界面已本地化;技能介绍与说明文档按所选语言由 AI 自动翻译,代码、命令、路径和专业术语保留原文,以原作者文档为准。
探索技能
效率与协作 · SKILL.md

ads-plan

ads-plan

【效率与协作】Create a professional paid-advertising strategy covering objectives, economics, platform selection, campaign architecture, audiences, budget, creative, measurement, experiments, governance, rollout, and reporting. Use for ad plan, media plan, PPC strategy, paid-social strategy, campaign architecture, advertising roadmap, or channel planning.

独立 GitHub 快照 · 自动分类 · 以原作者文档为准

技能说明

GitHub

name: ads-plan description: "Create a professional paid-advertising strategy covering objectives, economics, platform selection, campaign architecture, audiences, budget, creative, measurement, experiments, governance, rollout, and reporting. Use for ad plan, media plan, PPC strategy, paid-social strategy, campaign architecture, advertising roadmap, or channel planning."

Paid Media Plan

  1. Read the setup profile, business economics, evidence, current account state, and relevant industry template as optional input.
  2. Define objective, customer, offer, conversion, value, constraints, geography, regulated category, time horizon, and success criteria.
  3. Evaluate channel roles and exclusions from first principles; do not require every supported platform. For a channel outside the twelve-platform contract, load ads/references/additional-platforms.md and return a research lead unless its current buying, eligibility, measurement, and creative evidence is present.
  4. Before proposing Meta budgets, learning-phase expectations, bidding, consolidation, or performance forecasts, collect the account, Pixel, and conversion cold-start dimensions per the contract in skills/ads-meta/SKILL.md. When any dimension is cold or unknown, plan for measurement validation, explicit creative hypotheses, staged reversible tests, and confidence labels; do not apply mature-account benchmarks to missing history.
  5. Specify campaign architecture, audience strategy, creative system, budget and pacing, measurement, experiments, policy controls, and operating cadence.
  6. Phase prerequisites before launch, learning, optimization, and scale.
  7. Assign every action an owner, timing, dependency, guardrail, evidence, success measure, and rollback or exit condition.
  8. Return canonical JSON and render the requested human plan.

A plan is advisory. It becomes an account change only through launch or optimize mutation gates.

我的收藏