The fastest teams are not the ones with the best people. They are the ones where the person closest to a problem is allowed to act on it.
That idea comes from Stanley McChrystal, a retired four-star general who once commanded elite counterterrorism forces before advising companies on how they work. He watched capable units lose to a faster enemy, not because they lacked skill, but because their information and their authority sat in the wrong places. What he built to fix it transfers cleanly to any team that has to move quickly in changing conditions.
Getting there is a shift you can actually make, one layer at a time:
- Why a top-down structure quietly caps how fast your team can respond
- Connect separate teams into one network so information travels instantly
- Share far more than feels comfortable, and watch decisions improve
- Push authority down to where the work actually happens
- Build trust between groups that barely speak to each other today
The most useful realisation is this. The rule designed to protect your information is usually the same rule that slows your whole organisation down. Holding a steadier picture of your own state under pressure is the personal version of that same discipline. The free foundation course opens with a first practice for noticing what is going on inside you, a simple place to begin. Clarity at any scale starts with seeing what is actually there.
Why a pyramid slows a fast-moving team
A traditional structure sends decisions up to the top and instructions back down. That worked when the world moved slowly and tomorrow looked like today. It cannot keep pace once conditions change faster than a decision can climb the ladder and return.
McChrystal saw this directly. His task force was more skilled than the network it fought, yet it kept arriving a step late. The structure itself was the bottleneck. A single mind at the top, however brilliant, cannot pre-plan a response to a situation nobody predicted.
The freeing part is that the fix is structural, not personal. You do not need better people or a harder-working leader. You need the authority to sit closer to the problem. You also need a way to keep everyone working from the same picture. A leader who cannot let go of control is a pattern I meet often in my own work. The release usually starts on the inside, well before it ever shows up in a team.
Connect separate teams into one network
The answer is not to merge everyone into one giant group. It is what McChrystal calls a team of teams, where separate and capable units stay distinct but link into a single network. Intelligence gathered by one team becomes instantly available to all of them.
He learned the cost of the opposite the hard way. Different elite units operated in the same theatre without sharing what they knew, and raid intelligence sometimes sat unread in a supply closet for weeks because translators were overloaded. The capability existed. It simply could not be combined, and combination is where the advantage lives.
Picture the difference in a company. When your product team and your sales team have no real connection, customers ask for things that never reach the people who could build them. Scott's Miracle-Gro hit exactly that, with sales unable to deliver products customers wanted because the two functions barely spoke. A network fixes what a stack of separate boxes cannot.
Share more than feels comfortable
Most organisations run on a "need to know" default, where information only reaches people with an identified reason to have it. It sounds prudent. In practice it is the quiet source of the silo problem, because you cannot predict in advance who will need which piece of information.
McChrystal reversed it with a blunt principle. "I want everybody to know everything all the time." His task force ran a daily all-hands call with one rule. The price of admission was contributing real information, including problems and failures, not a tidy status update. Run consistently, that call eventually reached 7,500 people and produced something powerful.
That something is shared consciousness, a common understanding of what is happening and what the group is trying to do, held at every level. Once people carry the same picture, they can self-organise and make good calls without waiting to be told. The meeting is not really about the updates. It is about the shared picture the updates build.
Give people the why, not just the what
Shared information needs a shared story to hold it together. McChrystal uses a one-page narrative that combines an honest account of where things stand with a clear direction of travel. Honesty matters here. Aspirational language with no grounding in reality breeds cynicism rather than commitment.
Alongside the story sits Commander's Intent, which means explaining the purpose behind a task and not only the task itself. When people understand why something matters, they adapt sensibly the moment conditions change, instead of freezing until fresh instructions arrive. The purpose travels with them.
A simple check keeps this honest. The brief-back technique asks a team to repeat what they understood after being briefed, which catches a misunderstanding before any resources are committed to it. It costs a minute and saves a month.
Push decisions down to where the work happens
None of this delivers speed unless authority actually moves. Choke points, where coordination fails and work backs up, are usually a symptom of decisions concentrated at the wrong level rather than a sign of weak people.
McChrystal's remedy is refreshingly concrete. Name the small handful of decisions that genuinely can only be made at the top. A chief executive should keep only three or four, then push everything else down in writing. His rule for how far to go is memorable. "Delegate until you're uncomfortable, and then go one more level." The discomfort is the signal that authority has reached the right place, not a reason to stop.
One retailer captured the spirit of it by letting staff spend up to 3,000 dollars to resolve a customer problem without seeking approval. Given that freedom, employees used judgement rather than abusing it, typically spending far less than the cap when a situation warranted action. Trust, offered clearly, tends to be repaid. Helping someone move past the fear of getting it wrong is often the shift I work through with people one to one. A leader learning to release control is doing that same inner work on a larger stage.
Build trust between teams that barely speak
Silos are not the result of bad intent. They form through habit and the natural pull to talk mainly with people you already know. They are invisible from the inside too, because each silo looks like a healthy team.
You can manufacture the trust that dissolves them rather than waiting years for it to appear. McChrystal embedded selected operators from one elite unit inside a rival unit for four months, which built real personal relationships. Afterwards, when speed mattered, people could call a known human being instead of a faceless department. Remote and cross-functional teams need the same deliberate effort, because functional-only interaction with no personal conversation slowly destroys the trust that fast collaboration depends on.
Keep judgement in charge of your data
A fast team still needs to see threats coming. McChrystal models this on the human immune system, which continuously scans for and neutralises the roughly 15,000 potential hazards a body meets each day. A business version senses market shifts with technology and data, then tests every output against direct experience and common sense before acting on it.
That last step is not optional. In 1983 a Soviet officer, Stanislav Petrov, held back from acting on a false nuclear-attack alert because his own judgement told him the machines were wrong. He was right, and the habit he modelled matters more as automated tools grow louder. Data informs a decision. It should never quietly make it for you.
Two exercises sharpen this. A red team is a group with no stake in a plan, tasked with finding how to defeat it. A pre-mortem assumes the plan has already failed and works backward to explain why, which releases the social pressure that normally stops people voicing real concerns. Both surface risks while they are still cheap to fix.
The character underneath the system
All of this rests on something the structure cannot supply. McChrystal defines character as convictions multiplied by discipline, because a belief with no daily discipline to act on it under pressure produces nothing you can rely on. Character has no external reward. It runs on internal commitment and honest reckoning with what went wrong.
That reckoning includes failure, even the public, career-ending kind. His own guidance is to treat a setback as a starting point rather than an ending. From there you build a team where every person, whatever their rank, is expected to act like a leader. A team that can decide and act without permission is really a team of people who have each been trusted, and who have learned to trust themselves.
If you are the one carrying that weight, you can explore the inner side of leading this way with me in a personal session. If you would rather start on your own, you can ask the chat how to push decisions down to your team. You will get a personalised answer drawn from this source and others in seconds.
