Amazon EC2 went down, taking Heroku down with it. That was May. Developer’s Mac freaked out. That was Tuesday.

Brought in a UI coder for some quick cleanup. Rails config nightmare on Windows, and 6 hours later he was just getting ready to dig in. That was Wednesday. It was a rough week.

Things go wrong. So what are you going to do?

In the first case, we just toughed it out and developed a backup plan for the future. We haven’t fully executed on that, which makes me nervous, but with few resources you do what you can when you can.

In the second case, he was going to have to wait up to 3 days to get his notebook back, so I had him buy a new Mac mini and work on that. I couldn’t afford to lose him for 3 days or longer, and this got him up and running within hours.

Now it cost $700 to do that, but we’ll be able to flip it online for maybe $650. That $50 saved lost development time, and getting the app out is more important.

In the third case, that was simply a series of bad choices that 4 otherwise smart people allowed to happen. We put that work on hold and I did it myself over the weekend. There’s a cost to everything, but that cost sanity and weekend time.

When things go wrong, adjust. Be creative. Question how you’re doing things. Examine your values. What matters? How could you do it differently?

We’re making this up as we go. That’s how startups are–you plan, but people talk about planning as though it’s some pre-defined thing that everybody knows. You’re making it up. And when things don’t go as planned–and they rarely do–you have to adjust.

Plans simply show how far off your original intent was from the reality–they are in essence historical documents.

The reality is a fluid set of adjustments and motion toward your ultimate goals.