Every leadership team believes it is aligned, because everyone in the room nodded at the same slides. Then the work reaches the people who have to do it. Donald Sull, Charles Sull and James Yoder analyzed 124 organizations for MIT Sloan Management Review in 2018 and found that only 28 percent of the executives and middle managers responsible for executing strategy could list three of their company's strategic priorities.
The instinct is to blame communication. The data does not support it. In an earlier study published in 2015, covering nearly 8,000 managers across more than 250 companies, the same research group found that 84 percent of managers could rely on their boss and their direct reports most of the time, while only 9 percent could rely on colleagues in other functions all of the time. Alignment up and down the hierarchy is mostly fine. Alignment across it is where strategy quietly dissolves.
That is the problem sitting in front of most chief executives. Each function is executing its own plan competently. The plans contradict each other at three or four specific points, none of those points is on anyone's agenda, and the contradiction only becomes visible one or two quarters later as a missed date that gets recorded as a delivery issue.
Alignment is not agreement
The word gets used to mean a pleasant meeting. What it has to mean is narrower: that if you asked each member of the leadership team the same four questions separately, in their own words, you would get the same four answers. Not similar answers. The same ones.
That is a testable claim, which is the useful part. Teams do not need to feel aligned, they need to survive the test. And most teams that fail it fail at a specific layer, which tells you what to work on.
The four layer test
Alignment has four layers and they fail in order. A team cannot agree the order of priorities if it cannot agree the list, and it cannot make a commitment if it has not agreed the trade-off. Almost every team that describes itself as aligned has completed layer one, half of layer two, and stopped.
Run the test cold. No preparation, no discussion, no shared document. Each member writes their answers alone, then you read them out without saying who wrote what.
Layer one: the list
Ask each member to write the organization's top five priorities for the next twelve months. Not the values, not the mission. The five things that must be true by this time next year.
Failure here looks like seven different lists with three items in common and a long tail of function specific items that have been promoted to enterprise level by their owners. It is common and it is fixable in a single session, which is why teams stop after fixing it and believe the job is done.
Layer two: the order
Now ask each member to rank their list. This is where the pleasant meeting ends. A team can genuinely share five priorities and still be pulling in different directions, because the second item on the chief financial officer's list is the fifth on the chief revenue officer's, and every resourcing argument for the next year will be a proxy fight about that gap.
Order is not pedantry. It is the only instruction that helps a director choose when two priorities want the same engineer in the same sprint.
Layer three: the trade-off
This is the layer that takes the day, and the one almost no team completes. Ask the question in its concrete form: if we can only have three of these five, which two do we drop, and what specifically stops happening when we drop them?
Teams resist this because dropping a priority has a name attached to it, and that name is sitting at the table. So the list stays at five, everything is a priority, and the choice is pushed down to people with less information and no authority to make it. They then make it in a hundred incompatible ways, which is exactly what the 28 percent figure looks like from the inside.
A team that will not make the trade-off in the room has not delegated the decision. It has abandoned it.
Layer four: the commitment
The last layer is the most mechanical and the most often skipped. For each surviving priority: who owns it, what they have committed to deliver, by when, and what they need from which colleague to do it. The last clause is the one that matters, because it converts a functional plan into a cross functional promise, and cross functional promises are precisely what the research says are unreliable.
Write those promises down with both names on them. A commitment with one name is a target. A commitment with two is an agreement.
Running the test in ninety minutes
Twenty minutes of silent writing on layers one and two. Twenty minutes reading the answers aloud, unattributed, with no discussion. This part is uncomfortable and it should not be rushed, because the discomfort is the evidence. Forty minutes on layer three, which will not finish, and it does not need to. Ten minutes to schedule the session where it does.
Do not run it as an icebreaker at the start of a two day event. It generates the real agenda, which means it belongs a fortnight earlier, in a normal meeting, so the offsite can start at the argument. That sequencing is the whole point of building an offsite agenda that decides things rather than one that fills two days.
What nobody tells you before you buy an alignment session
Three things are worth knowing before the first invoice, because none of them appears in the proposal.
No facilitator can give you the trade-off
An outside practitioner can force the question, hold the room in it and stop the team escaping into process. They cannot decide which priority dies, and any who imply otherwise are selling a decision they are not accountable for. The chief executive makes that call, in the room, in front of the person losing the priority. If they are unwilling to, the session will be pleasant and nothing will change.
Alignment is perishable
It decays with every reorganization, every senior hire and every change in market conditions. A team aligned in March is measurably less aligned in September without anyone doing anything wrong. Treat the test as a recurring instrument rather than a project, and run it twice a year. Teams that treat alignment as a one off event are the ones surprised by its absence.
The session will surface the question that has been avoided
Nine times out of ten, the failure at layer three traces back to a single unresolved question the team has been routing around for months: who owns the customer, whether the two divisions should be one, whether a named executive's remit still makes sense. It will come out. Decide in advance whether you are willing to hear it, because a chief executive who closes that conversation down has taught the team a lesson that outlasts the session.
Read your next leadership meeting differently
The useful change here is not a new framework, it is a different reading of things you already watch. When the same decision returns for the third time, stop treating it as indecision and ask which layer it failed at. When two functions dispute resources, stop hearing a personality clash and hear an unresolved ranking. When a director asks a clarifying question that seems obvious, take it as a report from the place where your priorities collide.
Most of what gets labelled poor execution is a leadership team that never completed layer three, and most of what gets labelled a difficult executive is someone defending a priority the team declined to drop. Read it that way for one month and the pattern is hard to miss. If several of the behaviors in the signs of a dysfunctional leadership team are also present, the alignment gap is a symptom rather than the cause.
When you want the test run by someone with no stake in which priority survives, that is what our leadership team effectiveness work is for. The first session is the four questions, asked cold, with nobody in the room defending a budget.
Frequently Asked Questions (FAQs)
What does leadership team alignment actually mean?
That every member, asked separately and without notes, gives the same answer on the priority list, its order, the trade-offs the team has accepted, and who owes what to whom. It is not shared enthusiasm and it is not consensus. A team can be aligned while two members still think the chosen order is wrong, provided they commit to it outside the room.
How often should a leadership team check its alignment?
Twice a year as a standing rhythm, and again after any reorganization, senior hire or material change in the plan. Alignment decays quietly through ordinary events rather than through conflict, so a calendar trigger works better than waiting for a symptom. The full test takes ninety minutes, which makes the cadence affordable.
Why do we lose alignment after the offsite?
Usually because the offsite produced a list rather than a set of trade-offs. A list of five priorities is compatible with every function continuing exactly as before, so nothing constrains anyone's behavior once they are back at their desk. Alignment that survives the return to work names what the organization is going to stop doing.
Is misalignment a strategy problem or a team problem?
Test it by asking whether the team can state the trade-off. If they can and still pull apart, it is a team and incentive problem. If they cannot, the strategy is genuinely ambiguous and no amount of team work will resolve it, because you are asking people to align on something that has not yet been decided. Fix the ambiguity first.