Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →When the user wants to set up, optimize, or manage LinkedIn Ads. Also use when the user mentions "LinkedIn Ads," "LinkedIn Campaign Manager," "Sponsored Content," "Lead Gen Forms," "Sponsored Messaging," "Message Ads," "Conversation Ads," "Accelerate," "Classic ad set," "Audience Network," "objective-based pricing," "Insight Tag," "job title targeting," "company targeting," "Matched Audiences," or "B2B paid ads." For organic posts, use linkedin-posts.
.claude/skills/mkurman-linkedin-ads/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 65% | 0% |
| case-11 | ✗→✓ | ▲ Improved | 83% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 56% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 85% | 0% |
-|------| | Campaign group | Groups related programs; shared status and budgets (optional). | | Campaign | Marketing objective; budget/scheduling as configured. | | Ad set | Accelerate or Classic; audience, placements, format, budget, schedule, bid; conversion/URL params. Compare types | | Ads | Creative; multiple ads per ad set for A/B. |
Billing model: Objective-based pricing — billed on the action tied to the objective (e.g. clicks for traffic-oriented objectives). Pricing overview
Budget pacing: Auction-based; on daily budgets, spend on a day may exceed the daily number by a margin; pacing smooths so 7-day spend does not exceed about 7× daily (per LinkedIn’s public budget article—always confirm the latest text). Budgets and pacing
Minimum spend: There is no one global fixed minimum across all cases; very low daily/lifetime budgets are often rejected in UI. A common practical floor cited in the industry is on the order of ~$10 USD/day per ad set (not a platform guarantee). Use Campaign Manager as source of truth.
| Funnel | Example objectives (names evolve—pick in UI) | Typical use | |--------|-----------------------------------------------|------------| | Awareness | Brand awareness | Broad reach | | Consideration | Website visits, Engagement, Video views | Traffic, company/page engagement, video | | Conversion | Lead generation, Website conversions | Leads, Lead Gen Forms, on-site events | | Talent | Job applicants, Talent leads (where eligible) | Hiring |
Full list: Marketing objectives
Ads Guide (authoritative for specs)
Aligns with organic shapes in linkedin-posts (image, video, document, etc.) as paid delivery:
| Type | Notes | |------|--------| | Message Ads | One focused message + CTA; body often ~1,500 characters in product docs (check Ads Guide) | | Conversation Ads | Branching buttons; long body allowance (often ~8,000 characters in docs); up to ~5 CTA buttons (check guide) |
Tracking
| Signal | Use | |--------|-----| | Job title, function, seniority | ICP and committees | | Company | Industry, size, name lists (ABM) | | Skills, interests, groups | Technical or topical | | Matched Audiences | Contact/company lists, retargeting (Insight Tag) | | Exclusions, expansion | Audience expansion (when offered)—broaden carefully |
Lookalike audiences: As of current LinkedIn public documentation, Lookalike is not the default expansion path; it has been deprecated/removed in many accounts—rely on first-party lists, Matched Audiences, and exclusion/expansion options shown in the UI. See Target audience size / practices and the live Campaign Manager for your tenant.
Principle: LinkedIn is expensive; start narrow, high-intent; then scale with measured expansion.
Audience size: A common operational band mentioned in industry is hundreds+ in targetable size; the UI will warn if the audience is too small—follow it.
| Topic | URL | |------|-----| | Ads Guide (formats) | https://business.linkedin.com/advertise/ads/ads-guide | | Pricing and auction | https://business.linkedin.com/advertise/ads/pricing | | Create campaigns | https://www.linkedin.com/help/lms/answer/a9509136 | | Objectives | https://www.linkedin.com/help/lms/answer/a424570 | | Placements (Text ads, etc.) | https://www.linkedin.com/help/lms/answer/a417880 | | Targeting options | https://www.linkedin.com/help/lms/answer/a424655 | | LMS help home | https://www.linkedin.com/help/lms | | Get started (what you need) | https://business.linkedin.com/advertise/ads/best-practices/what-you-need-to-get-started |
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 20,225 | 21,024 | +4% | 1 | 1 | 0% | 3,316 | 5,475 | +65% | 0 | 0 | — |
case-02 | fail→fail | 18,067 | 16,745 | -7% | 1 | 1 | 0% | 2,773 | 4,739 | +71% | 0 | 0 | — |
case-03 | fail→fail | 23,585 | 25,146 | +7% | 1 | 1 | 0% | 3,617 | 5,983 | +65% | 0 | 0 | — |
case-04 | pass→pass | 9,921 | 7,717 | -22% | 1 | 1 | 0% | 1,625 | 3,381 | +108% | 0 | 0 | — |
case-05 | pass→pass | 12,372 | 12,101 | -2% | 1 | 1 | 0% | 1,944 | 4,118 | +112% | 0 | 0 | — |
case-06 | pass→pass | 7,437 | 6,325 | -15% | 1 | 1 | 0% | 1,124 | 3,116 | +177% | 0 | 0 | — |
case-11 | fail→pass | 13,670 | 11,158 | -18% | 1 | 1 | 0% | 2,092 | 3,837 | +83% | 0 | 0 | — |
case-07 | fail→pass | 15,457 | 8,124 | -47% | 1 | 1 | 0% | 2,174 | 3,348 | +54% | 0 | 0 | — |
case-08 | fail→pass | 17,065 | 12,887 | -24% | 1 | 1 | 0% | 2,737 | 4,275 | +56% | 0 | 0 | — |
case-09 | pass→pass | 13,182 | 12,050 | -9% | 1 | 1 | 0% | 2,019 | 3,977 | +97% | 0 | 0 | — |
case-10 | fail→pass | 15,859 | 13,869 | -13% | 1 | 1 | 0% | 2,283 | 4,225 | +85% | 0 | 0 | — |
case-12 | pass→pass | 10,644 | 8,647 | -19% | 1 | 1 | 0% | 1,624 | 3,444 | +112% | 0 | 0 | — |
case-13 | pass→pass | 6,644 | 3,764 | -43% | 1 | 1 | 0% | 1,088 | 2,787 | +156% | 0 | 0 | — |
case-14 | fail→fail | 10,838 | 9,740 | -10% | 1 | 1 | 0% | 1,664 | 3,722 | +124% | 0 | 0 | — |
case-15 | pass→pass | 15,864 | 12,099 | -24% | 1 | 1 | 0% | 2,476 | 3,981 | +61% | 0 | 0 | — |
case-16 | pass→pass | 13,626 | 11,819 | -13% | 1 | 1 | 0% | 2,063 | 3,887 | +88% | 0 | 0 | — |
case-17 | pass→pass | 10,924 | 17,086 | +56% | 1 | 1 | 0% | 1,761 | 3,605 | +105% | 0 | 0 | — |
case-18 | pass→pass | 12,886 | 10,711 | -17% | 1 | 1 | 0% | 1,900 | 3,796 | +100% | 0 | 0 | — |
case-19 | pass→pass | 17,132 | 13,510 | -21% | 1 | 1 | 0% | 2,694 | 4,284 | +59% | 0 | 0 | — |
case-20 | pass→pass | 11,889 | 8,013 | -33% | 1 | 1 | 0% | 2,224 | 3,497 | +57% | 0 | 0 | — |
case-21 | pass→pass | 12,886 | 9,450 | -27% | 1 | 1 | 0% | 2,002 | 3,620 | +81% | 0 | 0 | — |
case-22 | fail→pass | 10,381 | 8,560 | -18% | 1 | 1 | 0% | 1,625 | 3,517 | +116% | 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 +27 percentage points is the difference between those two pass rates over the 22 comparable cases.
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.