Data Observability Demystified - A D/I O11y Framework Perspective

If you haven't checked out the amazing podcast that Malcolm Hawker and Barr Moses did recently on Data Observability, please do! It's excellent.

Malcolm's structured approach to getting the basics sorted whilst allowing Barr to showcase the incredible innovation happening at Monte Carlo demonstrates exactly why CDO Matters has become such an essential resource for data leaders.

The pace of innovation that Barr and the team at Monte Carlo are achieving in defining and expanding the data observability category is genuinely impressive—and something the entire industry should celebrate.

As practitioners implementing Data and Information observability capabilities across regulated industries, we wanted to build upon this excellent foundation with some additional perspectives from our recent projects.

Beyond Traditional Silos: The Need for Observability Frameworks

Barr's articulation of data observability as "knowing when your data is broken and why" provides an excellent starting point. We believe observability is about realising the "data as an asset" vision.

Malcolm's point regarding the engineering discipline and IoT telemetry is exactly right. Our analogy (aligning with Malcolm's) is F1 cars.

F1 teams have telemetry from almost every single component of the car—why aren't your data and information assets the same? If you can't see an asset, know where it is, monitor its health, track its supply and demand—how are you managing it?

Expanding the Definition: Our Position on Data & Information Observability

Whilst Barr's definition provides an excellent foundation, our work with global financial services institutions has led us to expand this definition along two critical dimensions:

First: Data AND Information Decision makers don't consume raw data—they consume information, which is data + context. When a Chief Risk Officer asks "Can I trust my regulatory reports?", they're asking about information assets that combine atomic data with business context and regulatory frameworks.

This is about measuring consumption of the data (usually in a context - i.e. Information) - in its various forms - by various stakeholders. And increasingly, that will be not just humans - other systems, AI (on behalf of humans) and AI agents.

Second: Beyond DataOps to CDAO Capability

At BearingNode, we define Data & Information Observability as:

"The capabilities to observe and steward Data and Information assets across their entire lifecycle, enabling organisations to understand health, performance, and impact whilst maintaining strategic alignment with business objectives."

This extends D/I O11y to encompass organisational impact and active stewardship capabilities.

That isn't criticism — we think Barr and the Monte Carlo team are focused on building and selling great software.

D/I O11y: Enabling Data Observability Platform Success

At BearingNode, our role is helping CDAO teams organise around the capabilities that platforms like Monte Carlo bring. Hence, we have developed the D/I O11y framework which takes a multi-dimensional view of the CDAO Operating Model.

Our five-capability framework (Value, Discover, Track, Comply, Govern) enables firms to leverage their data observability platforms effectively by providing the organisational context that technical solutions require.

The Critical Role of Open Standards

One area where we particularly appreciate Monte Carlo's approach is their recognition that data observability will thrive through open standards rather than proprietary solutions.

Just as OpenTelemetry created a rising tide that lifted all observability vendors, open standards in data observability will accelerate innovation whilst enabling the integration necessary for D/I O11y frameworks to succeed.

Our work with the OpenLineage community positions observability within this collaborative future.

Please Keep Building Barr and Thank You, Malcolm

The work that Barr and the Monte Carlo team are doing to define and evangelise data observability deserves recognition from the entire data community.

Their technical innovation, thought leadership, and commitment to open standards are creating the foundation upon which D/I O11y frameworks can build.

Malcolm's podcast demonstrates the quality of thinking driving this evolution—and we're excited to contribute our D/I O11y framework to this vital conversation.