Skip to content

Growth-Today/claude-skills / gt-linkedin-ads

gt-linkedin-ads

Verified

GT LinkedIn Ads — expert LinkedIn Ads strategist for B2B companies, built by Growth Today (growthtoday.co). Use when asking about LinkedIn advertising, LinkedIn campaign setup, LinkedIn ad targeting, LinkedIn bidding strategies, LinkedIn ad formats, LinkedIn retargeting, LinkedIn ABM campaigns, LinkedIn Thought Leader Ads, sequenced TLAs and Story Arcs, LinkedIn funnel architecture, LinkedIn ads measurement/attribution, LinkedIn ads reporting (weekly, monthly, QBR), LinkedIn ads troubleshooting, LinkedIn creative best practices, or any B2B paid social strategy involving LinkedIn. Triggers on: LinkedIn campaign, LinkedIn CPM, LinkedIn CTR, LinkedIn lead gen, B2B ads, demand gen on LinkedIn, sponsored content, thought leader ads, Predictive Audiences, Accelerate, Conversions API, LinkedIn ads report, monthly report, LinkedIn ads not working. Do NOT use for LinkedIn organic content (use gt-linkedin-content) or LinkedIn outbound messaging (use gt-linkedin-outbounder). More GT skills: growthtoday.co.

FileKindSize
.claude/skills/abm-strategy/gt-SKILL.mdmarkdown3.8 KB
.claude/skills/ads-outbound-sync/gt-SKILL.mdmarkdown3.4 KB
.claude/skills/audiences/gt-SKILL.mdmarkdown3.2 KB
.claude/skills/audit/gt-SKILL.mdmarkdown2.4 KB
.claude/skills/bidding/gt-SKILL.mdmarkdown3.3 KB
.claude/skills/campaign-setup/gt-SKILL.mdmarkdown3.8 KB
.claude/skills/copy/gt-SKILL.mdmarkdown3.3 KB
.claude/skills/creative/gt-SKILL.mdmarkdown7.8 KB
.claude/skills/measurement/gt-SKILL.mdmarkdown3.8 KB
.claude/skills/optimization/gt-SKILL.mdmarkdown3.7 KB
.claude/skills/reporting/gt-SKILL.mdmarkdown4.2 KB
CHANGELOG.mdmarkdown18.3 KB
README.mdmarkdown1.5 KB
SKILL.mdskill md9.8 KB
resources/linkedin-ads-knowledge-base.mdmarkdown14.2 KB
resources/references/abm/ads-outbound-signaling-guide.mdmarkdown11.4 KB
resources/references/abm/linkedin-ads-abm-guide.mdmarkdown4.8 KB
resources/references/accelerate-ai-campaigns.mdmarkdown3.8 KB
resources/references/ad-formats.mdmarkdown5.2 KB
resources/references/audit-checklist.mdmarkdown4.0 KB
resources/references/benchmarks.mdmarkdown5.4 KB
resources/references/bidding-objectives.mdmarkdown2.7 KB
resources/references/competitive-research.mdmarkdown1.9 KB
resources/references/creative-strategy.mdmarkdown3.1 KB
resources/references/crm-attribution.mdmarkdown4.9 KB
resources/references/funnel-architecture.mdmarkdown3.0 KB
resources/references/measurement-attribution.mdmarkdown7.4 KB
resources/references/predictive-audiences.mdmarkdown3.9 KB
resources/references/reporting-playbook.mdmarkdown8.6 KB
resources/references/targeting-audiences.mdmarkdown3.3 KB
resources/references/troubleshooting.mdmarkdown5.2 KB

CHANGELOG.md

View on GitHub
# Changelog — gt-linkedin-ads

All notable changes to this skill. Newest first.

---

## v2.3.0 — 2026-08-31

**Summary:** New **reporting** sub-skill — the recurring-report layer the skill was missing. Audit answers "what is broken once"; reporting answers "what happened this period, against what was expected, and what changes next." No existing content removed.

**Added:**
- `.claude/skills/reporting/gt-SKILL.md`: weekly delivery checks, monthly performance reports, quarterly pipeline reads. Data-source order, refusal to fabricate missing numbers, 8-section output, routing boundaries against measurement / audit / optimization / bidding.
- `resources/references/reporting-playbook.md`: reporting contract, metric definitions (incl. total clicks vs landing-page clicks), grading rules, per-cadence claim limits, report structure, reporting traps, automation notes.

