Translate every discrepancy into a specific behavioral…
When a feedback loop identifies a discrepancy between current and target state, translate the evaluation into a specific behavioral adjustment for the next cycle rather than stopping at awareness, because learning occurs during adjustment not observation.
Why This Is a Rule
The most common feedback loop failure mode isn't missing observation or evaluation — it's missing adjustment. People observe ("my writing sessions are only producing 200 words"), evaluate ("that's below my 500-word target"), and then... do nothing different next time. They've run 75% of the loop and stopped at the hardest part: translating awareness into specific behavioral change.
Awareness feels like progress. "I know I'm below target" creates a sense of having addressed the issue. But awareness without adjustment is like a thermostat that reads the temperature and reports it's too cold but never turns on the heat. The system has observation and evaluation but no controller — and the temperature stays cold indefinitely.
The adjustment must be specific and behavioral — not "try harder" but "next session, start with the conclusion paragraph rather than the introduction to reduce blank-page friction." The specificity requirement (Write agent actions as procedures a stranger could follow…) ensures the adjustment is executable, and the "next cycle" timing ensures the learning is applied immediately rather than deferred into abstraction.
When This Fires
- Every time a feedback loop produces a discrepancy between current and target state
- When review sessions produce insights but not changes — you're stopping at evaluation
- When the same discrepancy appears in multiple consecutive reviews — the adjustment phase is failing
- Complements A complete feedback loop needs three elements (feedback loop structure) by emphasizing the most frequently missing component
Common Failure Mode
Stopping at awareness: "I noticed my meetings run over time." This is observation + evaluation without adjustment. What will you do differently in the next meeting? Without a specific behavioral change, the next meeting will run over for the same reasons. The loop must complete: observe → evaluate → adjust → act differently in the next cycle.
The Protocol
(1) When a feedback loop reveals a discrepancy (current state ≠ target state), don't stop at "I see the gap." (2) Ask: "What specific thing will I do differently in the next cycle?" (3) The answer must be behavioral and specific: not "be more focused" but "close email and put phone in drawer before starting." Not "run meetings better" but "state the decision question in the first 2 minutes." (4) Write the adjustment down — it's a modification to your process for the next iteration. (5) In the next cycle, implement the adjustment. After that cycle, re-observe: did the adjustment reduce the discrepancy? If yes → keep it. If no → the adjustment targeted the wrong cause. Diagnose again and try a different specific adjustment.