Data Mesh is something new. That's why there are hardly any best practices and experience reports from the industry. Whether data mesh is really a sensible approach for your company can only be determined with a more detailed analysis. But with a few questions you can quickly get an indication in advance. That's why we present here a partial list of heuristics that can help you assess whether Data Mesh makes sense or is at least worth thinking about for your company. We have deliberately avoided a scoring approach, because such a thing always has its weaknesses, and instead worked on an unsorted list of topics with a rough classification of self-assessments into the three response categories "rather not", "worth thinking about" or "useful". This will give you a sense of how ready your company is for Data Mesh.
Usage: Go through it row by row. Click on the statement that best represents your company. Your selection is only stored in your local browser.
Data Mesh Fitness Test for INSERT_NAME_HERE
| Data Mesh might not be the right choice | Data Mesh is worth thinking about | Data Mesh seems useful | ||
|---|---|---|---|---|
| Domain-driven Design | Is unknown or does not matter | Partially known and understood | Is understood by all. There are bounded contexts and context maps | Don't know |
| Verticalization of the software system based on domains | No. There is a monolith for the entire business | The transformation to microservices has begun | The software system consists of domain-oriented microservices | Don't know |
| Number of software development teams | <= 5 | > 5 | > 10 | Don't know |
| Make-or-buy strategy for the core domains | Standard software (SAP, SaaS, etc.) | A mix of standard and individual software | Only individual software | Don't know |
| Cloud computing | No issue, many concerns | In the midst of cloud transformation / hybrid cloud | We are cloud native | Don't know |
| Team structure | Separate front-end, back-end and operations teams | Mixed teams consisting of specialists | Small cross-functional teams, supported by a platform team | Don't know |
| Trust | Micromanagement by leaders | Control through goals | Teams and employees are trusted to do the right thing | Don't know |
| DevOps culture | What is CI/CD? | We are establishing a DevOps culture | We are living the DevOps principles | Don't know |
| Product Thinking | Thinking in terms of projects. Roadmaps dictate implementation | Product development is attempted, but fails due to management | Autonomous product teams that (further) develop their products themselves | Don't know |
| Competitive advantage through ML and analytics | Not in focus | Potential is recognized and first lighthouse projects have been launched | Data-driven product development is strong | Don't know |
| Central data team as a bottleneck | The data team handles all requests at a reasonable speed | Analysis results often take several weeks or months | Product owners don't ask analysis questions in the first place | Don't know |
