Case 02 — Enterprise Revenue Trust Crisis
Situation
The board sees three different values for Net Revenue: - Finance reports from the ERP; - Sales uses a CRM-derived dashboard; - the enterprise warehouse publishes an executive KPI.
Investigation finds: - each group uses a different business definition of Net Revenue; - the glossary contains an old, unapproved definition; - the warehouse calculation mixes invoice and order data at different grains; - refunds arrive one day later than sales transactions; - one mart maps cancelled orders differently from the enterprise warehouse; - Product categories differ between Sales and Finance marts; - source-to-target transformations are poorly documented; - users cannot trace the KPI back to the originating transactions; - executives demand “one SQL formula by Friday” and suggest assigning the BI developer as owner of the metric.
Your task
Use the ten-layer worksheet and produce an ordered response that separates: - authoritative metric definition and accountability; - Business Glossary meaning vs Data Dictionary technical detail; - fact grain and dimensional conformance; - mapping vs transformation vs remediation; - lineage vs discovery/profiling; - timeliness/latency issues vs semantic inconsistency; - Data Steward/Data Owner roles vs BI implementation roles.
Minimum retrieval requirements
State closed-book: 1. which decision must occur before a durable SQL fix; 2. why a BI developer should not become the business owner merely because the KPI is implemented in BI; 3. why fact grain must be resolved before comparing measures; 4. what evidence would distinguish a late refund feed from a wrong Revenue definition; 5. how a conformed Product dimension helps cross-mart analysis; 6. what Metadata is needed so an executive can trace the KPI from report to source.