I’ve worked at three successful start-ups and one failure. I’ve also worked at post-IPO successes such as Borland, Netscape, and Apple, which means I’ve seen a lot of different founders who, if you measure success financially, were quite successful.

My backstory aside and with deep respect, most founders fail. You’ve heard of the stories of sucessful founder because they’ve become famous (or infamous). However, the majority of start-ups fail. No one tells and retells the stories of these companies because they never launch. No one became rich or famous. It is their defining characteristic. In his recent essay, Paul Graham talks about the successful founders. However, it’s not “Founder Mode,” it’s “Successful Founder Mode.” Lumping all Founders together would mean we should — statistically and more descriptively — call this “Failing Founder Mode,” which is neither clever nor inspirational.

As a person deeply in love with naming things, I like the framing of Founder and Manager Mode because it’s clever and instantly useful. If you’ve been reading me over the years, you’ve noted I’ve begun to detest the term manager for some of the reasons Graham highlights: unfamiliar with the details, management at a distance, lousy hiring, and siloed decision-making. I’ve gravitated towards the word leader both because I want to make it clear any motivated human can execute the skills of a good manager — leadership comes from everywhere — and, more importantly, I believe managers tell you where you are. Leaders tell you where you are going. It’s a philosophy thing.

It’s less clever and symmetric, but I would define these two modes as: “Founder Mode” and “Scale Mode” because one of the many things you need to do as a successful start-up is scale. Landing managers is one of those tactics, but it’s just one essential investment you make as you scale. Other tactics are equally important. Focusing on managers as the thing that makes you a larger successful company gets you… more managers. Gross.

My observation from three successful start-ups, three post-IPO starts where the founders still ran the show, and one failure: the founders were all eccentric humans. In a hypothetical room full of humans, which included these founders, you would eventually notice their behavior or conversational pattern. You would single them out in your head because the hair on the back of your neck stood up and wonder, “What… is going on there?” My one start-up failure? The founders? By the book, MBA types. Standard vanilla leadership. Trust me, you’ve never heard of them or the start-up.

Graham hints at some of the attributes of Founder Mode but mostly says it’s not as well-defined. It is. Founder Mode is the culture of a company, and a culture is defined by the character of the founders. Here are the values I’ve discovered over and over again working with these humans:

• The whole team is involved in the details. Anyone can argue about the product because everyone works to be a product expert.
• Everyone does the work. The stratification of responsibility is a red flag not just in rapidly growing team, but any company. Everyone files bugs because everyone uses the product. Yes, there is job specialization, but there is also a belief that we are equally accountable for the product.
• An organization chart doesn’t tell you who can speak with whom; it tells you who is accountable for what. It’s a map. Not a power structure.

And finally, hire leaders, not managers.