Sonar Success: Fireside chat with DATEV

Introduction to DATEV and Clean Code Excellence

DATEV is a unique IT cooperative serving over 40,000 members in tax consulting and legal advice, with approximately 2.5 million companies in Germany benefiting from their software solutions. With 8,500 employees and around 1,200 software developers working on approximately 250 different products, DATEV operates at a massive scale. During a recent fireside chat moderated by Ky Joshi, Andy Fisher, a software developer and leader of DATEV's DevOps Centre of Excellence, shared how the organization has leveraged clean-coding best practices to improve development velocity, productivity, and overall business growth. The conversation highlighted how hundreds of projects across DATEV have benefited from improved code health through systematic quality initiatives.

The Problem: Cost of Poor Code Quality

Ky Joshi opened the discussion by addressing a critical market problem facing modern enterprises. As companies undergo massive digital transformation, software has become central to all business operations. However, the underlying code quality directly impacts organizational costs, with poor code quality accumulating significant expenses that now reach into the trillions of dollars globally. These costs continue to increase yearly, making code quality not just a technical concern but a business imperative. Sonar's approach addresses this challenge by arming developers and development teams with solutions that help achieve and sustain clean code—code that is consistent, intentional, adaptable, and responsible.

DATEV's Journey Toward Code Quality Standards

Andy Fisher shared DATEV's evolution in code quality practices, beginning with his tenure starting in 1987 as a software engineer. Early in his career, an expensive error instilled in him a strong commitment to good code quality. Throughout his career working with languages including C, C++, Visual Basic, and C#, Fisher developed and maintained basis software frameworks—notably a large C# framework that continues to run DATEV's on-premise products. His introduction to static code analysis tools like FXCop and later ReSharper proved formative in establishing governance practices. However, early attempts to enforce a strict "zero issue policy" revealed a critical flaw: developers began suppressing issues in the tools rather than genuinely improving code quality, essentially gaming the system rather than addressing underlying problems.

Implementing Effective Code Quality Governance

The realization that developers were circumventing quality checks rather than improving code led DATEV to adopt a more sophisticated approach to code quality management. Rather than relying on easily-bypassed enforcement mechanisms, the organization recognized the need for tools and processes deeply integrated into the development workflow. This shift in strategy aligned with Sonar's philosophy of enabling organizations to achieve clean code as part of their existing software development lifecycle. By cleanly integrating quality checks into their DevOps workflow, DATEV moved away from punitive measures toward a more constructive approach that genuinely improved code health across their extensive portfolio of products.

Key Takeaways

• Code quality is a business priority: Poor code quality has enormous, accumulating costs that now reach trillions of dollars globally, making clean code practices essential for business growth
• Zero-tolerance policies can backfire: Simply enforcing strict quality rules without proper tooling and integration can lead developers to suppress issues rather than fix them
• Integration is critical: Effective code quality management requires tools deeply integrated into existing development workflows and DevOps processes, not external enforcement mechanisms
• Scale requires systematic approaches: Managing code quality across 1,200 developers and 250 products demands governance frameworks and consistent standards established at the platform level
• Clean code defines success: Code that is consistent, intentional, adaptable, and collectively owned enables teams to build software that is easy to understand, operate, and maintain over time