Why we install operators instead of running companies ourselves

Every venture studio hits the same wall at roughly the same point. You build something, it starts working, and then you have to decide whether to keep running it yourself.

Keeping it is the obvious choice in the moment. You know the product better than anyone. You have the context, the customer relationships, the sense of what to do next. Handing it to someone else means watching them make calls you would have made differently, and paying them to do it.

We hand it over anyway. Here is the reasoning, and the parts we have had to learn the hard way.

A studio that never hands over is not a studio

The clearest way to see this is to look at what happens if you keep everything.

Say you run three ventures yourself. Each needs a real decision from you several times a week. Your attention is now split three ways, and it does not split cleanly, because switching between three businesses costs more than a third of your time each. In practice you become the bottleneck on all three at once.

Now add a fourth. The fourth does not get a quarter of your attention. It gets whatever is left after the three that are already on fire, which is close to nothing.

This is the ceiling, and it arrives early. A studio that keeps the driver's seat caps every venture at the founder's personal bandwidth, and that bandwidth was fully committed before venture two existed. You have not built a portfolio of companies. You have built a set of departments inside one overloaded company and given them names that make them sound independent.

The handoff is what removes the ceiling. It is the only thing that does.

What the operator actually owns

The word "operator" gets used loosely, so it is worth being specific.

The operator owns the profit and loss. That is the real line. They decide what the company charges, who it hires, what it builds next, and what it stops doing. They live with those decisions, including the bad ones, and they hold meaningful equity so the outcomes land on them as well as on us.

What the studio keeps is narrower than people assume. We stay involved on strategy, on capital, and on the shared systems every venture inherits. We are a useful phone call and an informed board. We are not an approval step for a pricing change.

The distinction matters because a half handoff is worse than either extreme. If the operator has the title but every real decision routes back through the studio, you have kept the bottleneck and added a salary. Worse, you have hired someone senior enough to notice, and they will leave.

The failure mode is hiring late

The mistake we see most often, and the one we have made, is starting the search when the seat is already empty.

The search takes months when it goes well. The best operators are not browsing job boards, because the best operators are busy running something. Reaching them means a relationship that predates the opening, which means meeting people when you have nothing to offer them.

When you start late, every week of drift raises the pressure to fill the role fast, and pressure is the wrong condition for this decision. You start optimizing for availability instead of fit. You talk yourself into someone. Then you spend a year either managing around the mismatch or unwinding it, and the venture spends that year not compounding.

The fix is unglamorous. Keep meeting people continuously, including when there is no role. Treat the bench as something you build before you need it rather than a search you run when you do.

What we look for

Four things, in rough order of how much they predict.

Ownership instinct. The tell is how someone talks about a past failure. People who reach for context, the market moved, leadership would not fund it, tend to keep reaching for it. People who can name the specific call they got wrong, without theatrics, tend to keep doing that too.

Comfort with an unfinished thing. A venture at handoff is working but rough. Someone who needs a clean org chart and a settled roadmap will spend six months trying to create one instead of selling.

Judgment about what not to do. Most early companies die of doing too many things adequately. The operators who work out can hold three priorities and genuinely drop the fourth, rather than promising to do all four.

Willingness to be measured. The P&L is the score. Someone who wants the title but flinches at the scoreboard is telling you something.

Domain expertise is deliberately not on that list. It helps, and we take it when we can get it, but it is the most learnable item here and the one people most often overweight.

The handoff itself

The transition is a period, not a day. What has worked:

The operator arrives while the studio is still running things and takes over in pieces, starting where the cost of a wrong call is lowest and moving toward where it is highest. Pricing and hiring go last, because mistakes there are expensive and slow to reverse.

We write down what we know rather than transferring it in conversation. Every assumption we are carrying about the customer, the pricing, the competitive position gets written out plainly, including the ones we are unsure about, flagged as unsure. Verbal handoffs lose exactly the caveats that turn out to matter.

And then we stop. This is the part that takes discipline. There is a window after handoff where the studio can still see everything and is still tempted to intervene, and intervening once resets the whole thing. The operator learns that decisions are provisional and starts checking. From there you are back to being the bottleneck, except now you are also paying for a CEO.

What we still get wrong

Two things, honestly.

We have been slow on the bench more than once, which is the failure mode described above, diagnosed correctly and then repeated anyway. Knowing the search takes six months does not make you start it six months early unless you have built the habit of meeting people with no role to offer.

And we have been imprecise about what "involved on strategy" means at the boundary. It is easy to say the operator owns the P&L and then send a strong opinion about pricing that lands as an instruction. We are better at this than we were, mostly by being explicit in the moment about whether something is a view or a decision.

Neither is solved. Both stay a practice rather than becoming a policy.

For the wider context, the venture studio playbook covers the operating system, the economics, and the structures around all of this.