Data Contracts

Getting started with data contracts doesn’t have to be complex. Here’s 3 steps:

1. Run some proof of concepts

This doesn’t have to be on the most on the most critical dataset you have.

An ounce of prevention is worth a pound of cure.

Or put another way, the earlier you manage something, the cheaper it is.

Data is published by many different systems, and as the number of systems increase it’s likely your organisations data will become less standardised over time, with different identifiers, different semantics, and so on.

It’s important to celebrate small wins.

We recently decommissioned a small utility service called subcontractor, which managed some of the resources we create from our data contracts.

Many data platforms start with a change data capture (CDC) service to extract data from an organisations transactional databases — the source of truth for their most valuable data.

Data contracts are typically used to apply change management to data in tables, in a data warehouse.

Implementing data contracts at the platform layer gives it the capabilities needed to enable domain ownership.

Data Contracts are about setting expectations, including:

• How the data is structured, and how that can or cannot change
• Who owns the data, and the support they are providing
• When the data is updated, and how complete it is

And so on.

Software engineers use processes and tooling to instil discipline into the creation of software.

They:

There are 2 parts to implementing data contracts.

One is the tooling that allows your data producers to provide data through an interface. That’s where you implement change management and other data governance policies.