Tech Leader Chats: How to conduct human-centric, data-driven retrospectives

About the talk

We know that retrospectives are important for engineering teams for a number of performance reasons - enhanced processes, improved teamwork, and efficiency of delivery. However, there is a gap in practice between the objective data and the subjective data used in retros to evaluate performance. There are also so many ways to conduct effective retros, so identifying what fits your team’s evaluative needs is a unique challenge for engineering leaders to consider. 
‍
What barriers get in the way of running meaningful retros? How can engineering leaders mitigate team concerns around retro processes? How do you maintain a balance between human-centric reflections and data-driven insights when conducting retros? We welcomed Aino Corry from Metadeveloper to have a conversation with our Lead Data Scientist, Vivek Katial, to discuss all things to do with retrospectives.

In this session, Aino and Vivek shared insights on:

• Common pitfalls that leaders encounter when running retrospectives
• Why it’s important for teams to conduct data-driven retros
• The impact of people problems on retros
• How to maximize the benefits of retros for your team members

Our key takeaways from Aino and Vivek's talk are as follows:

Emphasize the 'Prime directive'

When engaging in retrospectives, it's good to enforce a Prime Directive. This refers to when you go into retros understanding that everyone did the best job they could, given what was known at the time, their skills and abilities, and resources. Living the Prime Directive enhances the psychological safety of the retro environment, and encourages open and frank participation in the process. Reminding people of the Prime Directive when they're invited to retros or at other appropriate points during the process ensures that participants actually take stock of what's being said, and approach retros in a non-combative way.

Don't skip the fluffy parts

It's one thing to have all the objective data in place with retros, and going over action points, but don’t skip the part where you generate insights. Working on producing those insights as a group is your team's chance to iron out any assumptions, and to make sure that the things you’re identifying as problems aren’t just symptoms of problems. Don’t just focus on rushing through the process of a retro; you don't get awards for finishing quickly. Taking the time to understand why things happened is just as important as being able to point out all the things that went wrong.

Facilitating a retro is a full-time job

When someone is facilitating a retrospective, let them focus fully on that.  It’s more than just note-taking. Facilitating a retro requires being hyper-aware of everything and everybody. Even though the facilitator misses out on having the retro themselves, they’re a necessary part of the team and critical to a retro’s success. An outside, neutral facilitator might provide a new perspective if a team is struggling with this part of the process - it can be difficult having someone who is invested in the outcome of a retro on a personal level to also be an objective and effective facilitator.

Resources

View the slides from the talk here, and the full talk transcript here.

Join our Tech Leader Chats community

Enjoyed watching Aino's talk? Be sure to join our community for future talk updates.

About the speakers

Aino Corry, author of "Retrospectives Antipatterns", is a teacher, a technical conference editor and retrospectives facilitator. She holds a masters degree and a PhD in computer science. She has 12 years of experience with Patterns in Software Development, and 20 years of experience with agile processes in academia and industry (in other words; she is old). She also teaches how to teach Computer Science to teachers, and thus lives up to the name of her company; Metadeveloper. In her spare time, she runs and sings (but not at the same time).

Vivek is Lead Data Scientist and a founding team member at Multitudes, where he drives and executes the company's Data and AI strategy. He also holds a PhD in Optimization via Quantum Computing. He is passionate about building data products that create positive change. As co-founder and Executive Director of The Good Data Institute (GDI), he's led an Australian NGO that helps charities build data capabilities, supporting over 65 organizations on more than 80 projects globally. In his spare time, Vivek loves to jam on data science, ML and ethics. He's recently spoken at Tech 4 Social Justice, OPTIMA-CON, and LAST Conf.