Business Intelligence Guidebook: From Data Integration to Analytics
One book, placed in its field — what Rick Sherman's book gets right, where it goes further than the rest, and what the rest of the field adds.
Mike West, Editor · Leading With Data · dataengineerprofessional.com/guides/profile-business_intelligence_guidebook_from_data_integration
Arrow keys or swipe to move · N for notes · F for full screen
Business Intelligence Guidebook: From Data Integration to Analytics
One book, placed in its field — what Rick Sherman's book gets right, where it goes further than the rest, and what the rest of the field adds.
Mike West, Editor · Leading With Data · dataengineerprofessional.com/guides/profile-business_intelligence_guidebook_from_data_integration
The problem
Business Value and ROI erodes when it is left to instinct instead of method.
You were taught the moves piecemeal, never the whole model.
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
The model
- 01Architectural Discipline
- 02Data Integration Rigor
- 03Dimensional Modeling Quality
- 04Tool-Based Development Adoption
- 05Requirements and Expectation Management
- 06Training Investment
- 07Organizational Governance and Sponsorship
- 08Information Quality (Five Cs)
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
Step 1 of 5 · Architectural Discipline
Sequence architecture as information → data → technology → product; reversing it creates the silos you will later pay to dismantle.
Misconception
Practitioners believe the reference architecture the vendor ships with their platform is their architecture.
Reality
A vendor architecture optimizes for that vendor's product footprint; your architecture must start from your information and data requirements and treat product selection as the last, not first, layer.
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
Step 2 of 5 · Data Integration Rigor
Integration rigor is the strongest predictor of information quality — invest here before investing in front-end analytics.
Misconception
Teams treat integration as plumbing — a one-time job of moving data from A to B that any developer can script.
Reality
Integration logic encodes business rules, and hand-coded extracts bury those rules in undocumented scripts that no one can audit, reuse, or safely change when sources shift.
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
Step 3 of 5 · Dimensional Modeling Quality
Conformed dimensions are what make numbers comparable across the enterprise — treat them as shared infrastructure.
Misconception
Modelers assume that if the source system is normalized and correct, a copy of it will serve analytics equally well.
Reality
Operational models optimize for transaction integrity, not analysis; without conformed dimensions and surrogate keys, you cannot compare subject areas or preserve history the way decision-makers need.
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
Step 4 of 5 · Tool-Based Development Adoption
Tool fit is a three-way function of needs, skills, and budget — a great tool the team can't use is a bad choice.
Misconception
Senior developers believe hand-written code is more flexible and cheaper than buying a tool.
Reality
Hand-coded solutions win on day one and lose over their lifetime — they lack the built-in lineage, scheduling, and metadata that tools provide and that governance later demands.
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
Step 5 of 5 · Requirements and Expectation Management
Requirements and expectations are two jobs — one defines the work, the other defines how the work will be judged.
Misconception
Teams treat requirements gathering as a one-time upfront phase and assume expectations take care of themselves once requirements are signed off.
Reality
Expectations drift continuously as stakeholders learn what's possible, so managing them is an ongoing communication practice, not an artifact you collect and file.
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
A framework to use
BI Architectural Framework
This framework is part of the full guide. See the guide →
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
The cheat sheet
- Architectural DisciplineSequence architecture as information → data → technology → product; reversing it creates the silos you will later pay to dismantle.
- Data Integration RigorIntegration rigor is the strongest predictor of information quality — invest here before investing in front-end analytics.
- Dimensional Modeling QualityConformed dimensions are what make numbers comparable across the enterprise — treat them as shared infrastructure.
- Tool-Based Development AdoptionTool fit is a three-way function of needs, skills, and budget — a great tool the team can't use is a bad choice.
- Requirements and Expectation ManagementRequirements and expectations are two jobs — one defines the work, the other defines how the work will be judged.
- Training InvestmentThree distinct training layers exist — foundational, tool-specific, and use-case — and skipping the first two guarantees shallow adoption.
- Organizational Governance and SponsorshipGovernance manages politics and process, not just data standards — its hardest work is arbitration.
- Information Quality (Five Cs)Measure all five Cs independently; a high score on one masks fatal gaps in another.
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data
Sources
- Business Intelligence Guidebook: From Data Integration to Analytics — Rick Sherman
Business Intelligence Guidebook: From Data Integration to Analytics · Mike West, Editor · Leading With Data