You ran the meeting. You laid out the problem. People asked good questions, you answered them, you agreed on the path. Heads nodded. Someone took the action. You left the room feeling like the issue had finally been resolved. Three weeks later, nothing has changed. You ask about it. You get answers that sound reasonable but do not actually explain why nothing happened. The thing everyone agreed to is still sitting where it was. And no one seems to be lying. This is one of the most quietly frustrating patterns in running a small business. It is not dishonesty. It is not laziness. It is something more durable, and more useful to understand once you can see it.
The gap between what we say and what we do
The Harvard organisational scholar Chris Argyris spent his career on a single observation: there is almost always a gap between what people say they would do and what they actually do. He called the first one the espoused theory — the principles, the values, the agreement reached in the meeting. He called the second one the theory-in-use — the rules people are actually operating from when no one is watching. Crucially, people are not aware of the gap. If you played someone a video of themselves behaving in a way that contradicted what they said in the meeting, they would be genuinely surprised. The espoused theory is not a lie. It is what the person sincerely believes they would do. The theory-in-use is what they actually do, governed by a different set of rules that operates underneath. This gap is not a moral failure. It is how human beings function under social pressure. Argyris spent decades documenting it and concluded that most adults in professional settings operate from one of two patterns. He called them Model I and Model II.
Why Model I meetings produce agreements that do not hold
Model I has four governing rules, usually unspoken. Stay in control of the situation. Make sure you win, or at least do not lose. Do not show difficult emotions. Be — or at least appear to be — rational. Most professionals run on this, including the ones who would not describe themselves that way. The reason agreements made in a Model I meeting do not translate into action is that the agreement itself is shaped by Model I. People agree because disagreeing risks losing the moment, or looking irrational, or making things uncomfortable. The agreement is genuine in the moment, but it has been made under conditions that filtered out the information that would have made disagreement honest. When everyone goes back to their desk, the actual rules — the theory-in-use — take over again, and the agreed action turns out to conflict with one of them.
What sits between the meeting and the action
What sits between is what Argyris called a defensive routine. A defensive routine is a pattern of behaviour that protects the group from threats to its current view of itself, while making the underlying problem undiscussable. The classic shape is this. Someone raises a concern that would, if pursued, force a difficult re-examination. The group acknowledges it, redirects gently, and moves on. The concern is now “addressed” without ever being engaged with. Everyone leaves with the same view they entered with. Defensive routines work. That is why they survive. They protect relationships, status, and the comfortable assumption that the room is rational. The cost is that nothing learned in the meeting actually transfers into action.
What Model II looks like
The alternative Argyris spent the rest of his career describing — Model II — has three principles. Use valid information. Allow free, informed choice. Build internal commitment. In a meeting it looks something like this. People share the data they actually have, including the data that contradicts their preferred direction. They ask for the reasoning behind other people’s positions, not to debate but to understand. When someone disagrees, the disagreement is treated as new information, not as a problem to manage. The decision that comes out of the meeting is one people are internally committed to, because they have examined it rather than complied with it. That sounds easy when you read it. It is not. Most adults need significant practice to operate this way under pressure, because Model I is the default we have all been rewarded for since school. Model II is a learned discipline, not a personality trait.
Three diagnostic questions for your own meetings
If you want a quick test on your own meetings, three questions will tell you most of what you need to know. How often does the meeting end with a decision that the most junior person in the room would feel safe disagreeing with publicly the next day? If the answer is rarely or never, you are running a Model I meeting. The decisions reached are espoused-theory decisions. They will not survive contact with the actual work. When someone raises an objection, what happens in the next sixty seconds? If the room moves to address the source of the objection (the person), rather than the substance of the objection (the data), you have a defensive routine. If the room gets curious about the objection, you have the early shape of Model II. In the week after the meeting, who tells you what is actually going on? If the picture you get from one-on-ones contradicts what was agreed in the room, the meeting did not produce learning. It produced compliance.
One question to ask in your next meeting
The shift to Model II is not a memo. It is a discipline. But there is a small move that costs nothing and reliably opens space. In your next meeting, when a decision is being reached, ask one question out loud: “What would have to be true for this to be the wrong decision?” Then stay quiet long enough for someone to answer honestly. That question, asked seriously, breaks one of Model I’s strongest rules — the rule that says we move on once a position has been taken. It signals that disagreement is data, not disloyalty. If people answer it, you have started a different kind of meeting. If no one answers, that is also data. It tells you that the room is not yet a place where the costs of disagreement feel safe. That is not a flaw in the people. It is the structure of the meeting. And it can be changed.
Why this matters more than it looks like it does
The gap between what teams agree to and what they do is not a sign that people are unreliable. It is a sign that the meeting was shaped by rules that filtered out the information that would have made the agreement durable. Those rules are old, deep, and human. They are also visible once you know what to look for. That is what Argyris meant by learning. Not “did people pick up new facts” but “did the group change the rules it operates from”. A team that learns to do that is a team where the action matches the agreement, more weeks than not.