Influencer Campaign Performance Analyzer
Turn creator campaign data into benchmarked scorecards, attribution findings, and prioritized next actions.
The skill declares Tier 1 operation, treats supplied reports and fetched content as inputs, and describes the YouTube connector as read-only. However, GA, sales, comments, and screenshots may contain sensitive data, while memory writes and promotion of durable facts lack explicit minimization, user-confirmation, rollback, or attribution-review requirements. Main risks are visible, but authorization, privacy, and recovery boundaries are incomplete, so points are deducted.
The ordered method, input/output contract, templates, and significance gate are fairly complete, and missing information is generally requested from the user. Static review cannot reproduce the key paths, there are no skill-specific tests, and handling for malformed inputs, missing fields, or attribution conflicts is thin. The claim of no live integrations also sits somewhat awkwardly with the optional YouTube script. The score stays below the static ceiling of 10.
Triggers, audience, inputs, outputs, and the non-fit boundary of excluding dollar-level ROI are fairly clear, with Chinese-language labeling present. However, practical input boundaries for platform coverage, attribution modes, and sentiment analysis are underspecified; geo-relevance is marked low, and regional reachability of platform exports or YouTube API access is not addressed. The scenario fit is plausible but boundary and environment evidence are limited.
The skill includes frontmatter, versioning, Apache-2.0 licensing, Quick Start prompts, a contract, numbered instructions, templates, examples, handoff guidance, and next-skill routing. Maintenance ownership, changelog practice, and an update path are not clear in the skill, and SECURITY.md describes version 17.x as current while the skill declares 18.0.0, creating a governance inconsistency. The unverified publisher is not deducted by itself, but incomplete governance lowers the score.
The skill covers metrics, platform and creator comparisons, content patterns, engagement quality, sentiment, conversion attribution, and ranked learnings, with fill-in templates that can support a usable report. However, the example includes ROI and budget-shift recommendations despite the stated exclusion of dollar-level ROI, and the output depends heavily on user-prepared multi-source data and manual judgment. Under static calibration it cannot exceed 7, and direct usability and correctness remain unverified.
The skill provides source labels such as Measured and User-provided, refers to a significance protocol and benchmarks, and supplies auditable report structures. This review is file-only; there is no skill-specific CI, key-path test coverage, or independently verifiable third-party execution evidence. The score therefore reflects limited static traceability only.
- Redact sensitive or personal data in GA, sales, comments, and screenshots before submission, and manually review facts before memory promotion.
- Do not treat the example ROI, budget-shift recommendation, or winner conclusions as verified facts; confirm targets, benchmarks, attribution windows, and significance first.
- The YouTube connector requires a user-supplied YOUTUBE_API_KEY; platform reachability, regional restrictions, and data latency may affect results.
- Note the inconsistency between the skill's 18.0.0 version and SECURITY.md's stated current supported 17.x line.
What does this skill do, and when should you use it?
Performance Analyzer is designed for mid-flight and post-campaign influencer reviews. It scores core results against targets and benchmarks, then ranks platforms, creators, and content. It also reads engagement quality, comment sentiment, and conversion attribution before producing ranked learnings. It does not perform dollar-level return math; that work belongs to roi-calculator.
It reads a campaign name and date range, native platform analytics, influencer reports or screenshots, website and conversion data, sales and promo-code data, available targets and benchmarks, and creator baselines when present. It scores reach, impressions, engagement rate, video views, clicks, promo uses, conversions, and revenue; compares platforms and formats; ranks influencers and content; evaluates engagement intent, sentiment, and purchase signals; attributes conversions through promo codes, UTMs, direct attribution, or estimation; and writes an analysis containing scorecards, rankings, attribution, and three to five learnings. Durable findings may be promoted to the repository's specified memory locations.
- A marketing lead needs a mid-campaign read on which platforms and creators deserve continued investment.
- A brand team has finished an influencer campaign and wants to identify the strongest creators and content formats.
- A growth team has platform exports, analytics data, and promo-code redemptions and needs an attribution breakdown.
- A campaign manager wants to distinguish meaningful engagement and purchase intent from surface-level interaction.
- A planning team wants to carry proven creator, format, and platform patterns into future benchmarks.
What are this skill's strengths and limitations?
- Covers core scoring, platform and creator rankings, content analysis, engagement quality, sentiment, attribution, and recommendations.
- Runs at Tier 1 using data supplied by the user, without requiring live integrations.
- Keeps Measured, User-provided, and Estimated evidence labels distinct.
- Uses a significance bar and Keep-testing status when evidence is insufficient.
- Hands results forward to roi-calculator, report-generator, and fit-scorer.
- Does not calculate dollar-level ROI, cost metrics, or payback; roi-calculator is required for that.
- Results depend on the completeness and quality of supplied platform, web, sales, target, and benchmark data.
- Apart from the optional YouTube connector, the source does not show automatic retrieval for specific social platforms.
- The source does not provide independent platform test results or a complete standalone fixture for this skill.
How do you install this skill?
In Claude Code, run /plugin marketplace add aaron-he-zhu/aaron-marketing-skills, then /plugin install aaron-marketing@aaron. On an Agent Skills-compatible host, run npx skills add aaron-he-zhu/aaron-marketing-skills; the repository can also be installed with git clone https://github.com/aaron-he-zhu/aaron-marketing-skills. The README documents a single-skill installation syntax, but does not provide separate platform-by-platform validation for this individual skill.
How do you use this skill?
Use Analyze performance of [campaign name] influencer campaign, or compare creators with Compare performance of these influencers from [campaign]: @handle1, @handle2, @handle3. Provide the campaign period, platform analytics, creator reports, web or sales data, and any targets or benchmarks. The skill should ask for missing inputs. For YouTube campaigns, the optional connector command is python3 "${CLAUDE_PLUGIN_ROOT}/scripts/connectors/youtube.py" videos @creator --limit 20.
How does this skill compare with similar options?
The source defines a clear boundary with roi-calculator: Performance Analyzer handles performance scoring, rankings, attribution, and learnings, while roi-calculator handles dollar-level return math. Compared with report-generator, this skill performs the analysis; report-generator packages an analysis into a formal stakeholder report.