We had just finished it. Weeks of work: market analysis, leadership alignment, a strategy everyone could stand behind. It was signed off and on its way down the organization, from leadership to teams and from teams to individual goals.

Shortly after, Epic Games launched its store with an 88/12 revenue split and started writing EPIC checks to secure distribution rights.

The strategy was outdated the moment that news hit.

It wasn’t wrong. It was simply written for a market that no longer existed. The economics of PC distribution shifted overnight, and the pipeline of games we were counting on was suddenly being bid on by someone with far deeper pockets.

What came next was worse than a broken plan. It was a void.

Continuing the cascade made no sense: why turn a dead strategy into team goals? But putting together a new one took time. So our teams went through a strange sequence. First they saw the presentation. Then they watched the ground shift under it. Then, for weeks, they got nothing.

They weren’t idle because they lacked skill or commitment. They were idle because we had given them a plan and nothing to steer by when the plan broke. If they had held a clear mission, they could have rebuilt sensible plans themselves, adjusting to the new reality while leadership regrouped. We had handed them a fish that rotted fast, instead of a fishing rod.

That’s when I understood that the problem wasn’t our strategy. It was what we passed down with it.

The problem isn’t speed — it’s what the cascade strips out

The usual diagnosis is that strategy cycles are too slow. Plan quarterly instead of yearly, the argument goes, and you’ll keep up.

That misses the point. The cascade doesn’t just delay the strategy. It hollows it out.

Watch what happens at each layer. The leadership team knows why: the market bet, the trade-offs, the risks we chose to accept. The next level down receives what: priorities, targets, budgets. The level below that receives tasks. By the time the strategy reaches the people closest to customers, partners and product, the reasoning is gone. What’s left is a list.

A list works fine as long as the world holds still. When it doesn’t, a person holding only tasks has two options. Execute a plan they can see no longer makes sense, or escalate - if you’re lucky! Both are slow. Both teach the organization the same lesson: thinking is someone else’s job.

And here is the irony. Cascading assumes the top knows best. But in a fast-moving market, the edges see change first. When Epic started writing checks, the people talking to developers and publishers every day felt it before any board slide did. They had the information. What they didn’t have was the intent that would let them act on it. So they waited, and the organization stopped.

That’s the hidden cost of the cascade. Not that the plan dies — every plan dies. It’s that when it does, nobody below the top knows what to do next.

A mission is a fishing rod: it lets people catch the next plan themselves, and gives the HQ the time needed to regroup.

The three gaps

None of this is new. Armies figured it out 150 years ago.

In the 19th century, the Prussian general staff under Helmuth von Moltke faced the same problem at a far higher cost: detailed battle plans that fell apart on contact with the enemy. Moltke’s conclusion is usually paraphrased as no plan survives first contact with the enemy. His answer was not to stop planning. It was to change what a plan contains.

Stephen Bungay, in The Art of Action, translates Moltke’s thinking into business terms. He describes three gaps that open between any plan and reality:

  • The knowledge gap — between what we’d like to know and what we actually know. We never have full information.
  • The alignment gap — between what we want people to do and what they actually do. Instructions never land exactly as intended. And even when they do, people quietly resist anything that feels like a personal loss, e.g. automating part of their own job.
  • The effects gap — between what we expect our actions to achieve and what they actually achieve. Customers never respond exactly as modeled.

The instinctive response to each gap is more of the same. More analysis to close the knowledge gap. More detailed instructions to close the alignment gap. More control to close the effects gap.

Every one of those responses makes things worse. More detail means more brittleness: the more precisely a plan specifies action, the more ways it can be invalidated by reality. More control means slower reactions and less ownership. You end up with exactly what we had after Epic: a detailed plan that died on contact with reality, and people who had nothing to replace it with.

The lesson isn’t to plan less. It’s to plan differently.

What management by mission actually is

Management by mission (mission command, as the military calls it, or Auftragstaktik in its original Prussian form) rests on a simple idea: tell people what you want to achieve and why, not how to achieve it. Then trust them to figure out the how, and to change it when reality demands.

In practice, it comes down to four habits.

Intent over instructions. A mission states the purpose, the desired end state, and why it matters. Not “sign 500 titles this year”, but: “Your mission is to secure titles our audience will love, so every deal is a win for both us and the developer. We aim to grow new-catalog sales by 20% this year while reducing the capital we put into licenses. You’ll need to balance onboarding capacity, the funds available for upfront payments, the revenue potential and the risk profile of each game.” The first goal dies when the deal market changes. The second tells you what to do next.

Boundaries over steps. Freedom without limits is chaos. A good mission defines what you can’t do, which resources you have, and which decisions stay at the top. Inside those boundaries, the how belongs to the team.

Backbriefing. Intent only works if it lands. So instead of asking “is that clear?”, leaders ask people to play back what they understood and what they plan to do. Misalignment shows up in a conversation, not in a quarterly review three months later.

Two levels up. Every leader should understand not only their boss’s intent but their boss’s boss’s. That’s what lets them adapt in the right direction when the plan breaks — because they know what the plan was ultimately for.

  Cascade Mission
What travels down Targets and tasks Purpose, end state, boundaries
Where decisions sit At the top As close to the information as possible
When reality shifts Wait for a new plan Adjust the plan, keep the intent
What gets measured Adherence to plan Progress toward intent
How it fails The void Drift, when intent is unclear

Management by mission doesn’t remove leadership. It changes its job: from writing the plan to making sure everyone knows what the plan is for, and from having the answers to asking the hard questions and equipping people to find their own.

Every plan rots. The Epic moment wasn’t bad luck. It was a normal day in a market that moves faster than any planning cycle. The question isn’t whether your strategy will be overtaken by reality. It’s what your people will have in their hands when it is.

A leader’s job isn’t to produce a plan that survives. It’s to produce a purpose that does.

If it works, why do so few companies manage it?