By Stephanie Simone
</span>
<span itemprop=”>
Effective data modeling enables value creation, efficiency gains, risk reduction, and strategic alignment in an environment of uncertainty and disruption.
At Data Summit 2026, Pascal Desmarets,CEO,Hackolade, led the pre-conference workshop, “From Strategy to Structure: Accelerating Business Impact Through Hands-on Data Modeling,” moving from high-level strategy to concrete data models that power modern analytics platforms.
The annualData Summitconference returned to Boston, May 6-7, 2026, with pre-conference workshops on May 5.
Desmarets’ workshop picked up where the previous one, featuring John O’Brien, principal advisor and industry analyst at Radiant Advisors, left off.
“We’re going to align with it and then we’re going to build up,” Desmarets said. “This is about how to get things done to do data modeling.”
Participants explored practical techniques to connect strategy, architecture, and analytics so that data investments consistently deliver business value.
“Nowadays, and in particular with the emergence of AI, we’ve seen a lot of interest in data modeling,” Desmarets said.
Data modeling matters to the business in the following ways:
- Value creation: Data modeling transforms raw data into a structured, strategic asset that drives better decisions, fuels innovation, and unlocks new revenue streams.
- Risk reduction: Data modeling reduces risk by enforcing governance, auditability, and data consistency—ensuring compliance, protecting reputation, and preventing costly errors and fines.
- Cultural transformation: Data modeling drives cultural transformation by enabling a data-driven mindset, building trust in reliable insights, and establishing clear accountability for data and outcomes.
- Efficiency gains: Data modeling streamlines operations by optimizing resources, reducing costs, and accelerating time-to-market through better data quality, integration, and automation.
- Strategic alignment: Data modeling aligns data with business strategy, creating a scalable foundation that supports long-term goals, digital transformation, and sustained competitiveness.
- Strategic business enabler: Data modeling is not just a technical discipline. It delivers measurable value by improving performance, efficiency, compliance, strategy execution, and organizational culture, making it a cornerstone of sustainable, data-driven transformation.
Most organizations stop at strategy or architecture diagrams. But business impact only happens when those ideas are translated into data structures that systems can execute, he said.
There are three phases/layers of data modeling that include conceptual graph modeling, logical polyglot modeling, and physical modeling, Desmarets explained.
“Logical modeling is important to understand, identify, and define business rules,” Desmarets said. “Graph modeling, logical modeling and physical modeling are important phases because they help you identify things from the top-down.”
AI has upended the traditional way to data model, he explained. Now, trust is the most important thing to establish. A semantic layer provides meaning and context so AI can understand the data it is fed.
According to Desmarets, one data modeling method includes:
- Identify Entities: Business nouns that matter—things the business tracks, measures, or manages.
- Define Attributes Properties of each entity: Classify: identifier, descriptor, measure, foreign key. Consider data types, constraints, and governance tags.
- Model Relationships: Connect entities with cardinality. Consider normalization vs. denormalization trade-offs for the target analytics workload.
- Apply Governance: Tag sensitive attributes (PII, PCI). Define data quality rules and ownership. Link to the business glossary.
- Forward-Engineer: Generate DDL for Snowflake/BigQuery, JSON Schema for APIs, Avro for streaming. Validate against the target platform requirements.
AI-assisted data modeling always requires a human in the loop, he concluded. Intelligent assistance accelerates the expert, AI just removes the tedium.
Many Data Summit 2026 presentations are available for review at https://www.dbta.com/datasummit/2026/presentations.aspx.