Coaching three data teams back to predictable delivery with Scrum

The challenge

Like many teams, Rio Tinto's Geoscientific Data team has more stakeholders wanting work done than it has people to do it. Over time its Jira backlogs, workflows and project configuration had gone stale and no longer made for an efficient way to work.

The team's three sub-teams had concentrated so hard on getting work done that their ways of working had lapsed. It was difficult to tell which items were in progress, which were high priority and which were quick wins, and very difficult to manage each person's workload.

Our approach

As agile coach, we reviewed how work was done in each of the three sub-teams, then planned and carried out a move to Scrum that addressed most of their pain points.

We started with a sailboat retrospective to find out what was driving the teams forward and what was holding them back. Then we worked with each team to clean up its Jira issue types and workflows, and changed the Jira projects step by step to support sprints and Scrum backlogs.

We helped each team develop its ways of working and start running the standard Scrum ceremonies: sprint planning, stand-ups, retrospectives and showcases.

The outcome

Work demand and work in progress are now visible, estimated, sized and planned, without undue effort. That makes delivery to stakeholders more predictable and easier to communicate.

Working in iterations gave each team more consistent throughput and higher predictability than before. The teams said they benefited quickly from cleaner backlogs and from taking in just the right amount of work to complete each sprint.