Overnight Success

[John Uhri]

I'm working on a web application in my spare time. Some days that means little programming time time, others more. The hardest part is the times where nothing seems to be moving forward - the grunt work can be depressing. It is easy to look at software companies that have had phenomenal success in the past few years and feel a little jealous.

Or is it?

Jonathan Fields wrote about so-called overnight success. It really isn't anything more than overnight exposure:

[W]ith rare exception, overnight success is really about extremely hard work, coupled with a defining event that allows large numbers of people to become instantly aware of the fruits of years’ or decades’ labor.

37Signals is one of those companies. Their products are loved by their users and were one of the original Web 2.0 darlings. One of the partners, David Heinemeier Hansson, wrote how their success was hardly overnight.

To get today’s levels we’ve relied on the compound interest of attention. Every year a steady stream of new readers and customers have joined the flock while still keeping the bulk from the year before.

I love that term: compound interest of attention. It definitely implies that building something great happens one customer at a time.

Building a software product comes from writing software. Many have chimed in on the difficulty of programming. Jeff Atwood at Coding Horror:

You will literally have to spend multiple years of your life grinding away at this stuff, waking up every day and doing it over and over, practicing and gathering feedback each day to continually get better.

Peter Norvig says it even takes ten years to learn how to program properly, like any skill:

There appear to be no real shortcuts: even Mozart, who was a musical prodigy at age 4, took 13 more years before he began to produce world-class music.

Malcom Gladwell belives it takes 10,000 hours of practice to become successful in a learned endeavor. Building the software with mad programming skills is just as difficult. Joel Spolsky says Good Software Takes 10 Years.

By those numbers, it will take 5 years of full-time coding to become a good programmer with a half-way good piece of software.

Believe it or not, I find that inspiring. I'm in this for the long haul, and can't wait to see where this thing ends up.

And now, back to the code...

[John Uhri]

You make a good point, Matt. I was being a bit facetious, though. I'm clearly aware that things need to happen faster than that... and I'm already a programmer with 13 years of experience. Mostly I was hoping to just get over the fact that it feels like nothing is happening, while, in fact I move forward each day.

I'm building software, but it is the business model around it that will make it successful. However, until I have something to show software-wise, the business doesn't exist.

[Matt Zentz]

John are you trying to become successful by building software or by building a business?

The scary thing with your 5-10 year time frame is how fast things change in that amount of time. From 2001-2002 I built a web-based email platform that rivaled exchange but used for small business. From 2002-2004 I built a hosted e-commerce solution that had really top notch functionality. In 2006, me and two other partners started a hosted web content management system.

In all of those, I was among the early providers, but everyone of those products have become just another choice among hundreds or thousands of like products. Sure, I can sell you all day on our differentiating factors but in the end it's the business mission and its model that drives the business and not the software. The software is just a mechanism to achieve the mission.