**New rules introduced:**
- **Effective CPC-to-landing-page** (spend / LP clicks) is the grading metric for link-driving formats, not raw CPC. Derived from the LP-clicks-per-$1K figures already in `benchmarks.md`: TLA ~$3.06, single image ~$14.08, video ~$16.13. Raw CPC flatters formats with high social-action volume — TLAs most of all.
- **Cadence claim limits:** weekly = delivery hygiene only; monthly = format/creative/audience/CPL; pipeline verdicts are quarterly only. A 30-day pipeline read in a 3-6 month cycle is not a finding.
- **Minimum sample before reading a change:** ~$100/creative, ~1,000 impressions/ad set, 7 days outside learning phase.
- **TLA follow/save data:** Ads Manager under-reports; the creator's native post analytics is the control source. Where they disagree, name both.

**Changed:**
- `SKILL.md`: version v2.2.2 → v2.3.0, reporting added to the sub-skill routing table, shared reference table, and routing rules (reporting = recurring for an audience; audit = one-off with a fix list).

**Platform orientation:** ZenABM MCP (`https://app.zenabm.com/api/mcp`) is named as the preferred data source, with the Campaign Manager CSV export as the standalone fallback. Deliberate change of position — the skill is platform-oriented on the ABM data layer, not platform-locked: every section works from a plain CSV export. Benchmarks remain the ZenABM 2026 dataset, as since v1.1.

**Sources:** existing `benchmarks.md` (ZenABM 2026, 161,256 ads / 211 companies) and LinkedIn Campaign Manager reporting field definitions — accessed 2026-08-31.

---

## v2.2.2 — 2026-07-22

**Summary:** QA sweep — removed competitor-agency references and traffic-driving blog links from the knowledge base. No strategy content changed.

**Changed:**
- `resources/linkedin-ads-knowledge-base.md`: removed a header that credited competitor agencies + their blog URLs, and deleted the "Section 10 — Sources & References" list of competitor-agency blog articles and courses. Replaced the header with a neutral GT + official-LinkedIn-docs provenance line.
- `CHANGELOG.md`: trimmed a bare vendor blog URL to the report name (ZenABM kept as a named SaaS source, no link).

**Policy:** name only SaaS products / official docs, never competitor agencies; never link out to competitor sites. Now 0 agency references and 0 external non-GT/official links.

---

## v2.2.1 — 2026-07-22

**Summary:** Housekeeping — removed named third-party agencies from two reference "Sources" lines (`predictive-audiences.md`, `accelerate-ai-campaigns.md`), leaving official docs + neutral "public best-practice guidance." Consistent with the competitor policy (name only SaaS products, never outside agencies). No strategy content changed.

---

## v2.2.0 — 2026-07-21

**Summary:** P2 coverage additions from the review — rigorous measurement, EU messaging compliance, and sharper ABM routing. No restructure.

**Added / expanded:**
- `measurement-attribution.md`: **Incrementality & Lift Testing** (native Brand Lift with its budget minimums + settle window; DIY conversion incrementality via holdout / geo-lift — LinkedIn has no robust self-serve conversion-lift) and **View-Through Conversions** (window guidance, report VTC separately from CTC, weight below clicks, sanity-check with incrementality). This is the causal answer to "CTR ≠ pipeline."
- `ad-formats.md`: **EU Sponsored-Messaging compliance** on Conversation Ads and Message Ads — blocked for EU members from Dec 2021 (ECJ/ePrivacy), consent-gated for opted-in EU members since ~Oct 2024; don't build an EU motion on these formats.

**Changed:**
- Tightened `abm-strategy` vs `ads-outbound-sync` descriptions + master routing to remove overlap: **abm-strategy = ABM planning** (structure, budget, formats, benchmarks); **ads-outbound-sync = ad-engagement → sales signals** (intent, BDR alerts, ZenABM/Fibbler sync). Clear "Do NOT use … (use the other)" boundaries.

**Note:** Conversation/Message Ads were already documented in `ad-formats.md` — v2.2.0 only adds the EU-compliance nuance (no duplicate format section).

**Sources:** LinkedIn Marketing Solutions Help (Brand Lift, Sponsored Messaging EU targeting), plus public incrementality-testing best-practice guidance — accessed 2026-07-21.

---

## v2.1.0 — 2026-07-21

**Summary:** P1 coverage additions from the critical review — the CRM-attribution loop and a structured account audit. Sub-skills now number 10.

