Back to Operations & Scaling
Operations & Scaling ChecklistIntermediate

Shipping On Fridays — A Checklist

If you must ship on a Friday, here's the 12-item checklist that makes Monday not suck.

EE
Published 1mo agoUpdated 1h ago 543

Friday deploys aren't always wrong. They're risky. Here's the survival kit.

Pre-deploy

  • Two people on call (not "available" — actually on).
  • Rollback plan tested in the last week.
  • Feature flag wraps the change, defaulting to off.
  • Database migrations are reversible.
  • Smoke test script ready to run post-deploy.

Deploy

  • One person drives, one watches.
  • Deploy starts before 2pm local.
  • Smoke tests pass before announcing.
  • Errors dashboards open in tabs.

Post-deploy

  • Monitor for 60 minutes minimum.
  • Customer support pinged about the change.
  • Internal Slack updated.

Why this checklist exists

Friday deploys go wrong not because Friday is special, but because everyone wants to leave early. The checklist forces patience.

Weekly digest

Get new resources like this, weekly.

One email a week: new hubs, new tools, and the editorial pieces worth reading. One click to unsubscribe.

Discussion

0 comments

Sign in to join the discussion.

Be the first to comment. The Bible community reads every thread.

Keep reading

More from Operations & Scaling