A newsletter on building governed master data, semantic layers and roadmaps so that business intelligence and AI deliver dependable answers. Each issue answers a practitioner question with a clear opinion, cited sources and a distinction between forecast and recommendation.

Kevin Holland writes about healthcare data and AI products, arguing that teams should earn each commitment through evidence about the decision a product supports rather than building polished demos. His issues walk through staged product gates, the risks of cheap AI-generated dashboards, and lessons from a costly healthcare dashboard project.

Practical technical posts on why AI agents querying replicated enterprise data (Lightspeed, Square, PeopleSoft) return wrong totals when naive column names are trusted. Each issue walks through a concrete data-modelling trap and how to query around it.

Peter Nicholson writes practical, opinionated essays on AI, ERP systems and data, drawing on his manufacturing sales background and hands-on migration and Power BI work. Each issue argues a specific point, such as where human approval adds little value in automated ERP processes or why AI answers can hide conflicting business definitions.