Tomorrow, Sifter turns five. During those five years, I’d say that all of my mistakes and regrets boil down to fear. Sifter supports my family, provides some income to good friends who do some contract work for us, and has thousands of people that use it to run their business. So almost every meaningful decision I’ve made has been haunted by how it might affect these people.

I’ve been afraid to make any significant design changes because it inevitably creates some pain for some of our customers as they adjust to the changes. The reality is that long-term, our customers are usually happier, and if they’re not, we can always fix it or make changes. Big changes usually bring a handful of emails from concerned customers, but after chatting with them, they’ve almost all been happy with the changes.

I’ve been afraid to let a customer wait more than an hour for a response to a support request. After really messing up my sleep, I decided to stop waking up in the middle of the night unless the servers were on fire. Nobody noticed, and people were just as happy to get a response the “same day”.

I was afraid of what would happen if I got sick or injured and couldn’t work for months. Guess what? It happened. In the past four and a half months, I’ve averaged a quarter of my normal hours and even spent three weeks in the hospital. Everything chugged along just fine. Granted, I had some part-time help who all really stepped up, but the world didn’t end. A few bugs lingered for a couple of days instead of a couple of hours, but I’m the only person who noticed.

I was afraid to pull in more outside help because of the effort to get them up to speed and our inability to guarantee huge long-term contracts. We now have 5 folks who all contribute from time-to-time as their schedule permits, and they all have specialties where they run circles around me. I’m quickly learning that in most cases, their wisdom and experience far outweighs the time that it takes to bring them into the loop.

I was afraid that if we didn’t add certain features, that customers would begin leaving in droves. When we didn’t add those features, nothing changed. When we did add the feature, nothing changed. I have a ton of historical data, and while improving the product is important, individual features just don’t move the needle.

When I was writing, I was afraid I wasn’t spending enough time developing. When I was developing, I was afraid I wasn’t spending time writing or designing. When I was designing, I was afraid that I wasn’t spending enough time writing or developing. The end result was that I constantly jumped from task to task trying to do a little bit of everything instead of sitting down and pouring myself into a specific task and just working on that.

I was fearful about customers trusting us with their data and credit cards. At one point, we even lost several hours of customer data. If you asked me at the time, I would have bet you that we’d be out of business the next week. I felt terrible and expected all of our customers to immediately lose faith in us, but even in this worst case scenario, things kept on trucking. We explained the mistake honestly and openly, gave customers some free months, spent some time talking with them, and then poured a whole bunch of effort into improving our infrastructure. It doesn’t get much worse than that, but we survived, and even became stronger for it.

These days, the fear is still there, but I’m better at reminding myself that it’s likely trivial. I’ve found it helps to ask, “what’s the worst thing that can happen?” Then, I’ll break that down and think about how we’d respond if that thing happened. 9 times out of 10, the fear is overblown and unlikely. Even if it is likely, it’s rarely worth being afraid of. The worst outcome is usually little more than a few extra hours of work answering emails or making changes based on customer feedback. When you look at it that way, even the worst case really isn’t that bad.