bi-architecture

Designs BI architectures covering KPI frameworks, semantic layers, dashboard hierarchies, and self-service analytics governance.

Updated Jun 28, 2026
One-click install
npx skills add https://github.com/JaviMontano/claude-plugins --skill bi-architecture-javimontano
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: bi-architecture
Source: https://github.com/JaviMontano/claude-plugins/tree/main/plugins/claude-native-toolkit/skills/bi-architecture
Command: npx skills add https://github.com/JaviMontano/claude-plugins --skill bi-architecture-javimontano

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) and assets (resource) components.

What problem does it solve? Organizations struggle with inconsistent metric definitions, dashboard sprawl, and ungoverned self-service analytics. This Skill produces a complete BI architecture document that unifies KPI definitions, designs a semantic layer, structures dashboard hierarchies, and establishes governance so teams consume trustworthy, consistent analytics. ## Core Features & Use Cases - KPI & Metric Frameworks: Decompose north star metrics into supporting and input metrics with ownership, formulas, targets, and leading/lagging classification. - Semantic Layer Design: Compare and select headless BI options (Cube, dbt MetricFlow, LookML, AtScale) and enforce single metric definitions across tools. - Dashboard Hierarchy & Governance: Define L1-L4 dashboard tiers with performance budgets, refresh SLAs, lifecycle management, and platform evaluation matrices (Looker, Power BI, Tableau, Superset, Metabase). - Use Case: A company with 87 ungoverned dashboards and 23 conflicting definitions of "revenue" uses this Skill to consolidate metrics into a Cube semantic layer, rationalize dashboards into a certified 4-tier hierarchy, and roll out governed self-service zones for 150 analysts. ## Quick Start Ask the assistant to design a BI architecture for your organization, including a KPI metric tree, semantic layer selection, dashboard hierarchy, and self-service governance plan.

Frequently Asked Questions about bi-architecture

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I design a KPI framework and metric tree?▼

Start from a north star metric and decompose it into supporting and input metrics with clear ownership. Each metric gets a unique name, formula, grain, dimensions, and owner, with 5-7 KPIs per department and 15-25 supporting metrics.

What semantic layer tool should I choose for BI?▼

Choose based on your stack: dbt MetricFlow for dbt-centric teams wanting metrics-as-code, Cube for tool-agnostic API serving across multiple BI tools, LookML for Looker-committed organizations, and AtScale for large enterprises with heavy Excel usage.

How do I enable self-service analytics without losing governance?▼

Implement three zones: a certified zone with locked metric definitions for all users, an exploratory sandbox with compute quotas for analysts, and a raw zone restricted to data engineers. Label content clearly and reserve certified content for executive decisions.

Does this cover dbt transformations or data pipeline design?▼

No. This Skill focuses on analytics consumption, not upstream engineering. dbt modeling and SQL transformations belong to analytics-engineering, pipeline orchestration to data-engineering, and upstream data quality rules to data-quality skills.

What performance targets should BI dashboards meet?▼

The recommended budget is initial render under 2 seconds, interactive queries under 5 seconds, and filter response under 1 second. Freshness SLAs vary by tier: daily for executive dashboards, hourly for departmental, and near-real-time for operational views.

How do I prevent dashboard sprawl in my organization?▼

Manage dashboards as products with a lifecycle: request, build, review, publish, monitor, and archive. Run quarterly audits and archive dashboards with zero views in 90 days, targeting 5-15% archival churn per quarter.