Feedback Loops

Master the feedback loops. Recognize which are vicious, which are corrective, which are virtuous — then design systems that shift the balance.

Operating Stack

A loop system gets better when its source, measurement, and repair paths are all visible.

Layer	Job	Local instrument
Source	Improve what enters the system	Belief article standards: resonance, reader, instrument, proof, transfer
Loop	Run the work at the right grain	Mission Loop
Instrument	Measure structure without opinion	Open Knowledge Format
Repair	Act on the queue the instrument sees	Typed Context edges and proof edges
Wisdom	Raise the next setpoint	Wisdom — better shared judgment before lock-in

The complete pattern is:

better source -> measured structure -> repair queue -> proof response -> higher standard

If the source is weak, the graph becomes a cleanup machine. If the graph is absent, standards drift into taste. If repair is absent, measurement becomes theater. Wisdom appears when the next cycle starts with a better read than the last one did.

Three Loop Types

Every system runs on feedback. The loop type determines where it ends up.

Runaway (positive feedback): The output amplifies the input. Growth feeds growth — until it doesn't. Network effects, addiction, hyperinflation, viral spread. Feels like success while it's working. Becomes catastrophic when it flips.

Corrective (negative feedback): The output resists the input. Thermostats, market prices, body temperature. The system fights change to maintain equilibrium. Stable — but resistant to improvement. Most institutions are corrective loops protecting the status quo.

Virtuous (designed positive feedback): The output improves the setpoint. Each cycle makes the next cycle better. Compounding knowledge, reputation, relationships. Requires deliberate design — virtuous loops don't emerge naturally. They need a setpoint that serves beyond self, a gauge that measures reality, and a controller that adjusts.

The diagnostic question: Is your business model a runaway loop disguised as a virtuous one? Runaway loops feel virtuous while the signal is positive. The test is: what happens when growth stops? A virtuous loop continues improving. A runaway loop reverses.

Dig Deeper

• VVFL — The setpoint, gauge, controller, proof, and baseline lift pattern
• Mission Loop — The session-level learning loop: Picture, Ask, Mission, Practice, Calibrate, Reduce, Teach back
• Inner Game / Outer Game — How invisible intention becomes visible proof and returns as feedback
• Game Loops — How loops stack by timescale: rendering, gameplay, core, and meta
• Belief mass — The public story for why clear, connected, repeated, proven ideas become easier to act on

Context

• depends-on Systems Thinking — Loops need system boundaries before their signals make sense
• depends-on Open Knowledge Format — The graph instrument turns loop structure into measurable queues
• pairs-with Virtuous Feedback Loop — VVFL gives the controller pattern for improving the next cycle
• applies-to Wisdom — Wisdom is the human proof that a loop improves situated judgment
• proved-by Reality Scoreboard — A loop needs a visible proof surface, not only a story about improvement

Questions

Which loop layer is weakest right now: source standard, measuring instrument, repair action, or wisdom transfer?

• At what feedback cycle length does a virtuous loop become too slow to motivate participant behavior — and what design choices compress the cycle?
• How do you detect that a loop has shifted from virtuous to runaway before the damage compounds to the point of irreversibility?
• Which virtuous loop setpoint — beyond-self service, measurable reality, or compounding standards — is hardest to maintain as a venture scales?