start seeding it through the people the organization is already watching — the timeline compresses in ways that feel improbable until you understand the mechanics.
This is ultimately a reframe of what strategy execution means. Strategy defines the destination. Execution is not the project plan for getting there. It is the social architecture you build to make movement feel safe.
That is the part most transformation plans still under-design.
About the Author
Jason L Zimmerman is the founder of 3Fold Collective, a consulting firm that helps leaders close the gap between strategy and execution. 3Fold works with organizations navigating complex change, using behavioral economics, organizational network analysis, and operating model design to understand how work really moves through the business. The firm’s focus is simple: helping leaders build systems where strategy can actually take hold.
1 Some identifying details have been anonymized and lightly generalized to protect confidentiality while preserving the underlying transformation lesson.
References:
Avramidis, E., et al. (2019). Change agents and internal communications in organizational networks. Physica A: Statistical Mechanics and Its Applications, 528, 121385.
Cagan, M. (2024). Product model concepts. Silicon Valley Product Group. https://www.svpg.com/product-model-concepts/
Centola, D. (2010). The spread of behavior in an online social network experiment. Science, 329(5996), 1194–1197.
Centola, D. (2021). Change: How to make big things happen. Little, Brown Spark.
Cross, R., & Parker, A. (2004). The hidden power of social networks: Understanding how work really gets done in organizations. Harvard Business School Press.
Krackhardt, D., & Hanson, J. R. (1993). Informal networks: The company behind the chart. Harvard Business Review.
McKinsey & Company. (2019). Why do most transformations fail? A conversation with Harry Robinson. McKinsey & Company.
Olsen, P. (2018, May 30). Tesla Model 3 gets CR recommendation after braking update. Consumer Reports.
For a related look at why execution often breaks in places leaders cannot see, I explored a similar idea in “The Driver and the Bridge.” It’s about the very human habit of blaming the driver when the real problem is a collapsed bridge, or in organizational terms, mistaking poor performance for what is actually a broken system.
