When things go wrong during a launch

A few days ago we were working on an exciting launch we’d been planning for weeks. ~1h before release - having shared links in a few places - our site went down: an incident similar to the one everyone’s been experiencing this morning

Engage incident response mode ??

The irony being that I work for an incident management platform (incident.io) and the launch was related to a guide on how to deal with this exact situation:

For a non-engineer, incidents are both panic-inducing (will we get the site back up in time for the launch?) as well as paralysing (I don’t want to ask the engineers who are working on a fix for for updates - the priority is the problem at hand).

What followed was a masterclass in how to deal with incidents.

The team used @incident to:

1. Initiate an incident channel

2. Escalate/identify the best person to lead the incident

3. Discuss potential hypotheses/solutions

4. Assign actions that led to a fix

All without leaving Slack ??

All along, the response team kept everyone else up to date on what was happening. This was a serious case of dog-fooding both the product, but also the guide we launched.

This weekend we hit the front page of hackernews and received a tonne of interest in the guide.

If you think either the guide or the product could be beneficial to you or anyone in your network, please share away. On to the next incident ??

PS: In case you thought this wasn’t real ??