As Okta scales, I’ve realized a big part of being a successful CEO is influencing the company culture, which means your day-to-day actions have to be consistent with what you want to see from the team. That includes the way you communicate—if you want a fast-moving culture, for example, you need to be clear when you ask for deliverables that you don’t want a six-month research project. You have to say, “Come back to me with three choices and then we’ll talk about details. How’s tomorrow?”

Communication also includes your everyday behavior. Are you nice to people? Are you respectful? We think so much about the big things—the all-hands meetings, the major announcements, the user conferences—but that’s not really how you influence culture. It’s the small, daily moments. It’s steady progress over a long period of time. How you act day in and day out is what moves the needle.

It’s going to take a long time, and that’s okay. Even the so-called overnight successes took a while to build. There’s an old saying, “The journey of a thousand miles begins with one step.” I always say, “The journey of a thousand miles begins with someone who has no idea how long a thousand miles actually is.” If you did, you’d probably never start.

The hardest part for me was continuing to execute when I didn’t know how long something would take or whether it would even work. I had to get comfortable with doubt and uncertainty, because I’d never fully experienced it. I’d worked for good companies that were constantly growing, but I hadn’t been the CEO. It wasn’t all on me. With Okta, I had to learn that it’s okay to not have everything figured out and keep going anyway.

My wife and I decided early on that we would always have dinner as a family. Our daughter was six months old when I started Okta, and our son was born a year later. I think my natural inertia would keep me at work until 11 p.m., and the commitment to eat together has a nice governing effect. I’m sure you always have regrets as your kids grow up, but I don’t think I’m ever going to regret being around for dinner every night.

I wish I knew what was over-hyped and under-hyped—what will actually make us healthy and happy. Or, from a business perspective, which technologies are going to exceed expectations and which are going to underperform. It’s kind of like buying stocks. If you find a great company and everyone else thinks it’s great, too, that doesn’t help you. The same is true of a weak company that everybody knows is weak. The real opportunities come when you realize something’s good or bad before everyone else.

When I started Okta, for example, a lot of people thought cloud computing was over-hyped. I think it turned out to be under-hyped, if anything. Right now, the one I really wish I knew is blockchain. I know cryptocurrency is over-hyped, but I’m not sure about blockchain itself. It’s tough because there’s always a reason for the hype, right? It’s never total BS. But if you can separate the signal from the noise, that’s a huge chance to differentiate yourself, your company and the tools you use.

I’ve been reading Thinking Fast and Slow, by Daniel Kahneman, which is about instinctual versus calculated thought processes. It’s been out there for several years and I’ve read a lot of articles and blogs that were influenced by it, but it’s nice to hear it cleanly articulated, from the source.

I do like reading industry stuff. I just finished Losing the Signal, by Jacquie McNish and Sean Silcoff, which is about the rise and fall of BlackBerry. I also read How Google Works, by Eric Schmidt and Jonathan Rosenberg. There’s a lot of good stuff in there about how technically focused the company is and how they empower people to make bottom-up decisions. I do wonder how much of that is the way Google actually works versus the way they want Google to work, but it’s influenced my thinking regardless.

I almost always read nonfiction, but I did read a good novel recently called City of Thieves, by David Benioff. It’s set in Russia, during the Siege of Leningrad in World War II. It’s the complete opposite of my world, but it’s so well written that it makes you feel what it was like to be in that place, during that time.

How much time do you have? I’m wrong all the time. I think what’s most important is timing—when do you decide you are wrong? You should be difficult to convince. Otherwise, you won’t stick with the controversial decisions even when they’re right. But you don’t want to be too stubborn.

In the early days of Okta, we thought all we had to do was build a good SMB product that was easy to use, and it would take off. But we tried that for about a year. It didn’t work. So we shifted our focus and went upmarket. We had some flexibility in that case, because we knew that if we did go in another direction, we wouldn’t have to throw it all out and start over. We had some time to become completely sure we were wrong before we made the call. But if we’d needed to do a total 180, that would have required shifting gears much sooner.

I’ll start with what doesn’t matter most, which to me is the medium term. I think there’s a lot of pressure to think at least a couple of years out, especially after you go public. But I don’t believe a three- to five-year plan is particularly useful. Our concrete product plans are usually only a few months long.

The super-long term and super-short term are far more important. You need to know your ultimate destination—the vision you’re heading toward—and you need to check yourself against that in terms of what you’re doing today, tomorrow and next week. Does it serve that long-term vision? If not, you should be doing something else.