Data Mesh: The Tech-Agnostic Methodology

In the search for data-to-decision optimisation, it’s easy to overlook one of the most critical drivers: People and their ability to drive data as a Product and Asset. With the industry’s attention focused on technologies and digital tools that turn our data landscapes into functional analytical platforms, data mesh is turning our gaze back on the people behind the wheel of our data journey.

Why is data mesh necessary in data-driven businesses?

Data is the heartbeat of our business drivers these days. Every action and process funnels into our data landscape with technologies that let us see our organisations with an intensity that was never before possible. The hope is to leverage these monumental insights in the realisation of tailored experiences.

Today, however, the sheer flood of data has resulted in often ineffective data warehouses and data lakes, where data is managed from a centralised location. The result is employees who lack the ability to make immediate data-driven decisions because they have to wait for data to reach them from the centralised storage location.

These centralised architectures lack the ability to support the scaling needs of organisations working with multiple domains, where numerous data types need to be processed and fed back into the data stream before they can be connected to the people who need it.

Data mesh seeks to solve this problem by addressing the psychology that has dominated many businesses' approach to data for a long time: that bigger is always better. More data in one place equals a more thorough representation of business functions, right? Well, it’s true from a corporate sense, but not from an operational sense, where domain-specific data needs to be connected to the right people in real-time and the bigger picture is of secondary concern.

But what’s most important is that data mesh isn’t a new technology. Instead, it’s a methodology that is enabling teams to utilise their existing tools in a new way to connect domain-specific data to decision-makers in real-time. Giving your organisation the ability to advance & scale your tools and processes in parallel without the typical monolithic chaos.

Data-mesh: The?technology-agnostic methodology

The technology-agnostic approach of data mesh makes it easy for companies to augment their tools for optimised performance without restricting them to any specific software.

Unlike many other developments in the data analytics field, data mesh is a socio-technical methodology that connects people, processes, and technology. The focus of new developments usually lies heaviest on new tools, with a bit of attention given to training on the side to ensure people know how to utilise new technologies.

With data mesh, only 30% of the effort goes towards tech, while 70% of the focus lies in changing people’s perceptions and altering business processes.

Here’s how the shift can be broken down:

People

With the extinction of the once central data team comes the dawn of decentralised domains where people get direct access and full ownership of their data. With a slew of new processes and mindsets, as well as new responsibilities and accountabilities, it’s a whole new ball game that requires teams to change gears entirely. As such, the data mesh journey encompasses organisational restructuring and role changes.

Processes

Data mesh turns the flow and access of data entirely topsy-turvy. So while data management processes are being updated, aspects such as data governance need to be updated as well. The policies and processes that exist around data need to be updated to implement and enforce a secure environment.?

Technology

While the technology-agnostic nature of data mesh allows you to augment existing digital environments for domain-driven data pipelines, new technologies are likely to emerge that will help you make the streamlined efficiency even easier.

What’s most beneficial about the shift of data mesh is that it allows businesses to scale their application of data, and force the optimisation of processes and prioritization for increased domain performance without having to reinvest in new technologies.

Driving domain performance with data mesh

The idea of domain-driven data ownership and architecture allows technology to better serve the people connected to a specific business function. The goal is for each domain to manage the data related to and created by its functions. Bearing in mind, people and teams can own or work in multiple domains against multiple data products and this starts the conversation about the sticky pieces which dissolve silo’s in organisations.

The by-product of this facilitates cross-pollination of skill sets and this still happens within the reach of what would be the centralised Data Governance Strategy.

This allows data mesh to become a vital driver as it promotes organisational agility and empowers teams to respond to their environment with increased accuracy and efficiency. Data mesh aims to increase the resilience of data by simplifying often complex ETL processes between operations and analytics so those insights are produced faster.

Embracing a new relationship with data

While centralised data models like data warehousing and data lakes may offer adequate solutions for companies with a simple domain architecture, once you’re dealing with domain-rich environments where multiple data types and sources converge in a flood of data, pooling them all together isn’t the answer. This is where we start talking about Data Polyglots however we would then begin delving into technical conversations better served on a different platform.

Data mesh allows data to reach the teams that most intimately understand a specific domain in the shortest time possible and makes the data-to-decision journey richer.

So, the question is: Is your company connecting data as effectively as possible with the people who need it most? If not, data mesh may be the answer you’ve been looking for.