The Learning Lag

You made a change on Monday. Something broke. But which change caused it?

By the time you saw the problem, you'd made 47 more changes.

You suspected something was broken in your feedback loop. It always was.

This is the learning lag: the gap between action and visible consequence. The longer it takes to see impact, the less you can learn from each cycle.

Monthly deploys mean 30 days of flying blind. Make change. Wait. Forget context. See result. Try to remember what you did. Fix wrong thing. Wait again.

Contrast with continuous deployment: change, observe, adjust, repeat. Same day. Full context. Clear causality.

Forsgren's research on elite performers is brutal: they deploy 200 times more frequently than low performers. Not because they work harder. Because shorter loops mean faster learning.

Every week of delay is a week of decisions made on stale information. You're not being careful. You're being uninformed.

The Andon cord at Toyota worked because the feedback was instant. Pull cord, production stops, problem gets solved. Same hour. That's The Speed of Being Wrong in action.

Context degrades with time. Speed isn't recklessness. It's the only way to learn. The Momentum Engine shows how to collapse the loops.