Read this white paper to learn how Intel IT is accelerating Agile scalability and transformation with a leadership-first approach and good scrum planning and practice.
Large-scale, Agile transformations are extremely challenging for digital enterprises. Every organization’s path is unique and more often than not, failure is part of the process. Intel IT’s Agile journey is no different—at times oscillating between challenging to downright discouraging over the course of a decade.
Today, Intel IT is on course to double the value delivery capacity of all IT teams by 2023. This paper focuses on how we traversed the chasm between discouragement and sustained improvement. More specifically, this paper will discuss:
How Intel IT leveraged the role of leadership to accelerate and support our Agile transformation.
Why “good scrum” practices are essential for scale.
What novel approaches we took to accelerate our progress.
Lessons learned throughout our Agile journey.
Like many enterprises, Intel IT executives were concerned with shortening cycle times, eliminating waste and improving predictability. Understanding this upfront allowed us to create goals that kept our scrum implementation anchored to achieving real business results. This helped us to not only track progress, but also ensure continued leadership sponsorship and participation in the journey.
We achieved measured improvements in each of the areas that executives were concerned about. Improvements included:
A 26% increase in velocity, which is the value delivery capacity of the teams.
A 31% increase in the number of teams that reliably make and meet commitments for each sprint.
A 25% increase in the number of “happy” teams as reported in the sprint retrospectives score of 4+/5.
Here are some of the key lessons we learned along the way:
Anchor the scrum journey in the pursuit of real business results. We began by asking leadership a few fundamental questions: What problems must be solved? Do you believe scrum can help solve these problems? What is the current baseline? What is the target?
Only scale “good” scrum. Scaling teams that have not yet achieved good scrum practice will only amplify inherent problems.
Scale scrum for the right reasons. Scale scrum only when there is a need for coordinated teams to create a larger integrated work product. Scaled scrum is not a management structure, but rather a cross-functional integration structure to help teams deliver something larger than a single scrum can on its own.
Scaling scrum is fractal. Once good scrum is in place, scale only the things that work, such as sprint planning, scrum leadership roles, team agreements and so on. Strive for the “minimum viable bureaucracy” at the scaled level.
Scaling scrum is incremental. Avoid scaling all at once; instead, start gradually with 5-7 teams. Take note of what works and make that the standard for how to scale the next batch of scrums.
No one can dictate how to scale scrum. Regardless of anyone else’s experiences, including our own, organizations must find their own way to “good” scaled scrum using fast feedback, inspection and adaptation.