Agent skill · data analytics · aradotso
marketing-analytics-lab-engine
Multi-platform marketing analytics engine with canonical data model for unified campaign data ingestion and KPI analysis
Why this skill is useful
Provides a structured approach to ingesting and validating marketing campaign data across multiple platforms, which the AI wouldn't reliably generate on its own.
What it needs
Requires pandas, pydantic, pytest, python installed locally. About 6k tokens when loaded. Last updated 2026-08-05. 8 stars on the source repository.
What this skill does
Marketing Analytics Lab Engine Skill by ara.so — Marketing Skills collection. Marketing Analytics Lab is a Python-based marketing data ingestion engine that enforces a canonical schema across multiple advertising platforms (Google Ads, Meta Ads, etc.). It uses an adapter pattern to transform platform-specific CSV exports into a unified CampaignRecord format with strict Pydantic validation before any analytics processing. Installation Core dependencies: Python 3.11+ pandas pydantic v2 pytest (for testing) Project Architecture The system follows a strict data pipeline: Key components: CampaignRecord: Canonical data model (single source of truth) BaseCampaignAdapter: Abstract adapter contract UnifiedLoader: Source-aware entry point that routes to correct adapter Validation layer: Pydantic-based contract enforcement Core Data Model All marketing data must conform to the CampaignRecord canonical schema: Pydantic validation schema: Loading Campaign Data Basic Usage Supported Sources Currently implemented: googleads: Google Ads CSV exports In progress: metaads: Meta (Facebook/Instagram) Ads Creating a New Platform Adapter To add support for a new marketing platform: 1. Create Adapter Class 2. Register in UnifiedLoader 3. Test the Adapter Validation Validate Individual Records Batch Validation Working with Campaign Data Calculate Basic KPIs Aggregate by Date Compare Platforms CSV Format Requirements Google Ads Expected Format Column mapping: Date → date Campaign ID → platformcampaignid Campaign → campaignname Impressions → impressions Clicks → clicks Cost → cost Conversions → conversions Conv. value → revenue Testing Run All Tests Run Specific Test Test Coverage Common Patterns End-to-End Pipeline Error Handling Troubleshooting Issue: "Unsupported source" error Problem: UnifiedLoader doesn't recognize the source name. Solution: Ensure source name matches exactly (case-sensitive): Issue: Validation fails with date format error Problem: Date format doesn't match YYYY-MM-DD. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills aradotso/marketing-analytics-lab-engine