I started StatSheet a month after my first (and only) child was born. It’s been interesting to see the similarities between the two processes over the last 3 years.
Lots of attention required early on
Raising a baby is a full-time job and starting a startup is a full-time job. Both need constant attention, especially early on, because they can’t operate on their own. With a baby you are spending all hours of the day and night care and feeding it.
With a startup, you have to get the infrastructure in place, the technology built, and the operational processes defined. Without all of that, you don’t have a startup. It took a good 6 months to even get to the point where StatSheet had something to show. I had many long nights and early mornings with a baby in one hand (asleep) and the laptop in the other.
Occasionally you wonder what the heck you were thinking
Everyone has a bad day, but when babies and startups have bad days they are really bad. I can remember the first July 4th after my daughter was born as being one of the most challenging days of my life. She was having a really bad time and I spent half the day driving around my neighborhood (car rides soothed her back then) to get her to stop crying. That night my wife had to drive me around the neighborhood to get me to stop crying 
With StatSheet, there have been times when our servers have had a bad day or a script kiddie was trying to download our whole site or a particularly bad bug made its way into production. This has resulted in hundreds of emails being generated from my monitoring scripts, flames on messages boards and some sleepless nights. On more than one occasion I’ve wondered why I chose to inflict this kind of torture on myself.
Fortunately those moments are short-lived…
Small innovations can save time and your sanity
Ever heard of the Wiggles? My daughter loved them for her first two years. If I ever needed a break or needed to tend to my other baby (StatSheet), I could count on the Wiggles for some quick relief. She also loves to draw so I could stick some paper and crayons in front of her and I could get at least 30 minutes.
With StatSheet, I found out early on that since the company consisted of just one person (me), I had to automate everything or I couldn’t sustain it long term. Also, the StatSheet code base was growing so large that it was next to impossible for me to test every permutation of queries that a user could run on the site. Early on I’d get lots and lots of bug reports. It was embarassing. But then I figured out how to make the web server email me anytime a Server Error occurred on the site. Thanks to Google’s constant crawling of StatSheet.com, I’d get notified of 95% of the bugs before a user ever noticed.
Pace of change is dramatic
Baby’s grow and change at a very rapid rate. Every 3 months it’s like you have a new kid. Their personality, likes, and dislikes change about the time you get used to them.
StatSheet has been no different. People like to talk about “Internet time” and how it moves faster than typical cycles. Early stage startups change even faster than that. Every 3 months it was like I was dealing with a brand new company. New opportunities, business partnerships, and technology changes has required that I test different business models and products.
Nothing more satisfying than the pride of a parent
No matter how frustrating parenting can be, there is nothing more satisfying than hearing your baby say something surprising or when they do something you didn’t think they were capable of doing.
StatSheet gives me a similar (though not as emotionally significant) feeling. Accomplishing particularly challenging milestones, getting positive press coverage, and hearing great feedback from users and colleagues is not something that was ever replicated during my time at big companies.
