Implementing domain ownership with data contracts

Hey all, hope this weeks been a good one.

This is part 2 of my 5 part series on implementing data mesh. See part 1 here.

Also links on moving from outputs to outcomes, understanding the business context, and an Excel interface for the open data contract standard.

Implementing domain ownership with data contracts

Last week I wrote about how, when it comes to data mesh, it’s better to consider each of its principles as guides, which you can choose to implement in part or in whole to solve some of the problems you have in your organisation.

Today I’m going to look at the principle of domain ownership.

This principle is worth considering for central data teams that have become a bottleneck, as data owners and providers for the organisation, who you have more requests coming in than they could ever handle.

Having a central team as dependency for any use of data limits the application of data in your organisation, as only the most strategic and high priority requests will get resources from the data team.

But that leaves many good, impactful data initiatives slowed or blocked.

With domain ownership we empower domains to own and make use of their own data, so they can autonomously perform analysis, gain insights, and build data applications.

We also want domains to have the ability to quickly and easily publish their data to other parts of the organisation who could make use of that data - again, without the need to go through an over-subscribed central data team.

It’s here where data contracts can help with the implementation of domain ownership.

Whenever data is being made available to another team/group/domain it needs to be made clear who owns it, what the data is, and what the expectations are.

These can all be captured in the data contract.

In fact, it’s at these boundaries of ownership that data contracts are most useful, because it is here where ambiguity and assumptions can limit the trust and confidence someone has when they are deciding whether to make use of the data.

To fully enable domain ownership of data there is much more you need to do, such as providing the right platform and ensuring the data in these domains remain well governed. I’ll cover this in more detail later in the series.

For now, consider how much of a bottleneck your central data team is and how that impacts the speed and scale of your organisations data use, and how by adopting some of the ideas from data mesh’s domain ownership principle you can start to remove this bottleneck.

Interesting links

From Output to Outcome: Why Defining Outcomes in Data & AI Products Is So Challenging by Anne Claire Baschet and Yoann Benoit

One challenge in Data & AI is that we often confuse the asset—transformed data, algorithm, KPI—with the user’s actual experience.

Here’s the secret sauce:

Outcome = (Asset: Data or AI) × (User Experience)

Great post.

Understanding Business Context by Ergest Xheblati

The greater your understanding of the business, the greater the value you can deliver through the use of data.

datacontract/open-data-contract-standard-excel-template: Edit Open Data Contract Standard in Excel

When thinking about implementing data contracts, you need to think carefully about how users will create and manage their data contracts.

Some users might want to define them as YAML. Some might want to define them as code. Some might want to define them in Excel.

All are equally valid! Like the link above, it’s about focussing on the user experience to enable adoption from those users.

Being punny 😅

I was sitting drinking tea in my slippers this morning, when I thought to myself… I really need to wash some mugs.

Upcoming workshops

• Implementing a Data Mesh with Data Contracts - Antwerp, Belgium - June 5
  • Alongside the inaugural Data Mesh Live conference, where I’ll also be speaking.
  • ​Sign up here​

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.

Enjoy your weekend.

Andrew