Enable javascript in your browser for better experience. Need to know to enable it? Go here.
Purple banner image with text saying The business case for Data Mesh

Data Mesh Business case

Many organizations are using new solutions like Data Mesh to gain business insights faster。

 

For decades, companies have invested heavily in data architectures, pouring resources into building structures that quickly turn ever-increasing amounts of data into actionable insights。

 

Often, these investments fail to deliver the expected value。According to a study published earlier this year by MIT Technology Review and DatabricksOnly 13% of companies are doing well in implementing a data strategy。According to a study published last year by NewVantage Partners,只有26.Eight percent reported success in building a strong data culture in their organization。

 

For many enterprises, their choice of centralized data architectures, such as data warehouses and data lakes, is the source of frequent problems。Lengthy data entry times, difficult analysis bottlenecks, overworked centralized teams, and data quality and discovery problems can all be side effects of a centralized architecture。

 

Most importantly, in the rush to enter and process data, it is possible to lose sight of the key goal of availability of the data products generated by the architecture, resulting in teams across domains finding it difficult to utilize these data products effectively。In addition, developing the capabilities of teams across domains is critical to creating value, but can be overlooked in the context of a centralized architecture。

 

As a result, more and more enterprises are looking for more flexible solutions, and this is where Data Mesh comes in。

 

 

What is a Data Mesh?

 

Data MeshThoughtworks is a decentralized approach to data architecture, originally proposed by Zhamak Dehghani of Thoughtworks。In the Data Mesh, Data is not located in a centralized Data pool, but decomposed into different "Data products". Each domain team owns and manages the "Data products" closest to its own.。

 

 Zhamak proposed the Data MeshThere are four basic principles, which include:

 

  • Adopt a domain-oriented decentralized data architecture。In the Data Mesh, the Data is owned and controlled by the team closest to it, eliminating all kinds of links and handovers between the Data producer and consumer。

 

  • Manage data in product form。With custom products, teams that need data can easily get it。In this way, domain teams can quickly and easily access whatever data they need in a self-service format。

 

  • Deploy the self-service data infrastructure。Data Mesh is designed to be self-service and provide teams with an automated means to manipulate Data and extract value from it without the need for centralized expert manual assistance。

 

  • Implementing joint governance。Automated governance at the platform level ensures that standards are maintained without compromising flexibility or limiting the use of data across domains。

 

 

What does all this mean for your business?

 

As an architectural approach, Data Mesh perfectly fits the Data goals that today's enterprises want to achieve。It brings data producers and data consumers closer and enables teams across domains to adopt self-service and access highly relevant data products。Therefore, Data Mesh can help enterprises establish data-driven agile innovation culture and experimental culture, and integrate such culture into the whole enterprise。

 

 

Here are some of the transformative advantages that Data Mesh can bring to an enterprise:

 

 

Make smarter decisions faster

 

 

In a centralized data architecture, there is a complex and specialized manual process from data creation to action on the data。数据提取或录入工作量庞大,而需要数据的团队往往并不清楚这些环节;即使有数据可用,团队也可能需要等待漫长的分析周期才能将其转化为洞见。 

 

With Data Mesh, many steps can be automated or unnecessary, thus eliminating all of them。Teams in each field input data and manage data products by themselves。They know what data they have and can manipulate it in any way they want, at any time。这与集中式数据架构形成鲜明对比,集中式数据架构假设数据视图具有通用性,所以往往生成标准化的数据视图;而在Data Mesh的加持下,各领域团队可以按照自己的意愿采用定制化数据视图。

 

Therefore, Data Mesh can significantly accelerate the decision-making process of enterprises。Companies gain a competitive advantage by being able to run data faster and act on it, and they can extract more value from the vast amounts of data they collect and hold。

 

A large financial services organizationWith the Data Mesh architecture, the average time to value realization is significantly reduced and the results are almost immediate。With access to domain-oriented data products and the freedom to work with data quickly, executives can ask more questions, get more reliable answers, and act on valuable insights faster than ever before。The field teams are also able to integrate analytics data directly into the customer digital experience to provide a truly differentiated product in the marketplace。

 

 

 

Create a true culture of data-driven innovation

 

A major advantage of a decentralized architecture like Data Mesh is that the Data end user has control over the management and use of the Data。

 

In Data Mesh, domain teams are in the lead。As managers and controllers of data products, they are free to experiment with data。They can ask more questions, simulate more scenarios, and explore more data-driven "moonshot" ideas (i.e., continue to make important innovations based on ideas).。

 

Teams across domains are motivated to ensure that their data products are as coherent and well maintained as possible, as they directly impact the team's analytical capabilities and results。As a result, throughout the organization, there is a culture in which everyone in every field is actively involved in improving data quality, conducting experiments, and pushing the boundaries of data innovation。

 

Saxo BankWorking with Thoughtworks to build data-driven open banking institutions, Data Mesh has played an important role。By implementing the Data Mesh principle, the challenges of Data visibility, quality, and access were addressed, enabling teams to not only advance the goal of creating an open bank, but also to make continuous improvements。

 

 

Support for artificial intelligence and machine learning programs

 

Artificial intelligence and machine learning have rapidly developed from highly complex specialized technologies to widely used in all levels of modern enterprises。To achieve value, both AI and machine learning need two things: high-quality, relevant data sets and innovative thinking that can find powerful use cases for those data sets。

 

As each domain team controls its own Data products within the Data Mesh, it naturally begins to build and maintain the kinds of Data sets it needs to drive the implementation of disruptive AI and machine learning use cases。 

 

In addition, because domain teams are responsible for managing the data, they face far fewer obstacles when it comes to ai experiments and creating powerful new use cases。Data Mesh has become an enabler of AI and machine learning innovation, and teams are even free to create Data products specifically for AI and machine learning, enabling more teams and more domains to gain AI and machine learning capabilities than ever before。

 

 

 

Change starts with a successful business case

 

Together, these transformative advantages form the foundation of a strong business case for Data Mesh, which is both adaptable and relevant, but far more than Data Mesh can provide。Data Mesh is also well suited to help enterprises: 

 

  • Improve data quality and governance, and even automate many elements of governance and compliance with specialized data products 
 
  • Improve visibility, quality, and governance of the entire Data Mesh to respond to new regulations more quickly 

 

  • Create data products or participate in data product marketing activities, and securely share data products across different organizations, or even collaborate to create products

 

  • Discover more opportunities in enterprise data, as more people focus on important data and more teams are motivated to actively explore every potential data product use case

     

However, it is important to note that any business case created for Data Mesh needs to be tailored to the challenges facing the enterprise。Some of the strengths we've highlighted may be more resonant and uplifting than others, and that's where you need to focus。

 

 

Get more content