Integration Multiplies Opportunities and Problems

For founders and senior leaders deciding to take on more: the same move that made Rocket Lab the fastest in the world can also sink it, and the same math runs your life.

By Peter Plötner. Aerospace engineer and Wayfinder Life Coach. More about Peter →

Rocket Lab just agreed to spend 8 billion dollars to own even more of its own supply chain, buying Iridium and its 66-satellite network outright. Whether that turns out brilliant or fatal comes down to one thing most people get wrong about growth.

Ten days earlier, that same instinct, owning the whole chain, let Rocket Lab put a satellite in orbit just 16 hours and 42 minutes after the US Space Force gave the order. A world record, by more than ten hours. Integration is a multiplier. It made them fast because their parts were already tight. Add the same move to something loose, in a company or in a life, and it multiplies the mess just as fast.

This is the part people get backwards about growth. Adding more does not simply add output. It couples to everything you already carry and multiplies whatever state that system is in. I work with technical founders and senior leaders right at the moment they are deciding to take on more: a second venture, a bigger role, a whole new part of the business brought in house. The instinct is always the same, I will add this and it will make me more capable. Sometimes that is exactly right. Sometimes it is the worst possible move.

What actually got compressed

Be precise about what happened, because it is easy to overstate. The Pioneer spacecraft was already built and finished, waiting on standby for the order. What the 16 hours measures is the gap between the call and orbit, the response time, not the build time. The previous record for this kind of responsive launch was about 27 hours, set in 2023 by Firefly Aerospace with a satellite built by Millennium Space, an outside contractor. Rocket Lab beat it by more than ten hours.

The difference they point to is integration. When the call came, their guidance team needed about four hours to calculate the trajectories and update the flight software, because the parts were their own and the gaps between them were already closed. You cannot move that fast if every handoff is a phone call to a different supplier on a different schedule. By owning the whole stack, they had deleted those gaps before the order ever came.

Integration removes the slack, and slack is where errors hide

Here is the engineering truth underneath it, and it cuts both ways. When you bolt many parts into one tight stack, the errors add up. Every part has a tolerance, a little bit of allowed slop. On its own, each one is fine. But stack enough of them and the small errors accumulate, and the finished whole can be out of spec even though every single piece passed its own check. Engineers call this tolerance stacking.

So integration is not automatically good. It removes the slack between parts, and slack is not only waste. Slack is also where errors get absorbed. A loosely coupled system has buffers. If one supplier is bad, you swap them. If one part is off, the gap between it and the next part soaks up some of the error. You lose efficiency, but you gain protection. Vertical integration deletes those buffers on purpose. That is where the speed comes from. It is also exactly why it punishes you when the underlying parts are not tight.

A multiplier has no opinion about the sign

So the launch is not a story about integration being good. It is a story about integration multiplying what was already there. Rocket Lab could pull it off because their specs are under control. Their parts hold their tolerances. The Electron line is reportedly putting out a rocket every eleven days. Into that tight system they added more, and it made them faster.

Then, ten days later, they pushed the same bet much further: the 8 billion dollar deal for Iridium, all the way to owning the constellation and its 2.55 million users. Rocket Lab called it becoming a fully vertically integrated space company. If their processes stay tight, that addition multiplies their value. If they let those processes go loose while swallowing something that big, the same integration that gave them 16 hours will spread problems through the whole company at the same speed. Integration is a multiplier. It has no opinion about the sign of the number it multiplies.

Your life works like that stack

This is the part people get backwards about their own lives. When someone wants to grow, they reach for more. Another project. Another commitment. More surface area. And they assume more input means more output. But your life works like that stack. Adding something does not just add its own value. It couples to everything else you are carrying, and it multiplies whatever state that system is already in.

I learned this the wrong way around, which is to say I learned it from having less time, not more. Before I had children, I had a side project I worked on only when I felt like it. In theory I had loads of time. In practice I spent very little of it on the project, and even that time was loose and unfocused. There was always more time tomorrow, so today did not have to count.

Then the children came, and the time collapsed. On paper that should have killed the project. It did the opposite. With almost no free hours, every hour had to count, so I finally prioritized for real. I cut what did not matter. The work got more focused, not less, because the slack was gone and I could no longer afford to waste the little I had. My coaching practice exists today because my life got tighter, not because it got freer.

We add so we do not have to fix

That is the uncomfortable point, and it is the one most people avoid. We often add more precisely so we do not have to fix what is already loose. A new venture is more exciting than fixing the broken process in the current one. A big new goal feels better than repairing the part of your life you have been ignoring. So we stack something new on top of a foundation we know is shaky, and we call it ambition. But you are just tolerance stacking your own life. Every loose thing you are already carrying is still there, and now the new thing couples to all of it and multiplies the mess.

The honest question is not what to add. It is whether the rest of your system is tight enough that adding anything will multiply up instead of down. Rocket Lab earned the right to integrate by getting their specs under control first. Most people try to integrate before they have done that work, and then wonder why more made everything worse.

Try this week

Do not add anything new this week. Instead, find the loosest thing you are already carrying. Look at your current commitments, work and personal both. Find the one with the most slack in it. The process that keeps slipping. The relationship you keep deferring. The part of your day that is quietly out of spec, that you have been routing around instead of fixing.

Pick the loosest one. And before you try to improve it, ask whether it should exist at all. Often the fastest way to tighten a system is not to fix the weak part. It is to remove it. The best part can be no part. So look hard: does this thing need repairing, or does it need to be gone? Tighten what stays, cut what does not need to be there, and only then let yourself add the new thing.

Because whatever you add next will multiply what is already there. So the real question is this. What are you about to add, and is it going to multiply a tight system or a loose one?

Frequently asked questions

What does “integration only multiplies what is already tight” mean?

Integration removes the slack between parts so the whole thing moves as one. That makes a tight system better and a loose one worse, because the move that deletes waste also deletes the buffers that used to absorb mistakes. It is a multiplier with no opinion about what it multiplies.

How did Rocket Lab launch a satellite in 16 hours?

They owned the whole chain. The satellite was already built and waiting, and with the parts, software, and launch all in house, there were no outside handoffs to slow the response. Their own team needed about four hours to set the trajectory. The 16 hours measures the response, not the build.

Is vertical integration good or bad?

Neither on its own. It trades buffers for speed. Tight parts, and it multiplies your strength. Loose parts, and it spreads the same problem through everything just as fast. The Iridium deal makes that bet much bigger.

What is tolerance stacking, and what does it have to do with my life?

Every part has a little allowed slop. Bolt enough together and the small errors add up, so the whole can be out of spec even though each piece passed its own check. Your life is the same. Every loose thing you already carry is still there when you add something new, and the new thing couples to all of it.

I want to grow. Should I add something new?

First ask whether the rest of your system is tight enough that adding will multiply up, not down. Often the fastest way to get stronger is not to add but to find the loosest thing you already carry and fix it or cut it. The best part can be no part. Then add.

Cutting the loose part before you add a new one is the deletion step of an engineering framework for a life. The framework starts in Stop Optimizing Your Life. Start Specifying It. and the deletion step itself is The Best Part Is No Part. If you want a place to start, the Essential Self Diagnostic is fifteen questions that take about sixty seconds, a quick read on which parts of your life are still carrying their weight.

Get new essays by email

New essays, plus the occasional thing I only send by email.