Reducing the difficulty of being a data owner

Hey friends 👋

This week I write about reducing the difficulty of being a data owner.

Also links to articles on data contracts at VMO2, metadata as common language, and BlaBlaCars data copilot.

Finally, a snowman pun.

Enjoy!

Reducing the difficulty of being a data owner

It’s often difficult to find someone who wants to own data.

That’s because it’s difficult to be a data owner.

You become responsible for the management of that data, ensuring:

• Observability and alerts are set up
• The right people have the right amount of access
• Any personal data is deleted/anonymised when retention periods expire
• Change management is correctly applied as the data evolves
• Keeping accurate and tested backups

And so on.

This can be a lot of work - work outside of your day job - and therefore work that is often not completed. That then increases the data management risks you have, for example for data misuse or data breaches.

If we want to assign ownership and responsibility of data we need to make it easy to be a data owner.

We do this by automating many of these tasks through the data platform, driven by data contracts.

For example, within the data contract the data owner can specify fields that contain personal data. Given that context, the platform can automate the anonymisation of that personal data when its retention period has expired.

The data contract is providing the context that allows the platform to do the right thing.

That’s just one example, but I’ve seen each of those pain points automated through data contracts. In fact, the automation is quite easy to implement, once you have that context.

All of this helps reduce the difficulty of being a data owner, making it much more likely you will find data owners across your organisation.

I spoke more about this, and various other data contract-related topics, in my webinar with Nick Zervoudis earlier this week. Check out the recording on YouTube.

Interesting links

VMO2 uses data contracts to build scalable AI and data products by Chandu Bhuman and Dženan Softić (Google Cloud blog)

Great example of data contracts in the wild, using them to automate the running of data quality checks, amongst other things.

Metadata as Common Language: An Onboarding Guide to Breaking Silos by Gaëlle Seret

This really is the power of metadata.

Why We Built BlaBlaCar Data Copilot: Shifting Data Analysis Left by Thomas Pocreau (BlaBlaCar blog)

An interesting solution, but I particularly like some of their assertions, including:

The barrier isn’t a lack of skill; it is the friction of unfamiliar tooling, the fear of “doing it wrong” without a safety net, and the organizational silos that make it difficult to cross the line between engineering and data teams.

And:

The greatest value doesn’t always come from being the absolute best Java developer or the most advanced statistician. Often, the most impactful innovations happen at the interface: the messy, complex boundary between Software Engineering and Data Analysis.

Also the solution is a great example of meeting the users where they are, which is something I touched on a few weeks ago.

Being punny 😅

Have you seen the world’s best snowman? It’s outstanding.

Thanks! If you’d like to support my work…

Thanks for reading this weeks newsletter — always appreciated!

If you’d like to support my work consider buying my book, Driving Data Quality with Data Contracts, or if you have it already please leave a review on Amazon.

🆕 Also check out my self-paced course on Implementing Data Contracts.

Enjoy your weekend.

Andrew