Decision Velocity: How Fast Does Your Team Decide?

Ask a manager how fast their team moves and you will hear about story points, cycle time, or deploy frequency. Ask how long the same team takes to make a decision and you usually get a shrug. Almost nobody tracks it, yet every sprint and launch waits on it. Decision velocity is the rate at which a team turns open questions into committed, executed choices, and it is often the slowest loop in the system.

What Decision Velocity Means

Decision velocity has two parts. The first is speed: the elapsed time between the moment a question becomes live and the moment someone with authority records an answer, with an owner and a date. The second is durability: whether that answer holds. A team that decides in a day and reopens the call three times a month is slow.

Bain's framework for decision effectiveness captures both parts. It scores decisions on four elements (quality, speed, yield, and effort), where yield means the decision was carried out as decided. Bain finds that organizations strong across all four dimensions "win more contracts, get to market faster and otherwise beat out rivals" (Bain).

Why Speed and Quality Travel Together

The evidence links faster decisions with better ones. In a 2018 survey of 1,259 respondents, McKinsey found that faster decisions tend to be higher quality (McKinsey). Only 20% of respondents worked at what McKinsey called winning organizations, the ones that make high-quality decisions fast and execute them quickly. Respondents at those organizations were twice as likely to say their recent decisions returned at least 20%. These are self-reported, correlational results, so read them as a consistent pattern that stops short of proving cause.

Bain lands in the same place. Its survey of almost 800 companies found a 95% correlation between excelling at making and executing decisions and top-tier financial results, whether measured by revenue growth, return on capital, or total shareholder return (Bain).

Slow decisions also cost a lot of time. McKinsey's respondents said managers spend an average of 37% of their time making decisions, and that more than half of that time (58%) is used ineffectively. McKinsey then modeled the cost at an average Fortune 500 company: more than 530,000 days of lost manager time and about $250 million in wasted labor costs per year, based on the 2017 US median salary for management occupations of $102,590 (McKinsey). It is a modeled estimate, and the order of magnitude still stands out.

The strongest academic support comes from Kathleen Eisenhardt's 1989 study of top management teams at eight microcomputer firms (Academy of Management Journal). The fast deciders used more real-time information than the slow ones, developed more alternatives through a two-tiered advice process, and resolved conflict actively rather than letting it stall the decision. Their fast decisions led to stronger performance, while the slow firms did poorly or failed. The study covers one fast-moving industry, but the lesson travels: fast teams did more analysis, ran it in parallel, and didn't let disagreement sit unresolved.

Match the Speed to the Door

Jeff Bezos gave managers the most practical tool for calibrating speed. In his 2015 shareholder letter, he split decisions into two kinds. Type 1 decisions are "consequential and irreversible or nearly irreversible," one-way doors that deserve slow, careful deliberation. Type 2 decisions are "changeable, reversible," two-way doors that "can and should be made quickly by high judgment individuals or small groups."

The failure mode he described: growing organizations drift toward using the heavyweight Type 1 process on most decisions, including many Type 2 ones, and the result is "slowness, unthoughtful risk aversion, failure to experiment sufficiently, and consequently diminished invention." In the 2016 letter, he added a rule of thumb: most decisions should probably be made with around 70% of the information you wish you had, and waiting for 90% usually means you are being slow. It is an untested heuristic, and a useful gut check.

On most teams the Type 1 drift has a face: the manager. When every call routes through one person, that person becomes the bottleneck, and the team learns to wait. The manager ends up babysitting two-way-door choices the team could make on its own.

How to Measure Decision Velocity

No published benchmark tells you what a good number looks like for your team, so treat these as a practitioner's toolkit and compare your team against itself over time.

• Time to decision. Days from "decision raised" (a ticket, doc, or agenda item) to "decision recorded" with an owner and a date.
• Reopen rate. The share of recorded decisions revisited or reversed within a few weeks without any new information. A high reopen rate points to failed commitment.
• Decisions per meeting. Explicit decisions, each with an owner and a next step, per recurring meeting. Microsoft's 2023 Work Trend Index found that 55% of workers cite unclear next steps after meetings and 56% find it hard to summarize outcomes (Microsoft).
• Latency by type. Track Type 1 and Type 2 decisions separately. A slow Type 2 cycle is the exact problem Bezos described.
• Yield. The share of decisions carried out as decided.

A simple decision log makes all of this possible: the decision, the owner, Type 1 or 2, date raised, date decided, date revisited, and a date to check the outcome. Pair the metrics with each other, since time to decision alone is easy to game by recording vague decisions. When one decision keeps taking three meetings, the three-meeting decision problem traces it to four missing structures: an owner, a pre-read, a slot for dissent, and a decision record.

What Slows Decisions Down

Unclear decision ownership. When nobody knows who owns the call, every stakeholder behaves like an approver. Bain's RAPID® model (Recommend, Agree, Perform, Input, Decide) exists to fix this, and its core idea is a single "D": one person holds the decision, and veto power stays separate from advisory input (Bain). Naming the owner before the discussion starts gets you most of the benefit. The five team decision modes give you vocabulary for the rest.

Late disagreement. The most expensive dissent is the kind that shows up after the decision. Someone who stayed silent in the meeting raises the concern in a thread two days later, and the decision reopens. Bezos's "disagree and commit" works only when disagreement gets voiced first, which is why his 2016 letter pairs it with recognizing true misalignment early and escalating it immediately. If your team tends to swallow objections, it helps to know how each person handles conflict; the free conflict style quiz takes a few minutes.

Meetings that end without an outcome. In the same Microsoft survey, inefficient meetings ranked as the top productivity obstacle (Microsoft). Microsoft's 2025 telemetry adds that 57% of meetings are ad hoc calls with no calendar invite (Microsoft). A decision made on a call that half the team never heard about is easy to reopen.

Cross-functional handoffs. A single roadmap call can need design, engineering, and sales. The patterns in how product teams make decisions without getting stuck apply to any team with several stakeholders per call.

Practicing Decisions Under a Clock

A decision log records the outcomes of your team's habits. The habits themselves (who speaks up, who defers, how long the group circles before someone commits) surface under time pressure, and teams rarely get low-stakes practice deciding together on a clock.

QuestWorks, a team intelligence platform, is one way to get that practice. The whole team plays a 25-minute quest once a week, seated into groups of 3 to 6, and each quest puts the group under pressure to decide together. The weekly Team Intelligence™ Score (0-100, with an 8-week trend) includes named dimensions such as Decision Velocity, Role Clarity, and Launch Readiness, plus a "Do this week" list of concrete actions. Participation is voluntary and never tied to performance reviews.

The tradeoffs: it measures how your team decides in play, which approximates work without replacing it, and it costs 25 minutes of everyone's week. It won't replace a decision log either. It runs on its own platform and works with Slack or Microsoft Teams, at $199 per month per team, with a 30-day free trial covering four game days, no credit card required.

Start With One Week of Decisions

For one week, write down every decision your team makes, who made it, how long it took, and whether it was reversible. At the end of the week, look for two things: two-way-door decisions that waited on one person, and decisions that got reopened. Those two lists show where your decision velocity leaks.