Work can slow between capable teams even when each function is meeting its own commitments. A completed task may reach the next team without the information, capacity or authority needed to continue. Asking everyone to work faster leaves that dependency unresolved.
The leadership task is to identify what the receiving team is waiting for and make the appropriate decision. An incomplete submission needs better information. An overloaded specialist needs a capacity or sequencing choice. A disputed customer promise needs someone authorized to settle it. These problems can look alike in a status report, but treating them alike creates more coordination without necessarily improving delivery.
Follow one item to the point where it stops
OutcomesLab uses Execution drag to describe the loss of speed and effectiveness as work crosses boundaries. The useful unit of investigation is an actual item moving toward a result: an order, a customer request or a product launch. Follow it beyond the point where a function marks its contribution complete.
Consider a hypothetical launch. A late technical test reveals that a feature needs an exception for a particular use. Product can explain the limitation, but it cannot approve new customer language. Legal needs to understand the proposed promise before reviewing it. Operations needs the final specification to prepare support, while sales has already discussed a delivery date with customers.
Every team can report legitimate activity. The launch nevertheless depends on a choice that none can make alone: change the promise, narrow the release or accept a defined exception. More frequent progress updates will not supply that choice. They may simply make the unresolved disagreement more visible.
Start by locating the next decision and the person authorized to make it. Then establish what that person needs. If the technical consequence is unclear, further investigation is necessary. If it is already understood but leaders disagree about the commercial trade-off, another technical report postpones the judgment rather than improves it.
Distinguish the cause from the queue
Specialization gives an organization expertise and independent checks. It also separates information and authority. Wouter Dessein and Tano Santos model the trade-offs between specialization, adaptation to local information and communication. Advance planning can coordinate familiar work, but new information can expose a choice that the plan did not settle.
In the launch, the late test changes what can responsibly be promised. A useful handoff therefore needs more than a revised specification: it needs an account of the consequence for customers and the alternatives still available. Sending the document without that context may cause each receiving function to ask a different version of the same question.
This can be an expression of alignment debt. If leaders have not resolved how speed, customization and risk should be weighed, a launch team may discover the disagreement only after commitments are made. Better execution requires settling the choice and making it usable in the work. An instruction to “own your piece” cannot reconcile promises that depend on one another.
OutcomesLab’s Strategic friction directs attention to the operating conditions resisting a priority; execution drag concerns the delay and rework as work encounters those conditions. The distinction prevents a visible queue from becoming an automatic diagnosis. A reviewer may be waiting for usable evidence, working through excessive demand or holding a decision no one has authorized. Each calls for a different intervention.
A meta-analysis of team studies distinguishes task dependence from shared outcomes. That is a useful distinction here: agreeing that the launch matters does not provide the inputs or authority needed to complete it. Leaders need to examine the dependency itself, rather than infer from a common target that coordination is in place.
Use the Handoff Friction Map to choose an intervention
The proposed OutcomesLab Handoff Friction Map examines four features of a boundary: goal, information, authority and timing. Use records and conversations with the people who handled the item to establish what happened. The map should produce a decision about the work, not four separate improvement programs.
- Goal: establish what each function was trying to protect. If sales is preserving the announced date while operations is protecting readiness, identify who can decide how those commitments fit together.
- Information: inspect what the receiver actually obtained. A complete form may still omit the reason for an exception or the consequence of refusing it. Improve that input before adding another approval.
- Authority: distinguish responsibility for tracking the launch from permission to change its promise. Give the unresolved trade-off a decision owner while preserving specialist authority over required safeguards.
- Timing: separate necessary review from time waiting for input, capacity or a decision. Earlier notice may let operations prepare safely; starting before the specification is stable may instead create rework.
Compare several instances before concluding that the interface needs redesign. If the same missing fact repeatedly sends work backward, change the submission. If usable submissions wait at an overloaded reviewer, decide what work takes precedence or what qualified capacity can be added. If the question keeps returning because no one can authorize a compromise, clarify authority. A generic collaboration initiative can miss all three.
In the hypothetical launch, leaders might narrow the first release and revise the customer promise. Product can then supply a stable specification, legal can assess the actual language and operations can prepare the supported offer. The improvement comes from a settled dependency, not from requiring each function to accelerate regardless of what it receives.
Protect the handoffs that improve the result
The goal is not to eliminate every wait. Independent review can catch an error that an integrated team misses, and separation of duties can protect consequential decisions. Where the stakes warrant scrutiny, make the evidence, authority and sequence clear before work reaches the reviewer. Removing the check to improve cycle time could damage the outcome it protects.
More contact is not automatically the answer either. A small 2025 study of hospital workgroups found no statistically significant overall improvement in relational coordination after huddles and cross-role shadowing in that setting. That does not show that those practices never help. It cautions against adding meetings without identifying which dependency the interaction will resolve.
Three existing management perspectives sharpen the diagnosis. Execution: The Discipline of Getting Things Done, by Larry Bossidy, Ram Charan and Charles Burck, connects people, strategy and operations. Eliyahu M. Goldratt and Jeff Cox’s The Goal: Excellence in Manufacturing directs attention to the limiting point in the flow. In The Fifth Discipline: The Art and Practice of the Learning Organization, Peter M. Senge examines patterns arising from interrelated actions. Together, these perspectives invite leaders to examine whether the launch has the support it needs, where completion is constrained and which choices reproduce the delay.
After changing one interface, follow the whole item again. Did the receiving team get what it needed? Did action begin sooner without losing necessary scrutiny? Did the delay disappear or move to another queue? A faster transfer is useful only if it helps deliver the intended result.
When work repeatedly stalls between functions, leadership should make the missing agreement explicit: what is being promised, what evidence is required and who can resolve the exception. Teams can then be held accountable for delivering within a workable arrangement, rather than for negotiating the same unresolved choice every time work changes hands.