What happens when a data contract is breached?

I wrote earlier this month that data contracts shouldn’t focus on enforcement.

By which I meant, the outcome you’re optimising for isn’t enforcing rules on someones data, but instead using data contracts to facilitate a better quality dataset that allows others to build on it with confidence.

But clearly, once you start implementing data contracts, you’re going to implement a way of enforcing them!

So, what happens if that data contract is “breached”? What happens when the data doesn’t match the contract?

Firstly, the producer of the data should be the one that is alerted. They should own the incident, and be responsible for resolving it.

Those consuming the data are interested parties in that incident, and should receive regular updates on the resolution.

The aim should be to resolve the impact as quickly as possible whilst limiting its impact.

Once the incident is resolved you should follow as postmortem process to understand what went wrong and what can be done to prevent a similar incident happening in the future.

These “breaches” of the contract shouldn’t be unexpected!

These are complex systems and things will go wrong.

We accept that, assume everyone is working with the best intentions, and learn from our mistakes.

With data contracts, we’re assigning responsibility for these incidents to the data producers, as are the only ones able to fix the issue at source.

Which is the only place we can improve the quality of the data.