Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Plan and write a paid social advertising campaign. Use when asked to build a paid social campaign, create Meta/LinkedIn/TikTok/X ad copy, define a social ad strategy, or plan an advertising funnel across social platforms. Produces a complete campaign plan with audience targeting, ad set structure, copy for each ad format, budget allocation, and measurement framework.
.claude/skills/mohitagw15856-social-ad-campaign/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 24% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 151% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 179% | 0% |
| case-12 | ✗→✓ | ▲ Improved | 384% | 0% |
| case-16 | ✗→✓ | ▲ Improved | 152% | 0% |
This skill produces a complete paid social advertising campaign plan covering campaign objective, audience targeting, funnel structure, ad set architecture, ad copy and creative briefs for each format, budget allocation, bidding strategy, and a measurement framework. Output is ready for a media buyer, performance marketer, or social team to execute.
Ask the user for these if not provided:
Campaign objective: e.g. Lead generation — 200 qualified leads in 30 days] Platform(s): e.g. Meta (Instagram + Facebook), LinkedIn] Budget: £/$/€ X total over X weeks] Campaign period: Start date → End date] Owner: Media buyer / performance marketer / agency] Date: Date]
Why paid social for this objective: 2–3 sentences justifying the platform and format choice for this specific goal and audience. E.g. "LinkedIn is the right channel for this B2B SaaS campaign — we can target by job title, company size, and seniority, ensuring budget reaches decision-makers, not browsers."]
Funnel structure:
| Stage | Objective | Audience | Budget allocation | |---|---|---|---| | Top of funnel (TOFU) | Awareness / reach | Cold audience — interest/behaviour targeting | X%] | | Middle of funnel (MOFU) | Consideration / engagement | Warm audience — video viewers, page engagers, website visitors | X%] | | Bottom of funnel (BOFU) | Conversion / lead | Hot audience — retargeting, custom audiences, lookalikes | X%] |
Platform: Meta / LinkedIn / TikTok] Audience size target: e.g. 500K–2M — broad enough to learn, narrow enough to be relevant]
| Targeting dimension | Settings | |---|---| | Location | Country / region / city] | | Age | e.g. 28–45] | | Gender | All / specify if relevant] | | Interests / behaviours | e.g. SaaS tools, productivity apps, small business owners] | | Job titles (LinkedIn) | e.g. Head of Marketing, Marketing Director, CMO] | | Company size (LinkedIn) | e.g. 50–500 employees] | | Industry (LinkedIn) | e.g. Technology, Financial Services, Healthcare] | | Exclude | e.g. Existing customers — upload suppression list] |
Platform: Meta] Source: People who engaged with content / visited website in last 30 days
| Signal | Action | |---|---| | Watched 50%+ of a video ad | Retarget with a case study or testimonial ad | | Visited product page but didn't convert | Retarget with a direct offer / free trial CTA | | Engaged with Instagram / Facebook page | Retarget with social proof ad |
Platform: Meta / LinkedIn] Source: Website visitors (last 7 days), abandoned cart, form started but not completed
Retargeting message: More direct. Address the specific action they took. Time-sensitive CTA.
Source: Existing customers / email list / best-converting website visitors] Lookalike similarity: 1%–3% (tight match) / 3%–10% (broader reach)] Platform: Meta
Campaign: [Campaign Name] — [Objective: Lead Generation / Traffic / Conversions]
│
├── Ad Set 1: TOFU — Cold Interests
│ ├── Ad 1A: [Video ad — hook format]
│ ├── Ad 1B: [Static image — benefit-led headline]
│ └── Ad 1C: [Carousel — feature/use case showcase]
│
├── Ad Set 2: MOFU — Warm Retargeting (30-day engagers)
│ ├── Ad 2A: [Social proof / testimonial]
│ └── Ad 2B: [Case study / before & after]
│
└── Ad Set 3: BOFU — Hot Retargeting (7-day website visitors)
├── Ad 3A: [Direct offer — free trial / discount / demo]
└── Ad 3B: [Objection handling — FAQ / reassurance]Campaign Group: [Campaign Name]
│
├── Campaign 1: [Job Title Targeting — Awareness]
│ ├── Single Image Ad: [Thought leadership hook]
│ └── Video Ad: [Problem/solution story]
│
├── Campaign 2: [Company Size + Industry — Consideration]
│ ├── Single Image Ad: [Case study / proof point]
│ └── Lead Gen Form: [Gated asset / webinar / demo]
│
└── Campaign 3: [Retargeting — Conversion]
└── Sponsored Message / Lead Gen Form: [Direct CTA with personalisation]Hook (first 3 seconds — must stop the scroll): > "Pattern interrupt question or statement — e.g. 'Are you still doing painful thing] manually?']"
