Speed Without Direction Is like burning Jet Fuel on a Runway

Martin Eriksson darüber, warum KI strategische Klarheit dringlicher macht. The Decision Stack, Prinzipien die etwas kosten, und was 30 Jahre Produktarbeit wirklich lehren.

Table of Contents

Martin Eriksson has spent 30 years watching product teams spin. One of the founders of Mind the Product, author of The Decision Stack, advisor to nearly 150 portfolio companies. The pace of building keeps accelerating. The strategic problem underneath does not change.

We hosted Martin live on LinkedIn. Here are the moments worth keeping.

Learn product strategy that actually sticks.

Eight weeks. A cohort of product leaders. Hands-on from day one.

See cohort dates →

Speed is not the constraint anymore

For most of the past decade, product teams had a building problem. Shipping fast was hard. Getting something in front of users took months.

AI is removing that. And Martin argues this makes the strategic question more urgent, not less. When anyone can build anything fast, the question of what to build is the only place differentiation still lives.

He opens with a metaphor.

“Speed is a little bit like setting jet fuel on fire on a runway. It has no direction, it’s just pure unbounded energy, right? It’s spectacular, you might be able to warm your hands for a little bit, but you’re not actually going to get anywhere.”

Speed is fire. Velocity is direction. Momentum is the whole organization moving together. Most product teams are optimizing for speed.

* * *

The layer most organizations are missing

Martin has a direct diagnosis. Research from Harvard Business Review shows 95% of employees do not know their organization’s strategy. He has seen this pattern across the companies he has advised. The problem is rarely that a strategy does not exist. The problem is that nobody outside the leadership room has enough context to understand it.

“I actually think the biggest challenge as a leader is probably actually communicating the strategy you already have, right? Making that clear, repeating it, making it consistent for the rest of the organization.”

The Decision Stack is his answer to the connectivity problem: five questions that link vision to daily decisions, and back up again. Where are we going? How are we getting there? What matters right now? What actions are we taking? How do we choose between them?

The stack is not a new framework. It is the conversation most leadership teams stopped having after the last reorg.

The Decision Stack by Martin Eriksson

Strategy, objectives, principles. The full framework in one book.

Get the book →

* * *

A principle is a decision that never has to be made again

Strategy defines direction. Principles define how to make decisions inside that direction. Martin argues most organizations confuse the two. They post values that everyone agrees with but nobody uses to make actual trade-offs.

He traces the distinction back to his time at Monster in 1999.

“Hang on, job seekers always come first. Because if we build an amazing experience for them, then recruiters have to follow. … We just knew in any trade-off, job seekers always came first.”

Months of recurring debate ended in one sentence. A real principle is not good over bad. It is good over good, a trade-off between two things your organization actually cares about, pre-decided.

* * *

When everyone builds at machine speed, direction is everything

Martin wrote The Decision Stack before the current wave of AI tools. The book barely mentions AI. Deliberately, so it does not date. But in every conversation since it came out, the same theme keeps surfacing.

The chief transformation officer at Miro put it plainly while reading an early draft.

“Because agents need even more strategic clarity and directions and principles and trade-offs, because otherwise they’re gonna run in the wrong direction, not at human speed, but at machine speed, right?”

An organization without a clear decision stack has always paid a price in wasted effort and misaligned teams. That price is now compounding faster.

* * *