Startup Company Culture

Company culture is one of the most important things that determines whether a startup succeeds or fails. For my startups I decided to think ahead of time about the kind of culture that I wanted, write it down, and then work to bring it about by hiring the right people and continually encouraging employees in the right direction.

There are dozens of attributes that are desirable for a company culture, but I decided that the odds of actually implementing the culture would be best if I picked a small, focused set of ideas that would be easy for people to remember. I chose to focus on how people in the company should work together and how to build a culture of continuous improvement. When you start a company you know almost nothing: you have a vague idea for something that might someday become a product that maybe some people will buy. The key to success is to very quickly figure out the answers to a lot of questions and get a lot smarter on numerous topics than you were when you started the company. Most of the culture rules below follow naturally once you accept that you know almost nothing and must learn fast.

I should also point out that it takes a lot more than just writing down some ideas to achieve a good company culture. Lots of companies have wonderful-sounding culture documents that bear no relationship to the way the company actually operates. It is hard to put a culture in place in a new company, and even harder to maintain it as the company grows and new employees join. It takes continuous discussion and reminders, and above all the company leadership must "walk the walk"; even the tiniest bit of hypocrisy in the leadership will be detected by employees and doom the culture.

Here is the culture document for Electric Cloud:

1. It's OK to make mistakes

It is in the nature of any startup to take risks. In order to be successful we must move very quickly and we often have to make decisions with limited information. Thus we expect to make some mistakes along the way. It's OK to make mistakes: if you never make any mistakes, then you probably aren't being aggressive enough.

However, when you make mistakes it's important to recognize them quickly, learn from the mistakes, and correct them. It's not OK to keep making the same mistake over and over, and it's not OK to deny the existence of mistakes: this will turn a small problem into a big one. Thus you should constantly be on the lookout for opportunities to correct mistakes, both yours and other people's. Approach each task with the following philosophy: "Much of what I'm doing is wrong; I just don't know why yet". Your job is to find out why as quickly as possible and fix it.

2. We want to know about problems

In order to fix mistakes, we first have to know about them. The longer it takes to recognize a problem, the more difficult it will be to fix. Thus we want to know about problems, and the sooner the better. You are doing people a favor by telling them about problems and potential mistakes.

At the same time, it's important to remember that the real goal is to fix problems, not just to talk about them. It's good to point out a problem, but it's even better to find solutions. So, once a problem is known, stop complaining and move on to the solution. Above all, be constructive and non-personal in discussing problems. Avoid placing blame; this leads to a negative atmosphere where people are afraid to take risks.

When it comes to discussing problems, we distinguish between external communication and internal communication. Though we will be honest and forthright with our customers, we will shade our discussions more carefully when talking outside the company. Internally, on the other hand, brutal honesty is the right approach. You should not hesitate to offer constructive feedback to anyone in the company. You should also be prepared to receive feedback at any time from anyone; if you can't deal constructively with criticism then this is the wrong company for you. As CEO, I know that I make mistakes, and I depend on the rest of you to tell me about them.

3. Success = execution + consultation

One of the most important ingredients for working well as a team is to strike the right balance between execution and consultation. On the one hand, everyone in the company is expected to execute well: once you have agreed to carry out a task, we expect you to be totally reliable and complete the task without reminders or prodding. On the other hand, there are very few tasks that you will be able to carry out entirely by yourself; almost always, you will need input from others along the way. Furthermore, there will often be unexpected twists that change the nature of the task or affect your ability to complete it as promised. It's important that you know when to consult with others to get input. It's also important to let us know immediately about any risks to the successful completion of the task so we can work with you to revise the goals.

I've found that the people I most enjoy working with, and the people I trust the most, are those who have mastered the balance between execution and consultation. If I'm not sure that you can execute reliably, or if I'm not sure that you will come for help when you need it, then I have to micro-manage to be sure that the task gets done. However, if I can be totally confident that you will do you what you have said, and that you will let me know whenever problems arise or you need help, then there's no need for me to check up on you constantly; it's more fun for both of us.

One of the things that drives me crazy is what I call the "silent drop": this is when someone agrees to a particular task and then fails to complete it with no warning: the deadline date comes and passes without any communication from the responsible person. When asked about this, the dropper will usually have an excuse, and often it is a valid reason, but it is essential to speak up about the problems well before the deadline date: if you don't tell us about an issue you take away our options to help you or revise the plan. If you do silent drops you are asking to be micro-managed.

4. Decision-making: get lots of input, but move forward quickly

The smartest decisions get made when you combine the best ideas of many different people. For each decision you make, identify the people that can contribute to that decision and consult them appropriately. The more important the decision, the more people you will need to consult. On the other hand, we need to move quickly. Thus it's important to find efficient ways to collect input and agree on decisions. In general I think people tend to err on the side of getting too little input and being too autocratic or hierarchical in the decision process.

This topic is complicated enough that I have written a separate article on how to make decisions. Some of the key ideas from that article are that consensus-based processes actually works surprisingly well, and that it is possible to get input from a large number of people in an efficient way.

5. The Reasonable Person Principle

In spite of all our best efforts, disconnects and misunderstandings will still occur. The Reasonable Person Principle is an approach for working through them. When a disconnect occurs, it often seems that the other person is behaving unreasonably ("I can't believe X could have done Y!"). The Reasonable Person Principle says that you should always assume that other people are behaving reasonably, even if it doesn't appear that way. Rather than attacking them, ask questions to find out why they did what they did. Almost always, you'll discover that there is a reasonable explanation. This way, you can defuse the situation quickly and move forward constructively.

A corollary of the Reasonable Person Principle says that if someone truly is unreasonable, and behaves that way repeatedly, then they must leave. That way, we end up with a company where virtually everyone is reasonable, so it is even easier to apply the Reasonable Person Principle.


The overall goal of these rules is to create a highly iterative culture where we move quickly but make constant corrections and improvements. We also want a culture where everyone has an opportunity to contribute on any topic and in any way that they can. And finally, we want a constructive culture where criticism is viewed as a good thing and where we can work through problems and disconnects in a positive, non-personal fashion. The result will be an environment where we have lots of fun, learn and improve, and produce terrific results.