Compliance Isn’t Adoption

The mandate goes out on Thursday. By Tuesday, the dashboard shows 100% completion. Six months later, nobody is doing it. We’ve watched this sequence play out enough times in mid-market organizations to know it’s the default outcome of treating cultural change with structural tools—the gap between strategy and execution that no policy memo can close.

That gap costs change initiatives their credibility, and it explains a question we keep getting from executives: how do we make the habit stick without a stick? The honest answer is that we have to stop confusing compliance with adoption. Compliance is what people produce when they’re being measured. Adoption is what they do when nobody’s watching. Mandates can drive the first. They cannot produce the second, and most of the work that organizations actually need lives in the second category.

Argyris and Schön named the underlying mechanism more than fifty years ago. What we say we believe lives in town halls and policy memos. What we actually believe shows up in our calendar, our promotion decisions, and what we tolerate when no one’s looking. When a leader mandates a new habit, the espoused theory says, “We want you to practice this.” The theory-in-use revealed by enforcement is “we don’t trust you to choose it.” Teams hear the second message louder than the first. That’s why adoption stalls.

Karl Weick saw this clearly in 1984. His work on small wins, published in American Psychologist, made the case that durable change emerges through “concrete, complete outcomes of moderate importance” that build a pattern that attracts allies and deters opponents. The mechanism is social and observational, not directive. The general failure rate for organizational change initiatives runs about 70%, a figure cited consistently across Cameron’s research, Beer and Nohria’s Harvard Business Review work, and decades of McKinsey studies. The pattern beneath that headline number is consistent: top-down, mandate-driven change underperforms collaborative, incrementally built change. The math has been clear for forty years. The simple fact is that mandates feel decisive, and catalysts feel slow, so we keep choosing the wrong tool.

A catalyst creates the conditions under which a behavior spreads through observation and peer influence rather than through enforcement. The shift required is from pushing harder to building the proof that pulls. That requires a different sequence and a different timeline.

The work runs in four phases over 12 to 18 months, and each phase exists to create proof for the next.

The first two months are about finding the volunteers, not the appointees. Every organization has 5 to 10% of its people who are already curious about the new habit, experimenting, or practicing some version of it. Find them. The temptation is to recruit by hierarchy—a senior VP “championing” the initiative, a manager appointing direct reports as adopters. Both moves trigger reactance and signal top-down imposition. Recruit instead by genuine interest, across roles and levels. Give the early group practice materials, modest visibility, and protected time to experiment. Resist the urge to scale. The work in this phase is depth, not width.

Months three and four are about making the practice visible peer-to-peer. This is where most rollouts go wrong. They make adoption visible through executive announcements, which signal “the boss wants this” and trigger the same compliance theater as the original mandate. The work instead is to build channels where adopters tell their own stories to peers, in their own words, with specific examples. A regular forum where five adopters describe what they actually did, what changed, and what they got wrong creates social proof. Glossy testimonials curated by communications teams don’t. The texture of imperfect, honest peer accounts is what spreads. Polish kills it. Senior leaders should model here, but they should model the learning, not the mastery. A leader who says, “I’ve been trying this for three months and I’m still bad at it, here’s what surprised me,” produces more credibility than one claiming a mastery they haven’t earned.

By months five through eight, the early adopters can teach the next wave. Communities of practice form — small groups of practitioners coaching each other with a light structure and no central authority. The temptation is to formalize and scale. Resist it. Centralized scaling kills peer credibility. Let the network grow through invitation.

By month nine and into the second year, late adopters start moving on their own. Colleagues, not campaigns, are doing the work of persuasion. The executive role shifts from creating adoption to removing barriers—making sure systems, calendars, and incentives don’t punish people who practice the new habit. This is where the work feels like it’s happening on its own, which is exactly what we wanted. We just had to wait for the proof to compound.

Three failure modes will end the initiative if we don’t watch for them. The first is mandate creep. When adoption looks slow at month four, leaders panic and mandate, which destroys everything the earlier phases built. The second is performance theater—early adopters start performing visibility rather than practicing the habit. We mitigate by rewarding depth in the storytelling. “Tell us what you got wrong” produces more credible proof than “tell us your win.” The third and most fatal is senior leader hypocrisy. Leaders endorse the habit publicly while their calendars and decisions reveal they’re not practicing it. Peers see straight through this. The theory-in-use gap kills social proof faster than any other variable. If the senior team won’t practice the habit, the right move is to end the initiative rather than let the gap erode trust.

There’s an equity consideration that’s most often skipped. Social proof can entrench existing power dynamics if we’re not deliberate. In organizations under high workload pressure, “voluntary” adoption tends to skew toward people with greater autonomy—usually those higher in the hierarchy or in less-monitored roles. If our adopter stories all feature senior, white, or English-first voices, we encode the message that the new practice belongs to that group. We have to recruit deliberately from compressed roles, track adoption by demographic group, and investigate systematic non-adoption before interpreting it. The answer is rarely “they don’t want to.”

The protocol is the easy part. Holding the line at month four—when adoption looks slow and the instinct to mandate returns—is where the executive role matters most. That’s where leaders earn their pay: protecting the slow approach long enough for the proof to compound. Catalysts work. They just take longer than the quarter we wish they took, and they require leaders willing to defend the timeline against their own impatience.

This essay synthesizes research by Chris Argyris and Donald Schön on espoused theory vs. theory-in-use; by Karl Weick on small wins in American Psychologist; and the broader literature on organizational change failure from Cameron, Beer and Nohria, and McKinsey.

Contact us for a complete list of works cited.