📣 paidads — Paid Advertising Department

One head skill for the whole paid advertising department. Channel modes share one campaign doctrine — objective before budget, audience before creative, tracking before launch, and data before opinion. Each channel mode applies the doctrine to that platform's native objects.

Modes

Every invocation resolves to exactly one mode. Match the request, then load only the matched reference:

Only the resolved mode's reference is loaded — the rest stay on disk, saving tokens on every run.

When to Use

• Building or restructuring a paid campaign on any major ad platform.
• Planning budgets, bids, or pacing across channels.
• Writing platform-native ad copy and creative briefs.
• Setting up conversion tracking, pixels, and events before launch.
• Building retargeting/remarketing funnels from audiences.
• Triangulating blended MER and CAC across channels or enforcing 4-tier spend pacing gates.
• Auditing an underperforming ad account or scaling a working one.

Anti-Triggers

• Organic social strategy, calendars, community → smm.
• SEO / AEO / organic search visibility → seo.
• Building the measurement layer itself (GA4, dashboards, attribution) → analytics.
• Ad copy as part of broader site copywriting → content.

Quick Reference

Universal campaign doctrine (applies to every channel mode)

1. Objective first. Name the business objective in one word (awareness/traffic/leads/sales/retention). No objective, no campaign.
2. Audience before creative. Define who sees it before deciding what they see.
3. Tracking before launch. Pixel installed, conversion events verified, UTM convention applied — or the campaign does not launch.
4. Creative volume. 3–5 variants per ad set at launch (test messaging, not colors).
5. Data before opinion. Decisions after the data window (typically 7 days or 50 conversions per variant), never before.
6. Budget floor. Per-ad-set daily budget ≥ 10× target CPA, or the algorithm can't exit learning.

Intake gate

UTM convention (suite standard)

utm_source={platform}&utm_medium=paid&utm_campaign={campaign_slug} — derive utm_term/utm_content per variant. Apply on every ad; analytics reads this contract.

Message-to-market foundation (run before any channel mode; source: kim-barrett advertising-skills MIT — copy-chief/schwartz-awareness-mapper, copy-chief/mechanism-builder, copy-chief/headline-matrix, orchestrators/full-funnel-campaign-orchestrator)

1. Diagnose awareness first. Place the audience on the 5-stage ladder (unaware → problem-aware → solution-aware → product-aware → most-aware) by asking: do they know the problem, do they know solutions exist, do they know us? Then set messaging rules (what you can say / must not say) and pick the angle type: education, agitation, comparison, or offer. Wrong stage = dead ad; match message to awareness, not product.
2. Build the mechanism. Name what the audience tried before and why it failed, then define the new mechanism (what is different, why it works now) under a simple, memorable, ownable name. Believable and distinct beats generic.
3. Vary headlines structurally. Generate ~25 headlines across 7 types (curiosity, specificity, proof, urgency, contrarian, mechanism, identity) — each structurally distinct, specific over vague.
4. Chain the campaign in order. avatar → offer → awareness → mechanism → angles → creative → path → objection → language-kill; one consistent strategy/ads/funnel output, never a skipped step.

Platform table (verify current specs against platform docs before launch)

Procedure

1. Intake. confirm the campaign objective (one of awareness / traffic / leads / sales / retention) — every downstream choice (channel, audience, bid, creative, KPI) is derived from it.
2. Resolve the mode. Match the request against the mode table; exactly one mode. Ambiguous → ask one question, then proceed.
3. Execute the mode playbook. Load references/<mode>.md and follow its Intake → Deliverable → Procedure → Quality gate in order.
4. Gate and deliver. Pass this file's Verification checklist plus the mode's quality gate, then deliver the artifact where the client expects it.

Pitfalls

• Launching before tracking is verified — unmeasurable spend; tracking gate comes first.
• Structuring by gut instead of the platform's native objects (ad sets, groups, lines).
• Editing budgets/creatives mid-learning-phase — resets the algorithm.
• One creative per ad set — nothing to learn from; 3–5 variants minimum.
• Broad match with no negatives (Google) — burning budget on irrelevant queries.
• Treating all channels the same — Reddit punishes promo-speak; LinkedIn needs firmographic copy.
• Optimizing to clicks when the objective is leads/sales — optimize to the business event.
• No kill criteria — a loser without a target CPA/ROAS floor runs forever.

Verification

• Objective named in one word before any platform object was created.
• Awareness stage diagnosed and angle type chosen before creative; mechanism named before headlines.
• Tracking verified (pixel fires, events test-fire, UTM applied) before launch.
• 3–5 creative variants per ad set, differing on message not color.
• Budget floor respected: ≥ 10× target CPA per ad set.
• Kill criteria stated upfront (target CPA/ROAS, data window).
• Findings and optimizations cite the platform metric they came from.