MULTINATIONAL FOOD, SNACK AND BEVERAGE CORPORATION
Applying SAFe Principles to a Data Analytics Team
INTRODUCTION
Springbach has had the privilege of working with one of America’s oldest and largest beverage companies since Springbach began operations in 2018. So, when this client needed lean-agile leadership for its second largest program, they turned to Springbach. Its flagship program, aimed at improving front-line operations, has been applying Agile/SAFe principles for over two years. They are the gold standard within the sector - a true role model when it comes to what it can accomplish and produce. Agile/SAFe plays a major role in that success. All development teams have been doing well with what has been introduced and coached on, but the Data Analytics team needed a boost to get up to the same caliber as they moved into their next major funding phase of work.
In this case study, we will demonstrate how Agile/SAFe can be applied to teams working with data analytics - not just traditional software development teams. You may be surprised by how effective this can work!
THE OPPORTUNITY
Earlier this year, Springbach started providing Scrum Masters for many of the teams on our client’s Agile Release Train. One of those teams was their Data Analytics team, working with and training the data to provide individualized outcomes within an app for each individual user. In the past, there had been a lot of resistance to Agile/SAFe from the Data Analytics team. A resounding theme was “Agile doesn’t fit with the type of work we’re doing”. Because of this there were a lot of anti-patterns that were taking place. The team was far from reliable in what they could deliver and was consistently not delivering on what they committed to during the planning interval (PI).
Since the team was in a constant state of supporting other teams, it seemed like there was always confusion about what was going on and what was being worked on. During every iteration there were stories and story points being carried over into the next iteration. What was completed during an iteration was not being demoed during the Iteration Review. There was a severe lack of communication, and the work that developers did was hidden behind a team lead that was prone to micromanaging.
OUR SOLUTION
As you can probably guess, Agile/SAFe does in fact work for a team working with data analytics. Just by implementing the most basic Agile practices, the team’s daily work and overall predictability changed drastically, and what seemed like overnight. As the team got started, we implemented...
-
Having the proper ceremonies in place and making sure they were worth everyone’s time played a big role in making sure everyone was on task. Specifically, starting each morning with the daily stand up focuses the team for the day and allows space to talk through any support needed. Additionally, backlog refinement, sprint planning, and retrospectives have given good opportunities to look forward into the coming PI for planning purposes, and to reflect on what has been accomplished in the past sprint and how.
-
The team is spread across different time zones, so understanding everyone’s available meeting times has built understanding and trust. Also, knowing what is going on in each other’s lives (such as holidays, celebrations, children's accomplishments, etc.) has helped build camaraderie. The What’s New in SAFe 6.0 article says, “Respect for people is now a SAFe Core Value since it’s a basic human need.
Treating people respectfully unlocks their intrinsic motivation to learn and grow, evolve their practices, and contribute to their business and customer outcomes.” This value is held in high regard and has actually has actually helped in the fluidity of the team’s work.
-
Since the Data Analytics team does support other front- facing teams, we decided to create a form for submission that would help keep requests from these teams organized. Teams that rely on the Data Analytics team fill out the form and no work will start on the requested project until the form is filled out and we have the proper work slated in the proper sprints.
This helps with backlog refinements and eliminates the confusion of what is being worked on and where. It also encourages other teams to make sure they know what they need and how Data Analytics will support that vision. This form gives the Data Analytics team the minimum answers to make the stories slated actionable and greatly helps them understand how to test the work needed.
-
Defining the Data Analytics testing process has been imperative. The team has a different way of testing than the other development teams which requires verifying code against sample data and extracting data to be verified by another team while testing integration impacts. Having the specific process described in detail in the Ways of Working document to be revisited at the beginning of each PI is important for the team’s success.
-
Allowing the team to decompress and work on what they found valuable for the innovation and planning sprint allowed for great discussions surrounding larger processes, and led to new ideas, as well as more forward-thinking, exploratory work overall