RDEL #83: How can engineering teams use data in retrospectives more effectively?
Subjective insights take priority due to data asymmetry, psychological safety concerns, and preference for simplicity.
Welcome back to Research-Driven Engineering Leadership. Each week, we pose an interesting topic in engineering leadership and apply the latest research in the field to drive to an answer.
Retrospective meetings are a core Agile practice, helping teams reflect, adapt, and improve. But despite having access to rich project data—such as code commits, test coverage, and deployment frequency—many teams still rely on memory and anecdotal feedback in these discussions. This week we ask: How do engineering teams use data in retrospective meetings, and what challenges prevent its full adoption?
The context
Retrospective meetings (or "retros") are a key Agile practice, providing teams with a structured opportunity to discuss what went well, what didn’t, and how they can improve. These meetings are typically guided by frameworks like Scrum, often involving interactive tools and templates to facilitate discussions.
The core idea behind retros is to foster open communication, psychological safety, and continuous improvement. However, despite software development teams generating vast amounts of objective data—from commit history and test results to deployment frequency and team sentiment surveys—many teams rely on memory, anecdotal feedback, and gut instincts rather than systematically incorporating data into retros. While subjective experiences are valuable, ignoring objective data can lead to biased decision-making and missed opportunities for improvement.
The research
A recent study surveyed 19 agile teams to understand their retrospective meeting practices and how they incorporate (or fail to incorporate) project data. The researchers analyzed how teams prepare for retros, the tools they use, and the barriers they face in adopting a data-driven approach.
Researchers discovered a few key attributes on how retrospectives are generally run:
Varied structures and formats: Teams used custom boards and formats to better suit their teams. Some teams take a more structured process, while others employ more creative approaches (such as using non-traditional structures to encourage participation).
Subjective insights take priority: While most teams track various project and engineering metrics, few systematically bring these into retrospective discussions. Instead, they prioritize open-ended conversations based on personal reflections.
As for barriers for adopting data during retros, researchers discovered the following trends:
Psychological Safety Concerns: Some teams worry that introducing data could feel like surveillance or performance evaluation, discouraging open conversations.
Asymmetric Data Usage: Managers or team leads often review and interpret project data independently, but this information isn’t always shared transparently with the broader team during retros. For many teams, metrics weren’t integrated into the sprint planning or retro process.
Preference for Simplicity: Many teams favor lightweight tracking methods, like spreadsheets or whiteboards, over dashboards and analytics tools.
The application
This research highlights an opportunity for engineering leaders to enhance retrospectives by integrating objective data while maintaining the human-centric benefits of discussion and reflection. Here’s how:
Normalize Data-Driven Discussions – Frame objective metrics as tools for insight rather than judgment. For example, instead of using data to assign blame for slow code reviews, use it to identify patterns and make process improvements. A key part of doing this in a way that empowers the team is to anonymize and aggregate data to a team level.
Encourage Pre-Retro Data Review – Provide teams with key data insights before the meeting so they can reflect and bring informed perspectives to the discussion. When teams have access to their own data, they can proactively identify areas to refine.
Pair Data with Context – When introducing data, remember that numbers alone don’t tell the full story. Encourage teams to discuss the “why” behind the trends and combine objective insights with subjective experiences to build a complete picture.
By thoughtfully integrating data into retros, engineering teams can move beyond intuition-based decision-making to drive more effective, measurable improvements in their workflows.
—
Happy Research Tuesday,
Lizzie