Breaking Engineering Org Anti-Patterns
A recent analysis highlights common anti-patterns in engineering organizations, such as over-centralizing decision-making or mistaking complex design for effective design. The piece argues that successful leaders actively break these patterns by iterating on org structures and empowering teams with clear ownership. These ideas are intended to improve collaboration and delivery speed.
- An over-reliance on centralized decision-making can create significant delays and lower the quality of outcomes due to a lack of local context. In contrast, companies that empower small, autonomous teams, like Spotify with its "squads," can roll out new features and fix issues more rapidly. A study of 93 teams found it was more difficult for them to adapt to a centralized structure after working in a decentralized one than the reverse. - A common anti-pattern is the "Heroic Rescuer," where a leader personally solves all critical technical problems. This leads to bottlenecks and prevents the team from developing its own problem-solving skills. Another is the "Feature Factory," which prioritizes shipping numerous features without considering their actual impact on business metrics, leading to technical debt and stagnant customer value. - The transition from a Director to a VP of Engineering often requires a shift from managing teams in a familiar domain to overseeing areas where one is not a subject matter expert. This necessitates a greater reliance on hiring strong leaders for sub-teams and communicating in terms of business impact rather than deep technical details. The typical career path to a VP of Engineering role takes 10-15 years of progressive experience. - AI agents are increasingly being used to automate and optimize SRE and DevOps workflows, moving from reactive to proactive incident management. These agents can autonomously monitor systems, detect anomalies, and even execute remediation tasks like rolling back deployments or scaling pods. Platforms like PagerDuty Process Automation and Datadog Watchdog are examples of tools incorporating these AI capabilities. - To demonstrate the value of infrastructure and engineering initiatives to executives, leaders often rely on DORA metrics (Deployment Frequency, Lead Time for Changes, Change Failure Rate, and Mean Time to Recovery). Research indicates that teams excelling in DORA metrics are twice as likely to meet or exceed their organization's performance goals. - Beyond DORA, a focus on Developer Experience (DevEx) is gaining traction, measuring factors like cognitive load, feedback loop speed, and flow state. Improving DevEx has been shown to boost developer productivity, satisfaction, and retention by removing friction from daily workflows. - One of the most detrimental anti-patterns is measuring engineering productivity by individual output, such as commits or tickets closed. This approach discourages collaboration and overlooks crucial work like mentoring and addressing technical debt. A better approach is to focus on team-level outcomes and system efficiency. - The "Not-Invented-Here" syndrome, where teams insist on building all solutions internally, is a recognized anti-pattern that can slow time-to-market and increase maintenance overhead. A systematic "build vs. buy" analysis is a recommended strategy to counteract this tendency.
