By Prashant Gupta and Seema Bansal
Many software companies and startups have widely adopted “Move fast and break things” as the go-to formula when building new products and adding features. This approach prioritizes quick delivery of new features to users, often without fully recognizing the technical debt that mounts up behind the scenes.
Figure 1: Technical Debt (by Vincent DNL, available under a CC BY-NC 4.0 License at https://vincentdnl.com/en/drawings/technical-debt)
What is Technical Debt?
Technical debt refers to the consequence of choosing a suboptimal solution in the present time to ship something quickly at the cost of substantial effort that may be required later to fix problems and maintain a competitive product. While it’s tempting to focus on speed, unmanaged technical debt can slow teams down in the longer term. It often builds up slowly as teams cut corners to meet tight deadlines or add features to meet ever-changing customer needs. Over time this leads to the codebase becoming increasingly complex, fragile to changes and harder to maintain. This results in making systems prone to bugs, performance issues and may limit the organization’s ability to innovate and stay competitive.
Strategies for Balancing Technical Debt with New Features
The key to long-term success in software development is finding a balance between investing in new feature development and addressing technical debt. This requires leaders to develop foresight on the growing business and willingness to make difficult tradeoffs. Below are some strategies that teams can use to effectively strike this difficult balance:
- Reserving capacity for reducing technical debt: Instead of considering technical debt as something to be addressed only when it becomes a development blocker, teams should reserve some percentage of development capacity to consistently buy down technical debt.This could involve updating outdated dependencies, improving test coverage, and modernizing components with the latest recommendations. While this might reduce overall capacity available for current feature development, it helps in keeping the system robust. It has been observed that reserving about 15-20% of the tech budget leads to significant tech debt reduction over 3 years.
- Clearly communicating tradeoffs: During design discussions and planning cycles, it’s critical to be transparent to leadership and stakeholders about the potential impact of technical debt and its effects on long-term system availability and resilience.
- Promoting a culture of quality: Cultivating a culture that sets a high bar for code quality is key to preventing further technical debt accumulation. This requires establishing mechanisms to ensure the right processes, tools, and incentives are in place to model desired behaviors.
- Leveraging modern tools: Teams can leverage a variety of automated tools that help identify, track, and manage technical debt accumulation. AI-powered coding companions and code reviewers can help detect anti-patterns, flag use of non-recommended library versions, automate test creation and write boilerplate code to improve development time.
- Celebrating technical debt reduction: While it may feel natural to celebrate new feature releases, it’s also important to be intentional in recognizing efforts made by the team to reduce technical debt.
- Generalizing solutions over specializing: When building software solutions for a broad user base, there may be instances where specialized solutions are required to meet specific customer requirements. This often requires engineering teams to maintain complex configurations, which often leads to increased technical debt. Taking a more generalized approach when possible, can help the team in the long run.
Conclusion
Engineering teams have to carefully balance between maintaining system robustness and innovation. However, the approach needed to achieve that balance can vary depending on the specific needs and constraints of each organization. To strategically move fast, teams can take measures throughout the development phases to manage technical debt. Key strategies include clearly communicating design decision trade-offs, promoting a culture of quality, and celebrating technical debt reduction.
Digging Deeper:
Technical debt management automation: State of the art and future perspectives by João Paulo Biazotto, Daniel Feitosa, Paris Avgeriou, Elisa Yumi Nakagawa. https://www.sciencedirect.com/science/article/pii/S0950584923002306
Technical debt unpacked: Agile strategies for a debt-free development cycle by Dan Radigan. https://www.atlassian.com/agile/software-development/technical-debt
Mastering Technical Debt: A Comprehensive Guide to Understanding, Managing, and Overcoming Challenges by Jomsborg Lab. https://medium.com/@JacekWo/mastering-technical-debt-a-comprehensive-guide-to-understanding-managing-and-overcoming-19d2ced78942
Tech debt: Reclaiming tech equity Tech debt: Reclaiming tech equity | McKinsey https://www.mckinsey.com/capabilities/mckinsey-digital/our-insights/tech-debt-reclaiming-tech-equity
About the Authors
Prashant Gupta is a seasoned leader currently serving as a Software Development Manager at Amazon where he leads a team of engineers to deliver scalable software solutions for Amazon’s vast network of warehouses. With a Master’s in Computer Engineering from UC Davis and over 9 years of professional experience, Prashant specializes in software development and management, utilizing state-of-the-art cloud solutions. As a cross-functional leader, he fosters global collaboration to ensure the quality and reliability of their projects. Prashant’s contributions extend beyond Amazon; he also serves as a reviewer for Elsevier, where he contributes to advancing research in computer science. LinkedIn: https://www.linkedin.com/in/prashantgupta16/
Seema Bansal is a Microsoft Product Manager with over 8 years of Product Management experience. She is currently building features for Microsoft Teams. Before working on Teams, she was managing security products in Cloud and AI within Microsoft Azure. Her computer science engineering & visual design academic background, matched with her entrepreneurial endeavors, has framed how she approaches building and scaling products. It’s through her empathy-driven product building approach, curiosity, and incredible mentors that have helped her grow into a seasoned product leader. LinkedIn: https://www.linkedin.com/in/seema-bansal/
