Every engagement starts with the data model.
This is the Data First position. Not a preference — a principle derived from consistent observation. The most significant technical problems in most organisations trace back to data that was never properly defined, structured, or governed.
The data model determines what the architecture can do. The data governance framework determines whether the data can be trusted. The data quality standards determine whether the reports, integrations, and decisions built on top of it are reliable.
Getting data right is not preliminary work. It is the work.
What Data First means in practice
The data model is established before architecture decisions are made. Not after. The data model determines what the architecture needs to support — how data flows, where it is stored, how it is accessed, and what operations it must support. Architecture designed without a clear data model inherits the model’s ambiguities at a structural level.
Data ownership is explicit. Every material data asset has a named owner — a person accountable for its quality, accessibility, and governance. Not a team. Not a system. A person.
Data definitions are authoritative. The same field means the same thing in every system that uses it. Where definitions conflict, the conflict is resolved and the resolution is authoritative. Multiple definitions of the same concept are a governance failure.
Data quality is enforced, not assumed. Standards define what correct data looks like. Systems enforce those standards. Data that does not meet quality standards is not processed as if it does.
The commercial consequence
Organisations that have not built data discipline have data they cannot trust. Reports that require manual verification. Integrations that require reconciliation. Compliance demonstrations that require manual assembly of data that should be available immediately.
The cost is real. It is distributed across every process that touches data — which is most processes.
Start a Conversation