Core message (seconds 4–20): > "Agitate the problem → introduce the solution → show the specific outcome]"
CTA (final 5 seconds): > "Clear, single action — e.g. 'Try free for 14 days — link in bio' / 'Get your demo today']"
Visual direction:
Caption copy: > Headline — max 40 chars] > Body copy — 1–3 sentences max] > CTA button label: e.g. "Learn More" / "Sign Up" / "Get Started"]
Ad variant A — Benefit-led headline:
| Element | Copy | |---|---| | Headline | "Single-sentence benefit statement — e.g. 'Cut reporting time by 80% with Product]']" | | Body copy | "Problem → solution in 2 sentences. Proof point if available.]" | | CTA | "Start free trial" / "Book a demo" / "Get 20% off" | | Image | Product UI / result visual / human context shot — no stock photos of people in suits] |
Ad variant B — Social proof headline:
| Element | Copy | |---|---| | Headline | "'Result] in timeframe]' — real customer result, or '500+ teams use Product] to...']" | | Body copy | "Expand on the proof. 1–2 sentences. Add a second proof point if available.]" | | CTA | "See how it works" / "Try it free" | | Image | Customer photo + quote overlay / logo wall / before/after data visual] |
Ad variant C — Curiosity/question headline:
| Element | Copy | |---|---| | Headline | "'Common misconception or challenging question]' — e.g. 'What if painful process] took 10 minutes, not 2 hours?']" | | Body copy | "Answer the question → introduce product → specific outcome]" | | CTA | "Find out how" |
Headline (shown above carousel): "Problem-first statement or benefit hook]"
| Card # | Headline | Description | Image | |---|---|---|---| | Card 1 (hook) | "Compelling hook — why this matters]" | "1-sentence setup]" | Eye-catching visual / stat] | | Card 2 | "Use case / feature 1]" | "Specific outcome this delivers]" | Product UI or illustration] | | Card 3 | "Use case / feature 2]" | "Specific outcome this delivers]" | Product UI or illustration] | | Card 4 | "Use case / feature 3]" | "Specific outcome this delivers]" | Product UI or illustration] | | Card 5 (CTA card) | "Strong CTA headline]" | "Reinforce the offer / urgency]" | CTA-focused visual / button] |
Intro text (shown before form): > "1–2 sentences on what they'll get and why it's worth 60 seconds of their time]"
Form headline: "Value-led headline — e.g. 'Get your free asset] / Book your 20-min demo']"
Form fields (keep to minimum — each extra field reduces conversion):
Privacy notice: Standard GDPR / CCPA compliance text — "By submitting, you agree to our Privacy Policy and may be contacted by Brand] about relevant products and services."]
Thank you message: > "What happens next — e.g. 'Thanks! You'll receive asset] in your inbox within 5 minutes. Our team will be in touch within 1 business day.']"
For website visitors (7 days) — direct offer: > Headline: "Specific nudge — e.g. 'Still thinking about Product]? Here's 20% off to make the decision easier.']" > Body: "Reinforce the primary benefit. Add urgency if genuine — e.g. 'Offer ends date]'.]" > CTA: "Claim offer" / "Start free trial" / "Book demo"
For video viewers (50%+) — social proof bridge: > Headline: "Continue the story — e.g. 'See what 50/100/500] teams achieved with Product]']" > Body: "Customer result quote or specific outcome. Bridge from awareness to consideration.]" > CTA: "Read the case study" / "See how it works"
Total budget: £/$/€ X over X weeks]
| Ad Set | Stage | Budget | % of total | Expected CPM | Expected CPC | Expected conversions | |---|---|---|---|---|---|---| | Ad Set 1 — Cold interests | TOFU | £X/week] | X%] | £X] | £X] | X leads / clicks] | | Ad Set 2 — Warm retargeting | MOFU | £X/week] | X%] | £X] | £X] | X] | | Ad Set 3 — Hot retargeting | BOFU | £X/week] | X%] | £X] | £X] | X] | | Total | — | £X/week] | 100% | — | — | X total] |
Bidding strategy:
Budget reallocation rule: After 7] days, pause ad sets with CPL > £X]. Reallocate budget to best-performing ad sets. Review weekly.
