Migrate your Culture for the Cloud

Jim DuBois, Corporate VP & CIO, Microsoft Corp.
108
190
36

Jim DuBois, Corporate VP & CIO, Microsoft Corp.

Our company is in the midst of a cultural and digital transformation. We're aspiring to create a living, learning culture by championing a growth mindset across Microsoft. This will allow us to learn from ourselves and our customers. We want to make a difference for our customers, our partners and individuals within our communities. And we will better serve everyone by representing and valuing each and every different perspective. These are the key attributes of the cultural shift at Microsoft. I feel great about how it’s resonating with, and empowering, our employees.

At the same time, Microsoft is in the midst of a digital transformation. This transformation is grounded in the combination of technology, people, and processes, so that we can better engage with customers, empower our employees, optimize our operations, and reinvent our products and business models.  

As I reflect on this fast-paced, evolving environment for the information technology (IT) team at Microsoft, we’re challenged to meet customer and business needs while we efficiently manage infrastructure and operations. Cloud computing has been a big part of the solution. We have migrated more than half of our internal line-of-business applications to the cloud, and SaaS solutions are the default for productivity applications.

  ​Digital leadership requires CIOs and their business counterparts to champion a digital mindset, which emphasizes sharing technology leadership 

This cloud migration wasn’t done in isolation, but in parallel with our changes in culture and processes. The first step was the adoption of a modern engineering model within IT. This model has two components:

• First, we are focused on maturing our people and our culture. We are increasing efficiency by merging development and operations into a single role—DevOps. This move empowers any engineer to perform any task on the team.  
• Second, we’re applying agile development principles, methods, and tools to all of our service offerings. This allows us to further shorten development cycles and increase responsiveness.

Today, it doesn’t take us three months to release a new feature. With the flexibility and speed of the modern engineering approach, high-priority projects are released in as little as two weeks from inception. Ultimately, we plan to deliver incremental updates on a continuous basis—with new releases every day.

As a first step in the move to modern engineering, we realigned the IT organization so that it mapped to the company’s business processes, such as sales, marketing, product development, supply chain, finance, and more. This mapping removed organizational barriers and aligned the right resources to each project. Then we started the process of merging software engineering—development and test—and service engineering—operations—roles into an agile DevOps team. The goal is to make each team member aware of the issues that other roles face, so they can cooperatively develop solutions. This increases both the efficiency and effectiveness of the team. Ultimately, any sprint team member will be able to perform in any engineering role.

We identified four levels of people and culture maturity in the progression from traditional to modern engineering. Moving up through the levels, the operations and development/test functions gradually merge.  

• Level 1—Trust: Software engineers and service engineers learn more about one another’s roles. As they begin to understand each other better, they develop greater patience and trust.
• Level 2—Shared goals: Service engineers and software engineers continue to learn about each other’s roles, while they begin sharing a common backlog. In the completion of tasks, they step in to help each other wherever possible.
• Level 3—Role sharing: Roles begin to merge. Service engineers and software engineers start using common systems and tools and cultivate new skills, so they can begin stepping into one another’s roles.  
• Level 4—Fully integrated roles: At this level, there is no difference between service engineers and software engineers. Everyone is part of the same team, with common goals. The customer is the focus. There are no more hand-offs. Only common tools, processes, and projects.

Our engineering teams are progressing through these levels, albeit at different speeds. Most teams are currently in Level 2 or 3. One hurdle to moving up a level is the time required to learn a new role while maintaining responsibility for one’s existing daily tasks. Another challenge is the shift to experimentation. Some teams postponed moving to the cloud, until they evaluated their approach and migration path. However, more successful teams activated with a basic premise of approach and migration path, then incorporated a culture of ‘experimentation, iteration, and lessons learned’. The teams that embraced this shift to experimentation moved more rapidly to the DevOps model, with improved development velocity.

Putting the modern engineering model into action has already yielded significant benefits to Microsoft:
• We now meet business customer demand with faster release cycles—delivering new features in as little as two weeks.  
• Issues are corrected more quickly, so that internal stakeholders or customers no longer have to wait for the next major release.  
• Updates and enhancements to the application portfolio are delivered continuously, with the business value of development efforts realized much sooner than before.  
• We have reduced our exposure to risk by breaking down releases into smaller chunks. Features now represent only two weeks of effort, rather than several months.  

Digital leadership requires CIOs and their business counterparts to champion a digital mindset, which emphasizes sharing technology leadership, changing from a service mentality to an investment mindset, and letting go of legacy systems that are a drag on innovation. The biggest obstacle to transformation is organizational culture; the current culture may have served organizations well by getting them to where they are today, but is unlikely to include the behaviors and mindsets for the digital future.

Read Also

Data Design and Storytelling

Dona Wong, VP-Digital Strategy, Communications & Outreach, Federal Reserve Bank of New York

Back to the Networking Future

Jim Houghton, CTO of the Americas Region, CSC

Real Time Data Integration on Hadoop

Richard Winter, Principal Architect, Think Big Analytics, a Teradata Company [NYSE: TDC]