Getting a better understanding of data and its uses to create value is an increasingly shared ambition. However, companies continue to face challenges that Data Mesh could help overcome... but only if the process of transformation is successful and the whole company is engaged.
In this article, we give you some tips to get started and some feedback from companies that have successfully implemented one or more Data Mesh pillars.
In the previous articles, we identified several relevant steps for building the Data Mesh framework. To get started, we recommend listing the priority actions for each pillar
- Organizing an initial data domain. For this, we need to find a business line that is mature enough in data management and that seeks to gain autonomy in its optimization. The basis of the Data Mesh operation can be set up from this base.
- The implementation of a data product pilot. This is a way to extend the product culture within the teams and launch initiatives on the other three pillars.
- Creating a data platform. Apart from the Data Mesh approach, many companies are now undertaking this large-scale modernization project with a view to industrialization. With the Data Mesh approach, the idea is to build a platform to meet all the different needs and self-service.
- The application of federated governance principles based on the Data Owner and Data Steward functions. By this pillar, we mean that companies become more accountable and also integrate measurable data objectives into their business challenges.
First example: From Data Hub to a Data Platform as a Service
Our example is about the case of a large French multinational in the digital sector.
Within operational project, which extends over several years, the company plans to review the data architecture.
In a first phase, the company wants to identify the components of its systems that are compatible with the Data Mesh approach.
In a second phase, the company will then define its management of the data architecture, starting with identifying the components of its existing system that are compatible with Data Mesh. Taking advantage of the technological capabilities of the new tools, the company is redefining its governance and organization. The platform was previously operated in Data Hub mode based on Hadoop Spark. It was modernized and implemented with Infrastructure as Code and the Cloud.
In this case, the Data Platform is managed as a product by the IT department, which then can manage it according to their own business needs.
Second example: Governance, Product & Platform
In this other example, a company specialized in the retail sector took the modernization of its data platform as a first step to the Data Mesh approach.
Here, the approach is mostly technological and aims to support the existing data lake as there is a lack of agility for the company. To solve this, the company decided to migrate the entire data platform in the Cloud.
With this project, the approach is all about the direct involvement from the business domains in the implementation of a data product logic. This also leads the IT department to work on the project side.
