Content Signal Lab

Problem

Creators often treat “viral” as a vague goal: no testable expectation before posting and no disciplined attribution after. This skill reframes content work as an auditable decision chain: define audience tension, promise, one variable, and a comparable baseline; publish; then evaluate the hypothesis with fixed-window metric evidence instead of rewriting the conclusion after the fact.

How it works

It organizes the workflow as signal -> hypothesis -> experiment -> locked bet -> release -> evidence -> playbook. Core actions include:
- Initialize project state for platform, audience, objective, cadence, traffic lane, and historical data availability.
- Build an experiment card that names the primary variable, primary metric, guardrail, predicted range, failure condition, and review window.
- Lock the forecast before release so success criteria cannot be changed after seeing results.
- Classify outcomes as supported, refuted, or inconclusive, separating facts, inferences, and assumptions.
- Update the playbook only when evidence reaches a higher level, preventing one result from becoming a general rule.

Boundaries

It is useful for content accounts, short video, long-form writing, and mixed organic/paid distribution, including China-market compliance and platform differences. It does not guarantee reach, followers, sales, or legal compliance. When data is thin, variables are confounded, or baselines are missing, the better output is a clear evidence gap rather than forced attribution.