Somewhere in your organization, right now, a group of people is running a study on you.
Not consciously. Nobody has designed it and nobody would describe it that way. But your people are gathering evidence about what happens here — what gets escalated and what gets buried, which arguments get had and which ones cost more than they return, what happened the last three times somebody brought you something you did not want to hear.
They are doing this because they have to. Every organization runs on predictions about what leadership will do, and those predictions have to come from somewhere. So people watch, and they compare what they were told against what followed, and where the two disagree they go with what followed.
That is the whole premise of Leadership by Design, and it is why the framework treats leadership as a design problem rather than a communication problem.
Signal is not intention. It is consequence.
What is signal integrity in organizational leadership?
Signal integrity is the degree to which what a leader intends to communicate matches what their organization experiences consequently from them.
It is the foundational law of the framework and the first phase of the eight-phase organizational lifecycle, and it governs everything downstream. High signal integrity means the organization’s working model of you is accurate — people can predict what you will support, what you will challenge, and what will happen when something goes wrong, and they are usually right. Low signal integrity means the model has drifted. People are still predicting, because they must, but they are predicting from a version of you that you would not recognize and cannot correct, because nobody has told you it exists.
The gap between the two is where organizational drag comes from. Not from bad people, and not from insufficient effort. From a workforce operating accurately against a signal you did not know you were sending.
Every organization has a working model of its leader. The question is not whether it exists. It is whether it matches.
Why does what a leader says matter less than what they do?
Because your words are a claim about the future and your behavior is evidence, and people who have to act under uncertainty use evidence.
This is not cynicism about your team. It is the same reasoning you would use in their position, and you already do. When a supplier tells you delivery will improve and then misses two more dates, you stop weighting the assurance. You are not being unfair. You are being empirical, and so are they.
The asymmetry is what makes it hard to see from where you sit. You have one data point per decision, and it is the one nobody else can access. You know what you meant. You know the pressure you were under, the thing you had just come from, the constraint you could not talk about. Your team has none of that. What they have is the outcome — what you did, in front of them, and what happened afterward.
So they have a larger sample than you do, taken from the outside, of the only variable that predicts anything. Their model of you is built on better evidence than your model of yourself.
That is uncomfortable and it is not an accusation. It is a structural feature of occupying the position.
How do organizations learn?
Through reinforcement, continuously, and without anybody deciding to teach.
The behavioral mechanism here is not novel and Leadership by Design does not claim it is. Behavior that produces a favorable consequence becomes more likely; behavior that produces an unfavorable one becomes less likely. That has been established in the behavioral literature for the better part of a century, and it operates whether or not anyone in the building has heard of it.
What the framework adds is the observation that in an organization, the leader is the consequence. Not the only one, but the largest and the fastest. Approval, attention, escalation, and irritation all function as consequences, and they arrive within seconds of the behavior that produced them, which is the timing that makes reinforcement stick.
There are a few unavoidable rules of reality that come with reinforcement, and each pricks in its own way.
You cannot stop transmitting. There is no neutral state. Silence is a consequence. A delayed reply is a consequence. Choosing not to respond to something is one of the loudest signals available, because it tells the organization that this category of thing does not require a response.
You reinforce most strongly when you are least deliberate. Considered decisions come with explanation attached, which gives the organization context to weigh them against. Your reaction in the four seconds after something unexpected lands comes with no explanation, arrives faster, and is read as more truthful because it was not composed. People believe what you did before you had time to decide.
Consistency beats magnitude. One dramatic response teaches less than the same small response fifteen times. Your organization is not reading your peaks. It is fitting a curve.
What are you broadcasting without knowing it?
None of these appear in a communication plan. All of them are read continuously.
What you interrupt. Not what you disagree with — what you cut off. Interruption is a statement about whose thinking gets the room’s time, and it is registered by everyone present, including the people not being interrupted.
What you follow up on. The organization tracks which of your requests come back. A thing you asked about once, then never again, has been marked as something that can be absorbed. This is the most common source of the phrase nothing ever changes around here, and it is rarely about the request. It is about the follow-up that did not come.
Who gets promoted. Every promotion is a published specification of what this place rewards, and it outranks every values statement in the building. When those two disagree, nobody experiences confusion. They update toward the promotion.
What you spend time on. Your calendar is an allocation of the scarcest resource in the organization, and it is legible to anyone who can see it. Whatever you have told people is a priority, the meeting you moved to make room for something else is the more reliable account.
What you tolerated once. A single unaddressed exception sets a precedent, and the precedent does not require repetition to hold. People do not conclude that you missed it. They conclude that this is a thing that can happen here — and the next person who does it is not breaking a rule, they are following one.
None of these are failures of character. Most of them are the ordinary consequences of a full week. But they are running whether or not you have attention available for them, and they are being read by people whose working lives depend on reading them correctly.
Is this just a communication problem?
No, and treating it as one is the most common expensive mistake in this territory.
The communication interpretation says: the message is not getting through, so send it more clearly, more often, through more channels. That interpretation is what funds the all-hands, the values rollout, the manager cascade, and the internal newsletter.
It fails for a structural reason. Your organization does not have a comprehension problem. It has a credibility calculation. People understood the message. They also observed what followed it, and where the two disagreed they weighted the observation. Saying the message again does not change the weighting, because the weighting was never about the message. Every repetition without a change in consequence makes the gap more visible, not less.
This is why the second culture initiative works less well than the first, and the third less well than the second. Each one is a fresh claim delivered into a system that has already priced your claims against your behavior.
Can you fix this by being more careful?
Not for long — and this is where the framework moves from interesting to paying dividends.
The obvious response to everything above is vigilance: watch your reactions, manage your tone, be deliberate about what you reinforce. That works, and it works right up until the moment it matters.
Deliberate self-management runs on the same system, and uses the same resources, which pressure robs and degrades first. In the quarter that is going badly, in the meeting that went sideways, in the fourth difficult conversation of a day that started at five — that is when your automatic responses are most available and your deliberate override is least. Which is to say: your signal is at its least managed at the moment your organization is watching hardest, because a system under pressure is a system paying close attention to what leadership does next.
A repair that depends on you being at your best cannot hold at your worst. That is not a criticism of your discipline. It is what discipline is made of.
So the repair is not vigilance. It is architecture — reinforcement structures that produce the intended consequence without requiring you to supply it in the moment. What comes back to you and on what cycle. What carries consequence and what does not. Which decisions leave the center and which do not. Built once, running whether or not you have attention available.
You cannot performance-manage your way out of a design problem. You can only design your way out of it.
Where this leaves you
With a different question than the one most leaders are asking.
Not am I communicating well enough. The answer to that is almost always yes, and it has almost never been the constraint.
The question is: what has my organization learned from me, and is it what I intended to teach?
You cannot answer that from the inside on your own, and the reality is that you are the least reliable observer of your own signal — you are clear on your intention and neither observe nor experience any of the evidence. But you can stop spending money on the assumption that the problem is the message.
The message is fine. The gap is somewhere else, and it is somewhere findable.
Signal is phase one of eight. What it produces travels the rest of the arc, and the failures that show up seven phases later are often carrying its distortion.
How to trace a failure back to its source → The diagnostic descent, worked end to end.
