Some work should be made more legible immediately.

Operators should not use the value of illegibility as an excuse for fog. Many organizations are not suffering from too much visibility. They are suffering from unclear ownership, hidden dependencies, ambiguous decisions, buried risks, and commitments nobody can inspect.

Legibility is essential when coordination or accountability depends on shared reality.

Make decisions legible

Decisions decay when they are not written down.

People remember different versions of the conversation. The rationale disappears. The tradeoff becomes controversial again. Someone who was not in the room reopens the call. A team optimizes against an old assumption because nobody told them what changed.

A decision should become legible when it affects other people’s work.

That does not require a bureaucratic artifact. It requires the basics:

  • what was decided;
  • who decided;
  • why this option was chosen;
  • what alternatives were rejected;
  • what changes now;
  • who owns follow-through;
  • when the decision should be revisited.

A decision that remains private while other teams must act on it is not discretion. It is operating debt.

Make dependencies legible

Dependencies are where optimistic plans go to die.

The team says it is on track because its own work is on track. Meanwhile legal review is late, data definitions are unresolved, platform support is unavailable, sales enablement is waiting, and the customer migration plan depends on a team that has not committed.

A dependency should become legible before it becomes a blocker.

Operators make dependencies visible by naming owner, date, consequence, and escalation path. Not “we need support from Platform,” but “Platform needs to approve the migration interface by May 20 or the launch moves by one week; owner is Maya; escalation path is VP Eng if unresolved by May 13.”

That level of legibility prevents fake alignment.

Make commitments legible

A commitment is not real until the system can inspect it.

“We will improve onboarding” is not a commitment. “Growth owns a new onboarding experiment that reduces first-value time for self-serve users; first version ships by June 14; activation lift is the evidence; Product and Data are dependencies” is closer.

Commitments need outcome, owner, scope, evidence, timing, and tradeoff. They also need a visible “not doing” list. Otherwise every commitment is secretly competing with work nobody was willing to stop.

Legibility at commitment time is kinder than accountability theater after the miss.

Make risks legible

Risk hidden until it is certain is not risk management. It is surprise management.

Operators should make risks legible while they are still actionable. That means naming probability, impact, trigger, mitigation, and decision needed.

Weak version: “There may be some data concerns.”

Useful version: “We have a 30–40% risk that the attribution model produces inconsistent activation numbers across Sales and Product dashboards. If unresolved by the next operating review, Q3 planning will use two conflicting baselines. Data owns reconciliation; decision needed is whether Finance accepts the interim definition.”

The point is not to dramatize uncertainty. The point is to give the organization a chance to act before uncertainty becomes damage.

Make ownership legible

If ownership is unclear, everything else gets political.

People route questions to whoever answers fastest. Dependencies go to the most helpful team, not the accountable one. Decisions are made by whoever cares most. Work stalls because everyone assumes someone else has the authority.

Legible ownership names the person or forum responsible for outcome, decision, execution, and escalation. These are not always the same.

A project may have one accountable owner, several execution owners, one decision forum, and a different escalation path. If those roles are not explicit, power will still exist. It will just be illegible.

Make interpretation legible

Metrics do not speak.

A number can say activation is down. It cannot say whether the cause is traffic quality, product friction, instrumentation change, onboarding messaging, seasonality, or a segment mix shift. It cannot say what to do.

Operators should make interpretation legible when metrics enter decision forums. The useful artifact is not a metric dump. It is the operator read: what changed, what we think it means, what evidence supports that read, what uncertainty remains, and what action follows.

The boundary

Make work more legible when other people must coordinate, decide, fund, trust, or hold commitments around it.

Do not make work more legible just because someone feels anxious. Anxiety often asks for more visibility when the real need is clearer ownership, better standards, or a sharper escalation path.

The standard is not maximum visibility. The standard is enough shared reality for responsible action.