**Added:**
- `resources/references/crm-attribution.md` — HubSpot 3-layer connection (native + CAPI + Insight Tag) with the offline-conversion loop on lifecycle-stage changes (~+30–50% SQL at same spend), Salesforce offline conversions + Campaign Influence for account-level attribution, the B2B U-shaped model default, and a 7–14d setup / 4–8wk learning timeline. Cross-linked to **gt-hubspot-admin** and **gt-salesforce-admin** for the CRM-side build. Wired into the `measurement` sub-skill.
- `.claude/skills/audit/gt-SKILL.md` + `resources/references/audit-checklist.md` — a GT-native LinkedIn Ads **audit** sub-skill with a 28-control catalog across tracking, targeting, creative, bidding/budget/structure, and outcomes/governance, built on evidence discipline (mark n/a/unknown; benchmarks are conditional; betas aren't failures). Brings triad parity with `gt-meta-ads` (which already had an audit). Audit-catalog pattern is common in paid-media audits (incl. MIT-licensed AgriciDaniel/claude-ads); implemented GT-native.

**Sources:** LinkedIn Marketing Solutions Help and HubSpot Knowledge Base (official docs), plus public B2B attribution best-practice guidance — accessed 2026-07-21.

---

## v2.0.1 — 2026-07-21

**Summary:** Removed all references to competitor services and re-anchored ABM benchmarks to the authoritative source. No GT strategy lost — the ABM guides were rewritten GT-native.

**Changed:**
- Rewrote `resources/references/abm/linkedin-ads-abm-guide.md` GT-native: dropped every reference to a competitor agency, its credit block, and its operational tables (team-requirements/FTE plan, tool-stack cost table). Benchmarks now point to `benchmarks.md` (ZenABM 2026) with an evidence-discipline caveat (check objective/geo/methodology/sample/maturity before applying any figure).
- Cleaned `resources/references/abm/ads-outbound-signaling-guide.md`: removed the "Source:" lines and the competitor credit block; content kept.
- Added a note on EU restrictions for Conversation/Message Ads.

**Policy:** competitor agencies/services are never referenced in GT skills; SaaS tools (ZenABM, HubSpot, Clay, Fibbler, etc.) remain named as products.

---

## v2.0.0 — 2026-07-21

**Summary:** MAJOR restructure into a master router + 9 sub-skills (multi-sub-skill layout), matching the pattern used by gt-linkedin-content and the CRM admin skills. Progressive disclosure — Claude loads only the relevant domain. No strategy content removed.

**Changed:**
- `SKILL.md` is now a **master router**: dynamic `SKILL_BASE` Setup block (Glob for `**/gt-linkedin-ads/SKILL.md`), an intent → sub-skill routing table, a shared-reference table, and the cross-cutting Key Benchmarks block.
- The 9 sub-skills moved from `resources/sub-skills/*.md` to `.claude/skills/<name>/gt-SKILL.md`: audiences, ads-outbound-sync, bidding, campaign-setup, copy, creative, measurement, optimization, abm-strategy.
- **Fixed a latent path bug:** sub-skills referenced `{SKILL_BASE}/references/...` but the files live under `resources/`; all references normalized to `{SKILL_BASE}/resources/references/...`.
- Enriched routing descriptions with the v1.3.0 topics (Predictive Audiences / career signals → audiences; Accelerate → campaign-setup; CAPI + Revenue Attribution Report → measurement).
- Corrected the master "Do NOT use" cross-links to `gt-linkedin-content` and `gt-linkedin-outbounder`.

**Shared resources** (`resources/references/`, `resources/references/abm/`, knowledge base) stay at the top level — script/relative paths and requirements unaffected.

**Validation:** all 9 sub-skills referenced exactly once; every `{SKILL_BASE}` reference resolves; all descriptions ≤1024; footers on every file; `.claude/` present in git; no stray files.

---

## v1.3.0 — 2026-07-21

**Summary:** P1 expansion from a July 2026 platform-research pass. Added the AI-era LinkedIn Ads capabilities the skill was missing (audiences, campaign automation, and attribution), grounded in current LinkedIn Marketing Solutions changes and 2026 benchmark data. Additive only — no existing strategy removed.

**Added:**
- `resources/references/predictive-audiences.md` — the Feb-2024 lookalike sunset, Predictive Audiences (30/account cap, 4-day processing, seed-quality rules), Audience Expansion vs predictive, and Career Journey / career-signal targeting.
- `resources/references/accelerate-ai-campaigns.md` — Accelerate AI campaign mode (GA for lead gen + website visits; video/document support; +42% CPA vendor claim), when to keep manual control (ABM, tight ICP), AI Ad Variants / Shutterstock creative with human-review guardrails, and real-time CRM data in Campaign Manager (June 2025).

