The dilemma
SoulCode gave teams the freedom to move. Some got faster. Others got slower. Nobody could say why.
The instinct was to look at the teams. Quinn looked at the edges instead. Freedom to move only works if everyone agrees where the edges are, and at SoulCode the edges were described rather than defined. Guardrails were fuzzy. Decision rights lived in people's heads. Dependencies stayed invisible until they bit.
Autonomy was not the problem. Unbounded autonomy was.
What was unclear
Edges. Guardrails were stated as principles, not as outcomes. Teams read them differently and both readings were defensible.
Rights. Who decides, who advises, and who simply needs to know were never written down. Every ambiguous call cost a round trip.
Seams. Work crossed team boundaries with no contract, so a decision inside one lane quietly reset the clock in another.
Recovery. Nobody had agreed how much room there was to be wrong, so any error looked like a failure rather than a cost of speed.
Quinn's playbook: audit the freedom
Outcome guardrails. Two or three measurable outcomes per team, plus explicit non-goals and quality bars. Freedom inside the box, clarity about the box.
Decision canvas. One page per significant call, public: Decider, Recommenders, Inputs, Escalation. If the four cannot be named, the decision is not ready to delegate.
Failure budgets. Pre-agreed error rates and rollback windows. Speed is affordable when the cost of being wrong is agreed in advance rather than argued about afterward.
Interface contracts. Every seam documented: inputs, outputs, SLAs, change windows, test hooks. Dependencies stop being a surprise once they are written down.
Autonomy heatmap. Team against dimension, covering outcomes, tech, timing, and scope. Red zones show where freedom was granted without the structure to hold it.
Shadow escalations off. When a choice hits a guardrail, escalate once, together, with receipts. No quiet side channels, no second conversation in a different room.
The shift at SoulCode
The heatmap did the most work. Two teams that everyone assumed were struggling turned out to be operating cleanly inside well-defined outcomes. Two that looked fast were burning their failure budget every sprint and paying it back in rework nobody had counted.
Once the decision canvas was public, the round trips stopped. Once the seams had contracts, the hidden dependencies became scheduling problems instead of surprises.
Freedom did not shrink. It got edges.
Leadership takeaways
Autonomy without edges looks fast until the rework bill arrives.
Decision rights are only real when they are written down.
A failure budget is permission to move, not permission to be careless.
Teaser for Episode 26: The Interface Pact
The audit fixed what happened inside the teams. It also made something else obvious: SoulCode could sprint inside a team and stumble between them. Quinn turns to the seams.
