Ch 16 - Be Water. The Continuous Practice of Data Modeling

2026-04-26 · Source: Practical Data Modeling · Field: Technology & Digital — Data Science & Analytics, Software Development & Engineering, Artificial Intelligence & Machine Learning · Depth: Intermediate, short

Summary

Chapter 16 of "Mixed Model Arts, Book One" emphasizes the critical need for continuous data modeling, reframing it from a finite project to an ongoing program. The author highlights that, much like evolving combat sports, data modeling is never truly "done" due to shifting requirements, technological advancements, and business evolution. The chapter introduces the concept of the "Done" Delusion, challenging the common perception of data initiatives as having a definitive end. It advocates for treating data modeling as a perpetual business function, akin to finance or operations, requiring continuous sense-making and adaptation. A key issue discussed is the "Ownership Problem," where diffuse data ownership leads to a lack of accountability, underscoring the necessity for dedicated data stewards and clear governance to maintain model alignment with organizational reality.

Key takeaway

For Data Engineers and Data Architects designing data systems, recognize that data modeling is a continuous program, not a project with a "done" state. You should establish clear ownership for data models and integrate continuous evolution into your workflows to ensure models remain aligned with the business, rather than only fixing them when they break.

Key insights

Data modeling should be a continuous program, not a finite project, adapting to evolving business realities.

Principles

• Data models are living reflections of the business.
• Continuous practice is essential for data modeling.
• Diffuse ownership hinders data accountability.

Method

Treat data modeling as a continuous program with dedicated stewards and clear governance, rather than a one-time project, to ensure ongoing alignment with evolving business needs.

In practice

• Implement continuous data model evolution.
• Assign clear data model ownership.
• Foster modeling literacy across data teams.

Topics

• Data Modeling
• Continuous Practice
• Program Management
• Data Ownership
• Data Governance

Best for: Data Engineer, Analytics Engineer, Data Scientist

Related on AIssential

• No Free Lunch: The Debt, The Excuses, and The Reality of Data Modeling — Intentional data modeling is critical for agility, preventing chaos, and aligning data with business processes, regardless of company size or AI advancements.
• Building a Data Mesh in the Real World — Real-world Data Mesh implementation requires pragmatic adaptation of theoretical principles to organizational realities.
• MLOps vs DevOps: A Practical Guide for Data Scientists and IT Teams — MLOps extends DevOps to manage the unique lifecycle of ML models, data, and continuous training.
• Your LLM issues are really data issues — Inconsistent data and weak governance undermine analytics and ML, necessitating AI-ready data practices.
• From Models to Momentum: Uniting Architects and Engineers with ER/Studio — Enterprise data modeling provides a semantic backbone, crucial for data reliability, governance, and effective AI integration.
• Data Modeling is Dead (Again), 2026 Edition. Part 1 — New AI capabilities are fueling a recurring narrative that traditional data modeling is obsolete.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Practical Data Modeling.