**Expanded:**
- `resources/references/measurement-attribution.md` — new "2026 Deep Dive" section on the Conversions API (server-side, offline conversions, Salesforce/GTM/Tealium/Adobe; ~20% lower CPA / ~31% more attributed conversions; ~75% adoption) and the Revenue Attribution Report (CRM-connected, company-level, up to 365-day windows; engagement attributes ~7.7x more revenue).
- `resources/references/benchmarks.md` — 2026 Dreamdata attribution benchmark block.
- Sub-skills `audiences`, `campaign-setup`, `creative` now route to the new references.

**Sources:** LinkedIn Marketing Solutions Help (official docs), Dreamdata 2026 LinkedIn Ads B2B Benchmarks, and B2B marketing press (AdExchanger, Search Engine Land, Social Media Today) — accessed 2026-07-21.

---

## v1.2.1 — 2026-07-21

**Summary:** Repo reconciliation + attribution normalization. Brought the public repo copy in line with the org-deployed v1.2 and standardized every footer to the canonical Growth Today format. No strategy content changed.

**Changed:**
- Added the two ABM reference files that existed in the deployed build but were missing from the repo: `resources/references/abm/linkedin-ads-abm-guide.md` and `resources/references/abm/ads-outbound-signaling-guide.md` (external ABM source material; later rewritten GT-native in v2.0.1).
- `resources/sub-skills/abm-strategy.md`: added the ABM reference-routing block and fixed the resource path to `{SKILL_BASE}/resources/references/abm/...`.
- `SKILL.md`: added the two ABM reference files to the Reference Files table.
- All files: footer standardized to the canonical `Created by [Growth Today] … More open Claude Skills …` format (replacing the older "Visit and download more skills" line).

---

## v1.2 — 2026-06-25

**Summary:** Added new tactical content from a 2026 research pass (thought leadership ad guides, Metadata, and leading LinkedIn-ads agencies) and added a Growth Today attribution footer to every file. No content removed — only additions, plus the description and footer changes.

**Why:** The v1.1 refresh fixed the TLA *benchmarks* but the skill still treated TLAs as single-post tests, had no consolidated "default settings waste money" checklist, and gave no economic-fit rule for when TLAs are worth running. The research surfaced three repeatable, non-obvious tactics worth encoding, plus a stronger third-party cost proof point (Metadata) and updated 2026 cost figures.

**Changed:**

1. **SKILL.md**
   - Description: added "built by Growth Today (growthtoday.co)", "sequenced TLAs and Story Arcs", and a closing "More GT skills: growthtoday.co" for AI-search/SEO surface. No header attribution block (kept clean by design). Length 864 chars (under the 1024 org limit).
   - Version v1.1 → v1.2 with inline note.
   - Key Benchmarks: updated blended CPC to the 2026 $6-15 spread with format medians, added the Metadata $4.14-vs-$22.54 proof point and the 41%-of-B2B-budget / 80%-of-social-leads macro stats.
   - Added attribution footer.

2. **resources/sub-skills/creative.md**
   - New "Sequenced TLAs / Story Arcs" subsection: 5-7 post arcs, 4-6 variant minimum, deliberate ordering, tight fixed audience, 14+ day minimum.
   - New "Who TLAs are economically right for" subsection: ~$1M+ ARR + enterprise/ABM + active executive voice; smaller teams start with sponsored content.
   - New "TLA Anti-Patterns" section: hard-sell, over-broad targeting, single-post campaigns, killing too early, ignoring comments, click-only tracking.

3. **resources/references/troubleshooting.md**
   - New "0. Default Settings Audit" as the first troubleshooting step: Audience Expansion OFF, LinkedIn Audience Network OFF, ABM budget concentration check, frequency capping, manual/target-cost bidding over uncapped Maximum delivery.
   - Quick Diagnostic Checklist expanded with LAN, bid strategy, and ABM spread items.

4. **resources/references/benchmarks.md**
   - Cost Benchmarks table updated with 2026 CPC spread.
   - New "Cost proof points" block: Metadata experiment, channel-weight stats, and the $3-5K/month spend floor + learning-phase timing.

5. **resources/references/ad-formats.md**
   - Added a Story-Arc + economic-fit pointer after the 14 TLA strategies, cross-referencing creative.md.

6. **resources/linkedin-ads-knowledge-base.md**
   - TLA block: added Story-Arc and economic-fit lines so the KB matches the sub-skills.

