When to Kill a Portfolio Company You Run Solo

Kill a venture when it costs more attention than it returns and has stopped teaching you anything. Not when it loses money for a quarter. Not when it is hard. When the honest math says this thing takes a slice of your best hours and gives back less than the next-best use of those same hours would. For a solo operator, that is the whole calculation, because your attention is the scarce input, not capital.

I run around twenty companies. I have shut several down. Every one hurt, and every one was correct, because killing the weak ones is how the strong ones get enough of me to work.

Why killing is prioritization, not failure

When you run a portfolio alone, every company competes for the same finite resource: you. There is no separate team for the loser. Time spent nursing a flat venture is time not spent on one that is climbing. So a dying company is not just losing on its own terms. It is taxing everything else you own.

This is the opposite of how a funded startup thinks, where a losing bet mostly burns investor cash. In a solo portfolio the currency is your attention, and you cannot raise more of it. That means pruning is not an admission of defeat. It is routine gardening. I decide which venture gets attention today constantly, and part of that discipline is admitting when a venture should get zero attention, permanently.

When should you shut a venture down?

Run three tests.

First, the attention test. If this company vanished tomorrow, would the hours it frees go somewhere clearly better? If yes, that is a loud signal. A venture worth keeping is one you would choose to fund with your time even if you were starting fresh today.

Second, the learning test. Early on, a venture that loses money can still earn its place by teaching you something that compounds into the rest of the portfolio: a market, a technical pattern, a distribution channel. When it stops teaching and just consumes, the subsidy is over.

Third, the drag test. Some companies do not just underperform. They actively pull attention because they break, generate support load, or create a single point of failure you have to babysit. A small revenue line that eats your best mornings is worse than no revenue line at all.

If a venture fails all three, the decision is made. You are just delaying it for emotional reasons.

Why operators hold on too long

Sunk cost is the obvious trap. You built it, so killing it feels like erasing the work. But the work already happened. The hours you spent are gone whether you keep the company or not. The only live question is where the next hours go.

The subtler trap is that a slow-dying venture rarely forces the issue. It does not blow up. It just quietly underperforms, month after month, taking ten percent of you and giving back three. Because there is no crisis, there is no trigger to decide. So it lingers for years, and the real cost is invisible: it is every stronger venture that got a little less of you the whole time.

Set a review cadence so the decision is not left to a bad mood. My weekly review that runs twenty companies is where underperformers surface, and once a quarter I ask the kill question directly for anything flat.

How to kill cleanly

Do it in a way that does not leave a mess pulling at you afterward. Notify customers honestly and early. Offboard data. Cancel the infrastructure. Because I self-host most of the portfolio, winding a company down is mostly deleting an app and a database, not unwinding a web of managed subscriptions, which is one more reason owning your stack makes a solo portfolio easier to prune.

Then reallocate the freed attention deliberately, the same week, before it gets absorbed by noise. The point of killing a company is not to have one fewer thing. It is to give the winners more of you. Tools like Servo Agent can keep the mechanical remainder of a wound-down venture handled so it does not quietly creep back onto your plate. Prune, reallocate, and the portfolio gets stronger every time you subtract.