Campaign Iteration Playbook

Every campaign variant should test one explicit assumption. "We believe that [changing X] will improve [metric Y] because [reason Z]." This keeps experiments interpretable and avoids conflating multiple changes.

• Frame hypotheses around audience (who), payout (how much), or task framing (how the action is described).
• Keep one variable changing at a time for clean comparisons; if you change both copy and payout, you cannot attribute results to either.
• Set a review window and minimum sample size (e.g. 50 completions) before making decisions so you do not overreact to noise.

Optimization works when leading indicators (early signals) and lagging indicators (outcome signals) are both tracked. Leading metrics help you correct course mid-campaign; lagging metrics tell you if the campaign actually delivered value.

• Leading: completion speed, acceptance ratio, active participants—watch these during the run.
• Lagging: conversion quality, retention signal, budget efficiency—evaluate these after the run or at milestones.
• Benchmark each run against your previous two campaigns so you can see trend, not just point-in-time performance.

Iteration should produce reusable playbooks, not one-off wins. Document what worked and what did not so the next launch starts from a higher baseline.

• Promote winning variants (copy, payout, targeting) into your default campaign template for the next launch.
• Archive failed experiments with concise reasons (e.g. "Lower payout reduced completion but did not improve quality").
• Revisit strategy monthly as platform and audience behavior evolves; treat playbooks as living documents.