ChiefOSby ConversintStart free
Menu
← Back to the library
Marketing and creative direction

Paid media planning

Plan, audit, optimize, and structure paid advertising campaigns for Google Ads, Meta, LinkedIn, TikTok, Twitter/X, PPC, paid media, ROAS, CPA, ad budgets, retargeting, audience targeting, bidding, campaign setup, health scoring, and ad spend decisions. Use for campaign strategy, launch readiness, reporting, performance diagnosis, and optimization. For bulk copy generation use /ad-creative; for organic social use /social-content; for Conversint's own Facebook/Instagram demand-test workflow use /fb-ads when applicable.

Works with the context you provideVersion 2.0.1Historical revision

# Paid Ads

Use this skill for paid media strategy, campaign structure, targeting, tracking, budget allocation, and optimization.

The goal is to make spend accountable. Do not recommend launching or scaling ads without a clear offer, landing page, conversion event, budget guardrail, and measurement plan.

Context Loading

Before asking questions, check whether one of these files exists and read it:

  1. `.agents/product-marketing-context.md`
  2. `.claude/product-marketing-context.md`
  3. `./references/dependencies/conversint-current/references/facebook-ads-playbook.md` when planning Conversint's own Facebook/Instagram demand tests
  4. `./references/dependencies/conversint-current/references/paid-ads-audit-engine.md` when auditing, reporting, optimizing, diagnosing, checking launch readiness, or deciding whether to scale/pause paid spend
  5. `./references/dependencies/conversint-current/references/digital-marketing-funnel-best-practices.md` when the campaign includes a paid landing page, CTA, form, lead offer, or conversion-path diagnosis

For Conversint-branded work, also read `./references/dependencies/conversint-current/references/marketing-message-governance.md`. When selecting a message or offer for a Conversint test, validate and consume the active-message catalog instead of reusing copy from a previous campaign plan.

Ask only for missing context that changes launch or optimization decisions.

Intake

If not clear, ask for:

  • Campaign goal: awareness, traffic, leads, demo, sale, app install, retargeting, learning
  • Platform or candidate platforms
  • Offer and landing page URL
  • Buyer, geography, exclusions, and buying trigger
  • Monthly or test budget
  • Target CPA, ROAS, CAC, or acceptable cost per qualified lead
  • Existing conversion tracking and pixel status
  • Current performance data if optimizing

If the user asks whether to run ads, first determine whether the funnel can convert paid traffic.

Platform Selection

| Platform | Best for | Use when | |---|---|---| | Google Search | Existing high-intent demand | Buyers search for the solution, category, pain, or competitor | | Meta | Demand generation and retargeting | Strong creative angles exist; audience can be reached by interest/behavior | | LinkedIn | B2B targeting | Job title, company size, industry, or account list matters | | TikTok | Video-led discovery | The team can make native short-form creative | | X | Tech and real-time narratives | Audience is active there and founder/POV content supports ads |

Campaign Workflow

  1. **Confirm readiness.** Offer, landing page, conversion event, tracking, budget.
  2. **Choose objective and platform.** Match the platform to intent and audience.
  3. **Structure campaigns.** Avoid fragmenting budget across too many tiny tests.
  4. **Define targeting.** Include exclusions and retargeting windows.
  5. **Specify creative tests.** Route copy generation to `/ad-creative` when needed.
  6. **Set budget guardrails.** Define test duration, daily cap, stop-loss, and scale rules.
  7. **Diagnose weekly.** Use `./references/dependencies/conversint-current/references/paid-ads-audit-engine.md` to classify the blocker before recommending changes.
  8. **Measure economics.** Read CPA/ROAS with landing conversion, qualified lead quality, and blended CAC context.
  9. **Record learnings.** Send audience, message, creative, and channel findings to `/marketing-learning-loop`; send economic findings to `/marketing-roi`.

Campaign Structure

Use clear naming:

```text [Platform]_[Objective]_[Audience]_[Offer]_[Date or Ongoing] ```

Testing phase:

  • 70% proven or highest-confidence audience/offer
  • 30% experiments
  • 2-4 creative concepts per meaningful audience
  • Avoid budget split so thin that no ad set can learn

Scaling phase:

  • Consolidate winners
  • Increase budgets gradually
  • Wait long enough for platform learning before judging

Optimization Rules

When the user asks "how are ads doing," "what should we change," "are these ads good," or "should we scale," do not jump straight to tactical edits. First classify the issue across:

  • tracking and attribution
  • creative diversity and fatigue
  • landing-page message match and conversion readiness
  • budget sufficiency and kill/scale rules
  • audience and retargeting seed quality
  • follow-up and CRM source capture

Use a paid-ads health score when enough evidence exists. If evidence is missing, produce a readiness/gap score and route the missing evidence to `/analytics-tracking`.

If CPA is high:

  1. Check conversion tracking.
  2. Check landing page conversion and message match.
  3. Check audience intent.
  4. Refresh creative angle.
  5. Adjust bidding or budget after the basics are sound.

If CTR is low:

  • Creative or offer is not resonating.
  • Audience may be wrong.
  • Hook is too vague.
  • Ad fatigue may be present.

If CPM is high:

  • Audience may be too narrow.
  • Placement or platform may be too competitive.
  • Creative relevance may be weak.

Output Format

```markdown ## Paid Ads Plan - Goal: - Platform: - Offer: - Conversion event: - Budget guardrail:

## Campaign Structure | Campaign | Audience | Offer | Budget | KPI |

## Targeting - Prospecting: - Retargeting: - Exclusions:

## Creative Tests | Angle | Format | Why | Next skill |

## Tracking Checklist - [ ] Conversion event tested - [ ] UTMs set - [ ] Landing page mobile checked - [ ] Exclusions configured

## Optimization Cadence - Weekly checks: - Stop-loss: - Scale rule:

## Diagnostic Readout - Health score: - Likely blocker: - One change next: - Metric to watch: ```

Reference Files

  • `./references/ad-copy-templates.md` - copy formulas and creative angle prompts.
  • `./references/audience-targeting.md` - targeting and exclusion patterns by platform.
  • `./references/platform-setup-checklists.md` - launch readiness checklists.
  • `./references/conversion-tracking.md` - pixel, event, and UTM setup.
  • `./references/dependencies/conversint-current/references/paid-ads-audit-engine.md` - diagnostic scoring, creative diversity checks, tracking gates, kill/scale rules, and learning-loop routing.

Learning And ROI Rule

Every paid test must define:

  • the hypothesis being tested
  • the spend cap
  • the qualified-response metric
  • the kill criterion
  • the ledger entry expected in `/marketing-learning-loop`
  • the data threshold for `/marketing-roi`