Defining the future: How we’re building an AI-powered continuous improvement culture at Microsoft

We’re at a crucial point in the history of technology. Our need for greater efficiency and productivity is urgent and increasing. At the same time, AI is unlocking our human potential with advances in workflow automation that were unimaginable just a few years ago.

This is a time of enormous possibility, but it comes with difficult questions.

How do we put AI into action to solve problems and drive impact? How do we ensure we’re pursuing initiatives that deliver real business value? And finally, how do we equip our teams with the skills they need to realize this potential? Across Microsoft, we’re adopting a methodology to ensure we make the most of this moment: continuous improvement.

“We exist to empower every individual and team at Microsoft to embrace continuous improvement powered by AI in all that we do—simplifying and improving business processes to accelerate growth and boost performance.”

Kirsten Paust, corporate vice president of Continuous Improvement, Microsoft

Our vision for AI and continuous improvement

Continuous improvement is central to our ambition for an AI-powered, human-led future. At its core, it provides a systematic, repeatable framework of methods and behaviors that help teams operate with speed, clarity, and discipline as they progress toward an enterprise enabled by AI.

As a methodology, continuous improvement enables our teams to identify and solve high-impact problems, establish a customer mindset to drive accountability, and codify practices that make progress sustainable and measurable. It’s a fundamentally iterative process based on identifying opportunities, executing initiatives, analyzing results, and making course corrections.

“We exist to empower every individual and team at Microsoft to embrace continuous improvement powered by AI in all that we do—simplifying and improving business processes to accelerate growth and boost performance,” says Kirsten Paust, corporate vice president of Continuous Improvement at Microsoft.

Continuous improvement isn’t new, but generative AI is. That’s why our approach at Microsoft involves redesigning end-to-end workflows with AI at their center, aligning technology, people, and processes to reduce human effort and deliver outcomes more efficiently.

“At Microsoft, we see a lot of benefit to applying continuous improvement and AI together,” says Becky West, leader of the Continuous Improvement Center of Excellence within Microsoft Digital, the company’s IT organization. “The way we do that is, is first initiate our continuous improvement workflows, which makes our processes as efficient as possible, and then we apply AI. Conducting continuous improvement in that order keeps you from automating a broken process and focusing AI’s abilities in the wrong direction.”

Our Continuous Improvement CoE works closely with our overarching, companywide Continuous Improvement team and our Microsoft Digital AI CoE, which is responsible for guiding our internal AI transformation. These organizations partner with each other and teams across Microsoft Digital to build our AI and continuous improvement muscles as an organization.

The result is a process of constant, iterative improvement aligned with our organizational goals, with AI as one of its most powerful drivers.

“Continuous improvement has been around for decades, but AI is providing new opportunities for process improvement because the technology has reached the level of maturity where it can fill gaps and smooth corners,” says Nitul Pancholi, principal and leading member of the AI CoE. “AI is ready for action, and continuous improvement is here to show us how to use it best.”

Continuous improvement in action

Continuous improvement provides us with a structured methodology for establishing rapid learning cycles that deliver tangible improvements across our processes, whether those enhance security, quality, delivery, innovation, or productivity. It helps us apply rigor to solving the right problems to capture the value we want.

“Culturally, continuous improvement helps us emphasize progress over perfection and do a little bit better every day.”

Matt Hansen, director, Continuous Improvement, Microsoft

We define a high-performing continuous improvement system according to four principles:

1. A clear definition of winning based on expectations
   We define success through the lens of what our teams value most and align our priorities accordingly.
2. Disciplined execution
   Teams operate within a simple, repeatable rhythm consisting of four stages: Plan, Do, Check, Adjust.
3. Constrained problem-solving with urgency
   We focus on root causes, not symptoms, and tackle problems head-on.
4. Sustained replication and acceleration
   When we discover improvements, we standardize and embed them into our operations to compound improvements.

“Culturally, continuous improvement helps us emphasize progress over perfection and do a little bit better every day,” says Matt Hansen, a director of Continuous Improvement at Microsoft. “It’s really about understanding your teams, their needs, and the value they can deliver to the business, then focusing their efforts to become as efficient as possible and do it all at scale through replicable techniques.”

The process centers on the disciplined execution of a four-stage cycle: Plan, Do, Check, Adjust.

Our continuous improvement system

Our Continuous Improvement System represents a disciplined approach to process improvement that leads to a virtuous cycle. “Do” is only one phase, which means we don’t take action for action’s sake. Instead, deliberate planning guides our projects, and a highly intentional approach to measurement helps us adjust initiatives as we learn what works and what doesn’t.

To put this system into practice, our continuous improvement methodology includes several processes and tools that underlie a structured approach to identifying opportunities, executing changes, and learning from our experiences to drive greater impact. These processes are informing and guiding several initiatives already in progress within Microsoft Digital.

The Bowler Method

Named for the scorecards used in bowling, Bowlers provide a disciplined method and visual tool for defining, tracking, and driving the outcomes that matter most. They define ownership, use clear KPIs arrayed along a linear progression, and identify urgent actions to close performance gaps. We review our progress on these cards during monthly operating reviews to develop a consistent habit of accountability.

We translate strategy to execution by cascading ownership, accountability, and measurement to the point of impact where the work happens and results are produced. That involves two levels of Bowler scorecards:

• The top-level Bowler defines the enterprise outcomes that matter most, tied to customer value, growth, cost, and risk.
• Cascaded Bowlers translate those outcomes into operational drivers that help deliver the desired results.

