Read a sample: 1. The Commitments You Cannot Keep Alone
Maya leaves the leadership meeting with what sounds like a clear assignment: launch a new customer onboarding experience by the end of the next month.
Her team has already designed most of the experience. The remaining work appears manageable, and the requested date is far enough away that she feels comfortable saying yes. Other leaders are pleased. The sales team can mention the new experience in upcoming conversations, and customer support can prepare for the change.
Back at her desk, Maya begins turning the commitment into a plan. Within a few conversations, the picture changes.
The onboarding emails need an updated customer list from the data team. The data team can produce it, but needs a definition of which customers should receive the messages. That definition depends on a policy decision Maya thought had already been made. The policy owner wants legal review before confirming it. Meanwhile, the person who normally configures the email system is supporting another launch, and customer support needs time to update its guidance before customers see the new process.
None of these requests is unreasonable. No one is refusing to help. Yet Maya’s confident promise now rests on several pieces of work she did not discuss before agreeing to the date.
This is a hypothetical situation, but its shape is familiar. A commitment can look sound from inside one team and become fragile when viewed across the groups it relies on. The problem is not that Maya failed to predict every detail. It is that she treated the visible portion of the work as if it were the whole commitment.
Leadership intelligence begins with noticing that difference.
A commitment is more than a statement of intent. In a workplace, it creates an expectation that a particular outcome will be available in a particular form, often by a particular time. Other people may plan their work around it. They may tell customers, reserve resources, defer alternatives, or make promises of their own.
The path to that outcome may cross several organizational boundaries. A team may own the final deliverable while relying on another group for information, a specialist for configuration, a manager for approval, and a receiving team to put the result into use. Each connection is a dependency: a relationship in which some part of the work relies on something else being available, decided, completed, or understood.
Dependencies are not defects. They are a normal consequence of specialization. Organizations divide work among people and teams because no individual can provide every skill, decision, and resource. The difficulty comes when those connections remain invisible at the moment a commitment is made.
A dependency can be hidden in plain sight. Maya knew the company had a data team and a legal team. What she did not know was how their work connected to her launch, what they would need from her, or when they could provide it. Knowing that a department exists is different from knowing the conditions under which its contribution will be ready.
The same distinction applies to familiar collaborators. A colleague may have helped with a similar request before, but that does not establish their current availability. An approval may have been straightforward on a previous project, but the present change may raise a different question. A receiving team may be aware of the initiative without being ready to use what is delivered.
The practical question is not simply, “Who is involved?” It is, “What must happen between those involved for this commitment to hold?”
A bottleneck is a point where the flow of work is constrained. It may appear when more work needs to pass through a person, decision, resource, or process than can move through it in the available time. It may also arise when an essential input is missing and everything downstream must wait.
In Maya’s launch, the policy decision is one possible bottleneck. Without a definition of the intended customer group, the data team cannot confidently prepare the list. Without the list, the email configuration cannot be completed and tested. Until the messages and process are settled, customer support cannot finish its guidance.
The constraint is not merely the existence of an approval. The concern is the relationship between that approval and the work waiting behind it. If the decision arrives with enough time for the remaining activities, it may not cause a problem. If it arrives late, several teams may be forced to compress their work at once.
The configuration specialist presents a different kind of constraint. The work may be well defined, but the person with the necessary access and experience is already committed elsewhere. Naming that person in a plan does not create capacity. Someone must confirm whether the work fits alongside existing priorities, whether another person can do it, or whether the schedule or scope needs to change.
These situations can be misread as individual performance problems. The data team is “slow.” Legal is “holding things up.” The specialist is “unavailable.” Such descriptions focus attention on the place where the delay becomes visible rather than the conditions that created it.
A more useful description is specific and neutral: “The data team needs the customer definition before it can prepare the list, and the definition is awaiting a decision.” That statement gives the people involved something they can examine together. It also leaves room for a solution: clarify the decision, narrow the initial audience, sequence the work…