The People You'd Replace

When a founder says, with real frustration, that they cannot rely on their team, there is now a fashionable answer waiting for them. Replace the people. Software does not argue, does not resign, does not need managing, does not have off days or difficult mornings. Swap the humans for agents, the argument goes, and the whole problem of unreliable people simply dissolves.

It does not dissolve. It transforms into a different problem, one you will feel for far longer and regret more deeply.

The moment you start replacing people with machines because the people are inconvenient, you begin quietly dismantling the two things that are hardest to build and easiest to lose, which are the culture and the humanity of the business. Almost no founder set out to create a company with no one in it. They built something they wanted to belong to, populated by people they wanted in the room, and a great deal of what makes the business good, the trust, the shared history, the arguments that produce better answers, the sense that it is a place where humans do meaningful work together, lives entirely in those people. Trade that for frictionless output and you gain efficiency and lose the soul of the thing, and the loss keeps costing you long after the efficiency has become invisible and unremarkable.

There is a further point, and it is more uncomfortable. An unreliable team is very often a built thing rather than a hiring misfortune. Teams become unreliable when ownership was never really handed over, when context stayed locked in the founder's head, when initiative was corrected often enough that people learned to stop showing it. In a large number of the cases I have seen, the unreliability the founder is describing is the predictable output of how they themselves have led. Replacing the people does not address that. It simply removes the witnesses to it, and rebuilds the same pattern with tools.

Reliability, where it genuinely exists, is almost always something a leader constructed on purpose. It comes from handing someone a whole area rather than a string of tasks, from giving them the context you carry in your head, from agreeing plainly what good looks like, and then from the hard discipline of holding back while they do it their way, including through the early stretch when their way is rougher than yours would have been. That holding back is the actual work of building a team you can lean on, and it is uncomfortable precisely because it asks you to tolerate things being done less well than you could do them, in the service of a team that will eventually do them without you. No agent removes the need for that. It is the job.

I want to be careful here, because I am not making a romantic case against technology, and I would be a hypocrite if I did. I build these systems and I use them every day. Used consciously, they are extraordinary, and the right use of them is to make good people markedly more effective, to lift the repetitive load off human beings so they can do the work that actually needs a human. That is a very different act from removing the people because leading them is hard. One strengthens the culture you built. The other spends it.

This is the heart of what I would put against the prevailing advice of the moment. The current enthusiasm is to strip the humans out and scale the founder, and it treats people as friction to be engineered away. I think that gets the whole enterprise backwards. The businesses worth building, and frankly the ones that will prove more durable, are the ones with people still in them, led well. That is a leadership problem, not a tooling problem, and leadership is learned in the seat. A consultancy will not teach it to you from a deck, and most coaches have never held the seat to begin with. I have spent three decades in every chair a leader occupies, and the thing I keep returning to is that reliance is built into people, not bought by removing them. The Entrepreneur Gap scan will show you how you are operating across the eight roles a business runs on, the people role among them, and where your own defaults may be producing the very unreliability you want to escape. It takes about five minutes and it is free.

Continue reading: A Bigger Machine Won't Fill It →

Or find your blind spots: Take the free 5-minute Entrepreneur Gap scan →