Scrum.org Community Podcast

What happens to organizations when the cost of outputs drops to nearly zero? That is the central question Dave West and Dr. Mik Kersten are back to explore in this episode of the Scrum.org Community Podcast. Mik, the author of the upcoming book Output to Outcome, joins Dave to dig into why so many companies are failing to see meaningful value from their AI investments and why the organization itself has become the bottleneck.
They cover the AI productivity paradox, why complexity is growing faster than outcomes in most large organizations, and what leaders need to do differently. Mik introduces the "outcome loop" as a mental model for leadership, explains why modularity and decentralization are the only structures that scale (in nature or in business), and shares why many organizations are quietly reverting to command-and-control patterns at exactly the wrong moment.
If you work in or lead a Scrum Team, manage a product organization, or are thinking about how AI is reshaping the way we structure and run teams, this conversation will give you a lot to chew on.
Output to Outcome by Dr. Mik Kersten is available for pre-order now, with release on July 14.

For 17 years, the Give Thanks for Scrum (GTFS) event in Boston has reflected the evolution of Scrum itself, from early education and experience reports to scaling, business agility, and now AI integration.

In this episode, Dave West sits down with GTFS founder and Managing Director of Improving Agility, Daniel Mezick to explore how the event began, sparked by bringing Jeff Sutherland and Ken Schwaber together during Thanksgiving week in Boston, and how its themes have evolved alongside the Agile movement.

The 2025 event returned to an in-person format and centered on three critical themes:

Delivering tangible results
Incorporating AI into Scrum practice
The evolving role of the Scrum Master

Daniel and Dave discuss why empiricism matters more than ever, how Scrum Masters must understand executive psychology, and what socio-technical design means in an AI-enabled world. They also explore whether large agile frameworks may give way to more focused, empirical approaches.

Looking further into the future, the conversation turns to AI’s expanding role, practical application over hype, and a blended in-person/online format to grow the global community.

If you care about the future of Scrum, stakeholder delight, and navigating uncertainty with empiricism, this episode is for you.

Dave West sits down with Mik Kersten, author of Project to Product and the upcoming Output to Outcome, to explore why AI amplification is exposing the real bottlenecks in how organizations work. Mik shares data from over 3,600 value streams showing that development teams account for just 8% of end-to-end delivery time which means making those teams faster with AI doesn't move the needle if the constraints are upstream and downstream.
The conversation digs into why most organizations are measuring the wrong things (hint: token consumption is not a productivity metric), why overlay agile structures have largely failed, and why the answer isn't fewer teams it's more empowered ones. Mik introduces the core models from his new book: the outcome loop, the outcome tree, and seven organizational shifts that together make up a new operating model designed for the age of AI.
Key Takeaways:
The bottleneck has moved from software delivery to planning, governance, and innovation and most organizations haven't caught up
Making development teams faster with AI delivers little value if the surrounding system isn't designed around outcomes
Agile as an overlay structure doesn't work it has to become the primary operating model and the actual org chart
Empowered, autonomous teams are not optional in an AI-driven world the speed of feedback loops makes half-measures unsustainable
Leadership roles need to be redefined and incentive structures realigned to match the way teams are actually working
The theory of constraints still applies in the age of AI the constraint just keeps moving, and finding it is now the critical management skill
Links
http://outputtooutcome.org/

Neste webcast “Conversa com PSTs: Scrum, o dia a dia e a IA”, dois Professional Scrum Trainers (PSTs) respondem às perguntas da comunidade e exploram como os pilares do empirismo em Scrum: transparência, inspeção e adaptação, são aplicados ao desenvolvimento de produtos, e nas situações do quotidiano, profissionais e também pessoais.
Durante a conversa, dois PSTs, Daniel Carrilho e Matheus Reis, irão refletir sobre como a Inteligência Artificial está a transformar a forma como as equipas trabalham, como pode apoiar a tomada de decisões baseada em evidência e de que forma líderes e profissionais podem tirar partido do Scrum para navegar num mundo cada vez mais complexo e assistido por AI.

In this episode of the Scrum.org Community Podcast, Dave West is joined by Tony Hinkley, Chief Technology Officer for Avanade UK. Tony brings a rare combination of perspectives: seasoned engineer, Professional Scrum Trainer, and senior technology leader helping some of the world's largest organizations navigate AI adoption. Together, Dave and Tony dig into what AI is really doing to product teams, Scrum practices, and knowledge work at large. This is an honest, experienced, and energizing conversation about where we are, where we're headed, and what it means for everyone who cares about delivering value professionally and sustainably.

Key Takeaways
AI has disrupted professional services more than almost any other sector because knowledge, the core asset of consulting and IT services, is rapidly becoming a commodity. That's a wake-up call for all knowledge workers.
Scrum Teams are seeing real productivity gains from AI  but those gains have moved the bottleneck. It's no longer about how fast developers can write code. It's about the quality of intent, requirements, and context being handed to the tools.
The principles behind Scrum haven't changed but how you implement them must. Your Definition of Done, for example, may now be enforced by an agent rather than a person. Are your standards clear and documented enough for that to work well?
Specialists get significantly better results from AI than generalists. First-principles thinking, clean code habits, and a strong sense of what "good" looks like are more valuable now than ever, not less.
Context is the new currency. Giving AI tools access to well-structured, well-governed organizational data and standards will unlock far more value than simply upgrading to the latest model.
Leaders face a real choice: use AI to cut costs, or use it to grow. Tony's strong recommendation is to invest freed-up capacity into the parts of your product organization that have always been under-resourced, strategy, ideation, stakeholder engagement, and product thinking.
Data governance isn't a dirty word anymore, it's a competitive advantage. Organizations that get serious about data quality, classification, and security will be the ones that get the most from AI. Garbage in still means garbage out, and the consequences are bigger than ever.
Mid-market companies should pay close attention. AI is leveling the playing field in a meaningful way, giving smaller organizations the ability to punch well above their weight in product delivery.
If you're feeling uncertain about your place in an AI-enabled world, start by embracing the tools. As Tony puts it: AI won't take your job, but someone who knows how to use it well might.