7. **All files**
   - Added a footer: "Built and updated by Brigitta Ruha (LinkedIn link). Visit and download more skills on growthtoday.co."

**Not changed (intentionally):** the GT-internal finding that single static images outperform animations/GIFs on Brigi and Jani's profiles is a GT-specific learning and was left as-is; the generic "moving elements increase clicks" line in creative.md is kept as a general-market note, not a GT recommendation. All audience, bidding, campaign-setup, copy, measurement, optimization, abm-strategy, and ABM guide content unchanged beyond the footer.

**Sources:** ZenABM and Metadata.io (published LinkedIn-ads benchmark/cost data) plus public LinkedIn-ads best-practice sources, accessed 2026-06-25.

---

## v1.1 — 2026-06-11

**Summary:** Benchmark refresh to the ZenABM 2026 dataset (161,256 ads across 211 companies) and expanded Thought Leader Ads creative guidance. No content was removed — only outdated benchmark figures were updated and new data was added alongside the existing structure.

**Why:** The skill carried the old LinkedIn-reported "63% higher CTR" figure for Thought Leader Ads. The ZenABM 2026 report gives far more specific, format-level data and surfaces an important counter-insight (CTR does not predict pipeline). Keeping the skill on the old number risked steering ad strategy by click volume instead of pipeline.

**Changed:**

1. **SKILL.md → Key Benchmarks**
   - Added a per-format CTR/CPC table (TLA 2.68% / $2.29, Single Image 0.42% / $13.23, Video 0.24% / $15.61, Carousel 0.32% / $13.30, plus Event, Document, Dynamic, Text).
   - Added the "CTR does not predict pipeline" caveat (Spearman rho = -0.170) and the use-a-mix rule.
   - Refined the blended CTR line from "0.4-0.6%" to "0.44-0.65%" and added enterprise/SMB CPL spread.
   - Bumped version v1 → v1.1 with an inline note.

2. **resources/sub-skills/creative.md**
   - Format Selection table: TLA metric changed from "63% higher CTR (mid-funnel)" to "2.68% median CTR, $2.29 CPC (ZenABM 2026)".
   - Thought Leader Ads section: added the ZenABM performance figures, dwell time (6.63s), and the two top-performer creative rules (65% first-person "I" voice; 75% place the link in the bottom 25% of the post), plus the open-with-a-concrete-statement / ~800-1,200 char guidance and the CTR-vs-pipeline caveat.
   - Example 2: updated the benchmark quoted in the answer.

3. **resources/references/benchmarks.md**
   - Replaced the CTR table with per-format ZenABM 2026 medians.
   - Added CTR-by-geography (NL 0.72% / UK 0.55% / US 0.52%), dwell-time-by-format, and a "good CTR by campaign purpose" table.
   - Added the CTR-vs-pipeline negative-correlation note.
   - Added a CPL-by-ICP-tier table (SMB $80-200 / mid $150-450 / enterprise $400-800).

4. **resources/references/ad-formats.md**
   - Thought Leader Ads "Stats" line updated from "63% higher CTR on mid-funnel content" to the ZenABM 2026 figures + CTR-vs-pipeline caveat. (Caught in QA — kept consistent with the rest of the skill.)

5. **resources/linkedin-ads-knowledge-base.md**
   - Same TLA "Stats" update as above, so the knowledge base doesn't contradict the sub-skills. (Caught in QA.)

**Not changed (intentionally):** all sub-skills other than creative, all other reference files (ad-formats, creative-strategy, funnel-architecture, targeting-audiences, bidding-objectives, measurement-attribution, troubleshooting, competitive-research, both ABM guides), and the full knowledge base. Structure and strategy were already current; only benchmark numbers and TLA creative rules needed the 2026 refresh.

**Source:** ZenABM 2026 LinkedIn ABM Benchmarks Report (ZenABM is a SaaS tool), accessed 2026-06-11.

---

## v1 — baseline

Initial GT LinkedIn Ads skill: orchestrator + 9 sub-skills (audiences, ads-outbound-sync, bidding, campaign-setup, copy, creative, measurement, optimization, abm-strategy), reference files, ABM guides, and full knowledge base. Built on $25M+ in managed B2B ad spend.

---

*Created by [Growth Today](https://www.growthtoday.co) — AI-native GTM engineering firm. Maintained and updated by [Brigitta Ruha](https://www.linkedin.com/in/brigittaruha/). More open Claude Skills for go-to-market teams: https://www.growthtoday.co/claude-skills*
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall