Speed is not the same as velocity
A faster recommendation is of little value if approval remains uncertain or teams cannot act on the choice. Improving decision velocity means finding where useful deliberation ends and avoidable waiting begins, while protecting the scrutiny that the consequences require.
In a hypothetical team, a recommendation arrives early but encounters an undisclosed approval and another request for analysis. While it waits, staff keep several plans alive. Shorter meetings have accelerated preparation without resolving who can commit the organization. The decision path needs attention before the team receives another demand for speed.
Decision velocity is the ability to make an appropriately supported choice, communicate it and act on it without avoidable delay or repeated reopening. It is not the number of decisions made per week. A quick approval that leaves execution contested has moved the decision into another forum, not resolved it.
In “The Effective Decision,” Peter F. Drucker argues that effective executives concentrate on important decisions rather than valuing speed for its own sake. The leadership challenge is to protect judgment while removing waits that add little information. That takes more than an instruction to be decisive: authority, evidence, escalation and review have to work together.
Where decisions lose time
A choice can slow before anyone formally objects. The people preparing it do not know who can commit resources, so they consult broadly for protection. Information arrives in stages because no one has defined what would be sufficient. A dependency surfaces late, sending the issue to a senior leader who has not seen the trade-offs. After approval, someone challenges an assumption that was never made visible.
Each step may look prudent in isolation. Together, they create a serial process: ask, wait, revise, escalate and begin again. Pressure to hurry can make matters worse if teams shorten challenge instead of removing waiting time. In experiments using constrained choice tasks, Charley M. Wu and colleagues found that time pressure changed how participants explored uncertainty: they explored less, directed their search toward value less effectively and repeated choices more often. These tasks cannot predict what will happen in a particular organization. They do challenge the assumption that a tighter deadline alone preserves the same quality of inquiry.
The tension between participation and authority appears in the management literature too. A review of Andrew S. Grove's High Output Management highlights the involvement of people with relevant knowledge alongside those with positional authority. Liz Wiseman's work on Multipliers emphasizes rigorous debate before a choice and ownership afterward; the original Multipliers: How the Best Leaders Make Everyone Smarter was coauthored with Greg McKeown. The practical implication is to bring challenge into the decision while there is still time to change it, then make responsibility for action clear.
The Decision Velocity Protocol
The proposed OutcomesLab Decision Velocity Protocol connects ownership, sufficient evidence, escalation and review. Use it to locate the step that could improve a judgment or make action possible, rather than give every choice a more elaborate process.
Name the owner. Frame the decision as a choice between actionable alternatives and identify one person authorized to make it within a defined scope. Name whose knowledge must inform it and who will carry it out. Make any genuine approval rights explicit at the start, including those arising from legal obligations, safety responsibilities or resource commitments. Bain & Company's RAPID framework distinguishes recommending, providing input, agreeing, deciding and performing. Its caution about unnecessary formality also matters: explicit roles are useful when they resolve costly ambiguity.
Define sufficient evidence. Before requesting another analysis, ask what finding could change the choice. Identify the criteria that matter, the assumptions still in doubt and the consequences of getting them wrong. Sufficiency depends on the stakes. A reversible change may warrant a bounded test; a difficult-to-reverse commitment may need stronger evidence and independent challenge. This is not permission to rely on instinct when useful information is available. It is a reason to stop requesting information that cannot affect the decision.
Set the escalation boundary. Specify which conditions exceed the owner’s authority: a risk threshold, a cross-functional dependency, an unresolved conflict between objectives or a commitment outside an agreed budget. Identify who receives an escalation and the question they must resolve. Without that boundary, every uncomfortable choice can move upward. With an unrealistically rigid boundary, serious risks can stay hidden. Escalation should resolve the exception, not restart the entire discussion.
Schedule the review loop. Record the choice, its rationale, its important assumptions, who will act and when the result will be reviewed. Agree what would warrant correction or reopening. A review is not an invitation to revisit every settled preference. It asks whether action occurred, whether conditions changed and what the organization learned. Examine the reasoning available at the time as well as the result: a defensible decision under uncertainty can still have a poor outcome.
These arrangements depend on one another. An owner needs a stable evidence threshold, known approval rights and a route for exceptions. Reviewing the result then shows whether those boundaries need to change.
Match the process to the consequences
In Amazon's 2016 shareholder letter, Jeffrey P. Bezos distinguishes decisions that are hard to reverse from those that can be changed after action. That is a useful starting point, provided leaders also ask how large the possible loss is, who bears it and whether harm could occur before a correction takes effect. A reversible choice can still have serious consequences.
Consider a hypothetical operations team deciding whether to test a new scheduling method in one location for four weeks. A local leader could own the choice if service and staffing safeguards are met. The team could establish a baseline, hear from affected staff, set conditions for stopping the test and review results at the end. Requiring enterprise-wide approval for that bounded experiment might delay the evidence needed to judge it.
Replacing the scheduling system across every location is different. Contractual commitments, implementation risks and potential disruption would justify broader scrutiny and a stronger case. The four protocol questions still apply; the authority, evidence and review arrangements change with the consequences. Calling both choices “scheduling decisions” should not give them identical approval paths.
Give consequential decisions enough time
A demand for speed can silence necessary dissent. A leader may call an issue reversible when someone else will absorb the harm. A fixed deadline may favor the most confident voice or conceal an unresolved safety concern. The process must make room for those objections before approval.
For consequential choices, seek relevant expertise, identify who bears the downside and document assumptions that would invalidate the case. Preserve independent approval where it is required. Time spent on genuine challenge can prevent a contested approval, a failed handoff or a costly reversal. Protect scrutiny that improves judgment while removing handoffs that obscure responsibility.
The protocol has a boundary of its own. If every routine choice requires a charter, four meetings and a formal retrospective, decision architecture becomes another queue. Use the lightest version that makes ownership and limits clear. Reserve fuller documentation for recurring, cross-functional or consequential choices.
Turn decisions into a capability
Start with one recurring decision that takes too long despite adequate information. Trace it from the first request to the start of action. Where did it wait? Which request for evidence changed the recommendation? Who had authority in practice, rather than only on paper? What caused the choice to reopen?
After changing the path, examine elapsed time alongside decisions reopened, action taken and risks surfaced. A shorter cycle is useful only if the decision remains workable. If a necessary voice was excluded or an unresolved risk went quiet, revise the process before extending it.
Decision velocity could become a compounding capability if each review improves an organization’s understanding of which information matters, where authority belongs and when escalation is worth the cost. That learning must change the next decision process; otherwise the organization repeats its delays with more experience.
A useful question for the executive team is: Which decisions need deeper judgment, and what prevents the others from becoming timely, durable action? Answering that question gives the team something more useful than a general demand to decide faster: a specific decision process it can improve.