Mark Pearl
  • Fixed hiring standards (or a hiring approach, e.g. test-driving)
  • Stop-the-line and fix the root cause of production issues
  • No flaky tests (tests that routinely fail, but are ignored)
  • Maximum limits for “unplanned” work
  • Ability to push changes into production in [some time duration]
  • Everyone does time in the support queue
  • Quarterly, well-facilitated continuous improvement events and immediately addressing the topmost items
  • Work in progress limits
  • Track data on every product initiative and do follow up post-release comparing adoption estimates to actuals
  • Always include a customer (or customers) on the team
  • Some regular rotation of team members to circulate information and learnings
  • Retrospectives for each initiative, and agreement to act on any of the key blockers whether or not it is solely within the team’s control
  • Immediate refocusing if rolling NPS drops below 40
  • CI/CD must not exceed N minutes/hours
  • Weekly goals
  • Chaos engineering in production environment (regularly bring down systems to test resiliency)
  • Periodically shortening the release cycle (e.g. 1 month, 2 weeks, 1 week, 1 day)

Scaling without imploding

blog comments powered by Disqus

Want to get my personal insights on what I learn as I learn it? Subscribe now!