#30 - Launch Fast. Roll Out Slow.

So my friends and I finally launched our app on iOS and Android: Wednesday Waffles.

I expected the road to get here to feel like a marathon.

It felt more like a Spartan Race.

We were constantly coming up against obstacles that challenged us in new ways and slowed us down tremendously. We ended every sprint tired, sweaty, covered in a mud and mildly disappointed that we didn’t cover more ground. And oh look, there’s another obstacle ahead. Terrific.

But hey, we launched an app in 3 months. We have 1,303 users. And last someone checked, we were sitting at #45 on the ‘Social Networking’ charts on the iOS App Store.

Let’s pat ourselves on the back.

Though I fought my team on it in the early stages, I have to admit that launching fast was a great idea. The rate of learning is faster. The pressure to improve is greater. And the thrill of seeing strangers use something you built is intoxicating.

But here’s my balanced recommendation to any product builders out there: launch fast, but roll out slow. By that I mean ramp up your user base slowly. Don’t “open the flood gates” to your entire wait list.

We had intended to launch our app last week, but had to push it back a week due to the App Store rejecting the version we submitted due to a couple of minor nits.

Thank god for that.

We rolled out that version to our small batch of external testers instead and the app had an unanticipated bug that made it completely unusable.

We then launched this past Tuesday to give users a day to get setup on the app in advance of Wednesday — the high-activity day for our app each week.

Thank god for that too.

Here are some of the things that immediately went wrong on Tuesday:

• Users’ data wasn’t syncing with the remote database due to another uncaught bug.

• Users couldn’t create accounts or login because we kept running out of Twilio credits faster than we were being auto-charged to replenish them.

• Speaking of Twilio, people in Norway couldn’t login because support for that geo-location was disabled by default.

• We got 429 and 544 rate-limiting errors in our database due to the number of people using the app all at once.

• Parsing of users’ phone numbers didn’t work as intended for certain country codes, so the app was unusable for people in Australia.

We patched all those holes as quick as we could and by the grace of god the app held together on Wednesday. But we really lucked out. If we had launched when we wanted to and fired all our sales & marketing bullets at once, we would have wasted a lot of great work that our Go To Market team has done over the past 3 months.

So launch fast, but roll out slow.

• Start with external testers that you know personally and will be patient with your bugs.

• Ramp up slowly, and see what bottlenecks you hit with your third-party services and back-end infrastructure.

• Spread out to different regions methodically. See how different folks are affected by query latency and security features of your third-party services.

…Or you could just stand in front of the thundering herd with your fingers crossed and pray your app holds up. Either way, you’re going to learn! And that’s all that really matters.

Quick reminder - If you appreciate my writing, please reply to this email or “add to address book”. These positive signals help my emails land in your inbox.

If you don't want these emails, you can unsubscribe below. If you were sent this email and want more, you can subscribe here.

See you next week — Rayhan