In 2014 Keith Rabois gave a Stanford lecture on how to operate a startup. It boils down a CEO’s daily activities based on 9 points that he or she should be focused on. While the principles laid out are great for current and upcoming founders or CEOs, they apply to team management in general and if you want to create a high-performing team these principles help you to do that.


It’s a natural tendency for every system to increase in complexity over time. Teams add more rules, processes, tools, and projects. This needs to be avoided. Your job is to look out for unnecessary complexity and reduce it. Simplify through omitting what is not that important and deleting no longer relevant documentation, tools, and processes.

As a general rule:

“The more you simplify the better people will perform”


A big part of your job is to ask questions. Often you’re not the expert on topics and issues, but with expertise comes tunnel vision and you need to uncover the knowledge gaps that people missed. Check if everyone is aligned and has a common understanding of the goals and the vision. See if you can bring light to the aspects that are still ambiguous and set a clear path for the team despite the unknown.

Allocate Ressources

There are always more things we want to do than things we can do. With limited resources, we are all investors. You decide where the team places its limited resources and where to place its bets. You need to consider trade-offs, chances of success, and expected value for your decisions.

Ensure Consistent Voice

Rabois is talking mostly about a consistent tone of voice here. The idea is that on every communication channel the company has one voice and is consistent in its communication. However, the principle can be applied even broader.

Align people not only on a consistent tone of voice but also on a consistent way of doing things. Define principles of conduct and instill a culture according to which the company operates. This allows each individual to act and decide autonomously in a way that fits the overall company direction and strategy.


You can’t run everything yourself so of course, delegation is a core aspect. Rabois offers to pointers:

First, delegate according to task-relevant maturity: This means that the more experienced a person is in doing a task, the more leeway and freedom you can give him in doing it. This is fairly straightforward but always worth a reminder.

The second approach is a bit more sophisticated. Map your decisions based on your level of conviction that the decision is the right one and on the magnitude of the consequences of the decision. If both these factors are high you might want to watch the execution closely and potentially step in and override the decision of the team. If both are low you should fully delegate to the team.

Delegation Matrix

Edit the Team

The team is only as strong as its members. It is your job to “edit” the team. On the one hand, you need to find the most capable people. This alone is a difficult task. But you also need to give your people challenging work. If the work is too easy good people get demotivated and bored. Not only are they performing below their level you might also lose them after some time.

Rabois' suggestion is to give people the most challenging work they can handle. That way you max out your team’s performance and each member has the opportunity to grow. He suggests to just increase the complexity of the task until the member fails. Then you know where the limit is and can subsequently provide her with the best tasks according to her skill level.

Insist on Focus

Peter Thiel, investor and one of the founders of PayPal, apparently refused to talk with employees about anything else than their top priority. This meant that if someone came to him to talk about another project or idea, he would shut them down and leave the conversation. It’s an extreme case but illustrates the principle perfectly.

Insist on focus because people have a natural tendency to work on too many things. We’re all bad at multitasking and there is ample literature on the importance of reducing work in progress to increase output.

Also insist on focus because people will do anything but the most important tasks. We procrastinate on challenging and difficult tasks.

“Most people will solve problems that they understand how to solve (…) they will solve B+ problems instead of A+ problems.”

Metrics & Transparency

First, measure what’s important. Then make sure that everyone knows about it. Metrics are only as useful as people look at them and base their decisions on them. Keep track of the corresponding indicators as well to not get surprised by a countereffect. For example, when you track the number of hires you should also track the quality of new hires. Otherwise, it’s easy to fall into actions that favor one metric over the other.

Rabois is big on transparency and thinks that practices like public meeting notes and glass doors for meeting rooms are helpful to calm people’s curiosity. We don’t like being kept out of the loop, so we want to sit in every meeting even if it’s not that relevant. By making meeting notes public there is no FOMO and meetings have fewer participants and are more efficient.

Office Space & Tools

Lastly, following the idea of “a startup as a cult”, Rabois says that you need your own space – no co-working. I’m not sure about the idea of structuring your team or company like a cult, but I do think his second point is important: Provide your team with the best tools to do their job. It sounds obvious but it’s often the first thing cost savings are realized. You can be the best craftsman in the world, if your tools are shitty you won’t produce good work.