The Gap Is Not Below You. It Is at Your Table.

In June, Harvard Business Review published a piece by Irina Wolpert called The Two-Organizations Problem. Her argument is that every company past a certain size runs as two organizations at once. The reported organization lives in dashboards, board decks, and town halls. The lived organization is what the work actually feels like on a Tuesday. Read it. She is right about the problem.

I have been making a version of the same argument for years, in different words. Every leadership team runs two operating systems. The visible one holds the strategy and the action list. The hidden one holds the unspoken agreements, the real priorities, and the things nobody will say. When the two diverge, the hidden one wins. What follows is not a disagreement. It is what I would add from inside the room.

Start with the sharpest thing in her piece, which is an observation I had not made and wish I had. The reported organization used to leak. The rough first draft. The hallway update. The half-finished deck someone forwarded by accident. Those leaks were how uncomfortable truth reached the top, not by design but by friction. AI has removed the friction. The board deck is now drafted by an assistant working from a prompt written by someone who already knows what the executive team wants to hear. What arrives is more coherent and more confident than any version that came before it. Whether it is more accurate is a question almost nobody in the room is asking. That is the last leak closing, and it deserves to be taken seriously.

Where I part company is on what kind of problem this is. The article treats it as an information problem. Layers summarize, incentives shape, long-tenured leaders get shown a curated version of their own company. So the fix is better signal. Drop into meetings unannounced. Ask sharper questions. Run a parallel instrument to measure the gap. All of that assumes somebody does not know.

In the rooms I work in, everybody knows. The team can tell you who really decides, whose behaviour gets managed around instead of addressed, and which subject ends the conversation the moment it comes up. They will tell me in a private call in under twenty minutes. Then they will sit in a room together for a full day and not say any of it.

It is not a shortage of information. It is a shortage of permission.

There is a moment in the article where a new CEO finds three functions where the reported and the operational truth have come badly apart, and the point is made that none of the leaders involved had done anything wrong. The structure had simply produced its predictable output. That sentence is why the piece is publishable. It is also why it will not change anything.

Structures do not defend themselves. People defend them. Every persistent gap has beneficiaries, people whose position, influence, or protection depends on nothing being named. Entrenchment is not passivity. It is active, and it has a payroll. Ask a leadership team why change is so hard here and you will get a list of external factors. Ask them these two instead.

The room goes quiet, because those questions have answers and several people already know them.

Then there is the direction of the gap. The article locates it vertically. Boardroom at the top, lived reality six or seven layers down, distortion accumulating in between. Executive teams like this framing. It puts the problem below them. It casts the leadership team as the audience for the distortion rather than a producer of it. The gap I find is horizontal. It sits between people at the same table, at the same level, who have worked together for years and have arrived at a stable arrangement about what will not be discussed. No layers required. Four people is enough.

Which is why the remedies cannot reach it. Every fix in the article is administered by the CEO. Drop into the operational review. Ask concrete questions. Follow up visibly. But the CEO is the power variable. Asking the person at the top of the structure to run the diagnostic on the structure is the oldest failure in this work. Nobody surfaces a system they are part of, and the more senior you are, the more of that system you are.

The second remedy is a recurring question put to a rotating sample of managers, asking them to name one thing their team believes that the executive team does not. Aggregate the answers and read them seriously. That is a survey. Two pages earlier the same article explains that engagement surveys are unreliable because they are shaped by the environment they are trying to measure. So is this one. People who have learned what gets rewarded do not become candid because the question got better.

So what does move it. Not more accurate reporting. The reported organization is not going anywhere and it should not. What moves it is a room where the people who already know say it out loud in front of each other, in a structure that makes it survivable, held by someone with no stake in what gets protected. That is uncomfortable, slow, and unpublishable as a tidy set of practices. In my experience it is also the only thing that works.