A Bowler scorecard

Action plans

Action plans are living tools that drive execution by clearly demonstrating what needs to happen, who owns it, when it needs to be done, and what impact it should have.​ This is where we initiate the Plan, Do, Check, Adjust cycle.

Gemba walks

As part of the planning process, Gemba walks involve observing teams and seeking an understanding of how they do their work. This concept originated from the Japanese phrase “Going to where the work happens.” During these sessions, we walk through a scenario to understand pain points, unpack the employee experience, and observe waste.

Kaizen events

A Kaizen event is a high-intensity, multi-day, in-person, team-based sprint designed to improve processes, solve problems, and close gaps. During the event, we define the opportunity, prepare for action by gathering data, observing work, and mapping processes, run a working session with the relevant parties, and deliver an improvement.

Value stream mapping

As part of a Kaizen event, value stream mapping involves charting the current state of a process to identify waste and pain points, pinpoint and prioritize improvements, design the future state, and create an improvement roadmap.

“Continuous improvement equips us to study our business processes in detail, uncovering how actions connect, where friction slows us down, and where automation can unlock new potential,” says Faisal Nasir, principal architect within Microsoft Digital, member of the AI CoE leadership team. “By combining this discipline with AI, we can turn those insights into transformative outcomes.”

Learning from our approach to continuous improvement

AI-driven continuous improvement initiatives are well underway at Microsoft, and some are producing results already, especially within Microsoft Digital. As the organization responsible for maintaining operational excellence and an exceptional employee experience within Microsoft, we’re applying AI-empowered continuous improvement to several different areas.

Our teams are using this framework to improve everything from third-party software license auditing to network hardware asset management. In one case, a new agent is helping designated responsible individuals (DRIs) on our Digital Workspace team save time resolving network outages, resulting in a 40% boost to a key network performance metric.

“What we’re building is a system of rigor around rapid cycles of accelerated learning to help us determine what works and what doesn’t for delivering the outcomes we want,” says Sammi Clute, a director of Continuous Improvement at Microsoft. “At the most basic level, it requires discipline around metric setting and review, but it also relies on establishing better connections between financial outcomes and executional work.”

As a result of our experience, we’ve established a process for launching continuous improvement initiatives. If you’re considering ways to use continuous improvement in support of your own AI projects, you may want to incorporate elements of our workflow.

First, think about who should be involved. Everyone has a role to play. When done properly, these efforts will have both horizontal and vertical implications, reaching across different teams and functions to foster participation at every level of the organization.

At Microsoft, we identify two major groups of stakeholders:

• The leadership team, responsible for defining business priorities and corresponding key metrics, assigning responsibility, and setting expectations for targets and pace.
• The execution teams build and test bowler cards, create and own execution plans, cascade key metrics if necessary, and conduct the work behind the initiative itself.

From there, consider working through a four-step process similar to Microsoft’s:

1. Understand your business’s priorities
   Clarify what matters most in terms of customer and stakeholder expectations, identify the capabilities your AI initiatives need to deliver, assemble clear and distinct priorities, and stack rank them.
2. Build your top-level bowler
   Once your business priorities are clear, you need to translate them into key metrics and assign ownership. Establish your measurements, set ambitious but realistic targets, and build out your bowler card.
3. Cascade your top-level bowler
   Connect your strategic priorities to the work that drives impact. This is where you clarify accountability, define how you measure progress, and ensure everything aligns with business goals. It’s largely a process of breaking your priorities into components and identifying leading indicators of success.
4. Build action plans for delivery
   With your measurement framework in place, it’s time to translate your key metrics into concrete, time-bound activities with named owners and clear outcomes. Start from your goals, define the components of progress, build actions to push those forward, identify enablers or prerequisites, and quantify the impact you want.

These four steps articulate the process, but AI has the potential to make it transformative. As you’re building out your action plans, put thought into ways that AI can enable continuous improvement by doing what it does best: automating routine tasks, eliminating waste, and augmenting decision-making.

Continuous improvement as a cultural mindset

As we move into the AI-powered future, we’re betting big on continuous improvement as a core methodology to ensure we put it to good use. By modeling this process internally, we’re creating a blueprint that enables our customers to reimagine their own operations.

“As Microsoft’s IT organization, we’re typically the first to scale out new technologies and processes,” says Nathalie D’Hers, corporate vice president of Employee Experience. “Now we’re excited to take on a leading role in defining how AI and continuous improvement work together to accelerate transformation, empower employees, and unlock greater value than before.”

All of this has required both cultural and operational development.

“Continuous improvement will serve as an engine for accelerating transformation and impact from AI.”

Carolina Dybeck Happe, executive vice president and chief operations officer, Microsoft

The “learn-it-all” mentality that Satya Nadella instilled in Microsoft when he became CEO in 2014 has been a crucial foundation as we operationalize and reinforce AI-enabled continuous improvement opportunities. In many ways, it’s an extension of the growth mindset that permeates our company’s cultural makeup.

We’ve also had to ensure we have the guidance, sponsorship, and skilling in place to help teams and individual employees feel comfortable taking accountability for continuous improvement initiatives.

It doesn’t happen by accident, but the effort is worth it.

“AI is a catalyst for innovation, growth, and value creation,” says Carolina Dybeck Happe, executive vice president and chief operations officer at Microsoft. “As we build a system of rigor around how we use these next-generation tools, continuous improvement will serve as the foundation for evolving Microsoft into an AI-driven Frontier Firm, transforming work as we know it to achieve more.”