Primary KPI (tied to campaign objective):
| KPI | Target | Why | |---|---|---| | Cost per lead (CPL)] | ≤ £/$/€ X] | Primary success metric — every pound spent measured against leads generated] | | Conversion rate (ad → lead form)] | ≥ X%] | Quality of targeting and ad relevance] | | Total leads] | ≥ X in X weeks] | Volume target] |
Secondary metrics (optimisation signals):
| Metric | Target | Action if off-target | |---|---|---| | CTR (click-through rate) | ≥ X%] | Test new headlines / hook variations] | | CPM (cost per 1K impressions) | ≤ £/$/€ X] | Broaden audience / test new placements] | | Video completion rate (if video) | ≥ X%] | Test shorter video / stronger hook] | | Lead form completion rate | ≥ X%] | Reduce form fields / test form intro copy] | | Lead-to-opportunity rate (post-campaign) | ≥ X%] | Review lead quality — tighten audience targeting] |
Reporting cadence:
Attribution model: Last-click / 7-day click + 1-day view / data-driven (if volume sufficient)]
Tracking setup checklist:
Run structured tests — change one variable at a time:
| Test # | Variable | Control | Variant | Success metric | Min budget to run | |---|---|---|---|---|---| | 1 | Hook / headline | Current headline] | Challenger headline] | CTR | £X / 500 impressions] | | 2 | Creative format | Static image | Video | CPL | £X / 1,000 impressions] | | 3 | CTA | "Learn More" | "Start free trial" | Conversion rate | £X / 200 clicks] | | 4 | Audience | Interest-based | Lookalike 1% | CPL | Equal budget split] |
Testing rules:
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 46,358 | 40,052 | -14% | 1 | 1 | 0% | 8,346 | 10,330 | +24% | 0 | 0 | — |
case-02 | fail→fail | 50,500 | 42,619 | -16% | 1 | 1 | 0% | 8,347 | 10,852 | +30% | 0 | 0 | — |
case-03 | fail→fail | 46,514 | 50,021 | +8% | 1 | 1 | 0% | 7,205 | 10,646 | +48% | 0 | 0 | — |
case-04 | pass→pass | 56,126 | 43,063 | -23% | 1 | 1 | 0% | 8,248 | 10,111 | +23% | 0 | 0 | — |
case-05 | pass→pass | 27,599 | 28,936 | +5% | 1 | 1 | 0% | 4,238 | 8,170 | +93% | 0 | 0 | — |
case-06 | pass→pass | 27,297 | 29,438 | +8% | 1 | 1 | 0% | 3,581 | 9,244 | +158% | 0 | 0 | — |
case-07 | pass→fail | 26,280 | 27,035 | +3% | 1 | 1 | 0% | 3,592 | 7,762 | +116% | 0 | 0 | — |
case-08 | fail→pass | 18,227 | 20,452 | +12% | 1 | 1 | 0% | 2,648 | 6,639 | +151% | 0 | 0 | — |
case-09 | pass→pass | 28,008 | 47,465 | +69% | 1 | 1 | 0% | 3,084 | 9,495 | +208% | 0 | 0 | — |
case-10 | fail→fail | 19,073 | 42,991 | +125% | 1 | 1 | 0% | 3,259 | 9,174 | +181% | 0 | 0 | — |
case-11 | fail→pass | 20,868 | 35,304 | +69% | 1 | 1 | 0% | 3,233 | 9,016 | +179% | 0 | 0 | — |
case-12 | fail→pass | 7,862 | 20,330 | +159% | 1 | 1 | 0% | 1,321 | 6,390 | +384% | 0 | 0 | — |
case-13 | fail→fail | 13,469 | 13,570 | +1% | 1 | 1 | 0% | 1,496 | 5,844 | +291% | 0 | 0 | — |
case-14 | fail→fail | 15,215 | 28,831 | +89% | 1 | 1 | 0% | 2,547 | 7,928 | +211% | 0 | 0 | — |
case-15 | fail→fail | 11,503 | 19,923 | +73% | 1 | 1 | 0% | 1,782 | 6,270 | +252% | 0 | 0 | — |
case-16 | fail→pass | 16,063 | 23,623 | +47% | 1 | 1 | 0% | 2,748 | 6,915 | +152% | 0 | 0 | — |
case-17 | fail→fail | 13,746 | 26,374 | +92% | 1 | 1 | 0% | 2,239 | 7,532 | +236% | 0 | 0 | — |
case-18 | fail→fail | 13,808 | 30,968 | +124% | 1 | 1 | 0% | 2,021 | 8,165 | +304% | 0 | 0 | — |
case-19 | fail→pass | 8,817 | 15,034 | +71% | 1 | 1 | 0% | 1,388 | 6,295 | +354% | 0 | 0 | — |
case-20 | fail→pass | 18,757 | 37,188 | +98% | 1 | 1 | 0% | 2,885 | 9,934 | +244% | 0 | 0 | — |
case-21 | fail→pass | 25,813 | 37,345 | +45% | 1 | 1 | 0% | 4,118 | 10,864 | +164% | 0 | 0 | — |
case-22 | pass→pass | 21,249 | 27,337 | +29% | 1 | 1 | 0% | 3,166 | 7,820 | +147% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +32 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.