Getting Ready to Launch

Sponsored by CodeRabbit

Teams using CodeRabbit report 50%+ reduction in code review time. It’s not just automated linting—it understands your codebase, learns your team’s patterns, and provides contextual feedback that actually matters.

SOC2 Type II certified. Zero data retention.

When I was a young engineer, I worked with the team at Tektronix about to launch the LT-1000 semiconductor tester. There was way more to do than there was time, but every morning all 100 engineers would gather in a big conference room, stand in a circle, & report status.

“What a waste of time!” thought young me, “why are we yapping when we’re so far behind?” Ah, youth.

Now that launching is cheaper than ever, I’ve been talking to a slew of teams launching new products. I’ve been reminded of that long-ago launch & the wisdom embedded in the practices I participated in without appreciating.

Pre-launch is its own game of engineering, with its own constraints, its own incentives, & its own tradeoffs. Here’s what I know about the engineering of launching well.

Count Down

How many calendar days to launch? Everybody should know this number without being told.

Don’t start the countdown too soon. You aren’t going to be working sustainably pre-launch. Start at 30? Too soon. Start at 5? Too late. You want to capture the pre-launch energy & direct it profitably.

You’re Fine

If you had to launch with what you have today, you’d be fine. (If you really wouldn’t be fine, then you’re launching too soon.) By the time you start the countdown the product should be usable. Bugs are fine, expected. Some slowness is fine, expected. (Know how you scale though & have a plan to throttle usage.)

It’s so easy to panic & say, “If we don’t have this ambitious feature, the launch will fail.” Nah. It’s not that important. Know that everyone will be thinking this about everything they do. And they will all be exaggerating. That’s just what a launch countdown does to engineers.

Less Upside, More Downside

The fundamental change of constraints during launch countdown is that there is way more downside from mistakes than there is upside from new features. You’ll have many new stresses on the system, less time to validate changes, & greater cost from errors than ever before.

The rest of the advice here supports avoiding costly errors while still getting the most important changes made.

You’ll ...