Opening Keynote: Breaking Conway's Law--or How to Work Differently and Not Ship Your Org Chart
Summary
Key Insights
-
•
Conway's Law means products often unintentionally reflect organizational silos, making collaboration across functions difficult.
-
•
All internal stakeholders can be right in their perspectives, but without shared customer focus, conflicts arise.
-
•
Customer obsession must be prioritized over revenue, competitors, technology, or business models to align organizations.
-
•
Leaders like CPOs and CEOs should spend regular, agenda-free time directly with customers to build empathy.
-
•
Real customer field research uncovers nuanced needs that proxies like surveys, sales feedback, or fictional personas miss.
-
•
Fictional personas dilute real customer complexity; using actual customer stories and photos is more effective.
-
•
Reframing product design around actual customer workflows—e.g., salespeople writing emails and holding meetings—can rapidly improve products.
-
•
Early, low-fidelity prototyping like sketches encourages honest, democratic feedback from customers.
-
•
Customer value drivers must dictate prioritization and metrics, as seen at Amazon (price, speed, selection) and Intuit (money, no work, confidence).
-
•
Aligning OKRs and dashboards to mission-driven customer metrics ensures product teams remain accountable to real user outcomes rather than shareholders alone.
Notable Quotes
"The products that you ship are direct reflections of your organizational processes and communication structures."
"Everyone has a different perspective and partial knowledge, and all are right—but that creates silos."
"Customer obsession instead of competitor obsession or technology obsession must be really clear across teams."
"Having your CEO and CPO spend two days a month in the field with customers without any agenda is critical."
"Sales objections are not customer insights; they serve different purposes and must be treated separately."
"Pipetting at a biology bench gives empathy no survey or photo can provide."
"Don’t make up fictional personas; use real people’s names, photos, and stories to capture real customer complexity."
"Showing sketches to customers early encourages them to redraw and better express their needs."
"If our employees are happy, they will take care of our customers—and shareholders will be just fine."
"Are we truly a customer-first company, or a shareholder-first company? That shows up in what we measure."
Or choose a question:
More Videos
"We need just enough structure and no more to avoid chaos or stifling control."
Stephen AndersonPuzzled? How to Coordinate Humans for Complex Challenges
May 27, 2021
"Nobody actually cares that we’re better researchers. The question should be, what makes us different? What makes us distinctive?"
Peter LevinSolve a Problem Here, Transform a Strategy There: Research as an Occasion for Expanding Organizational Possibility
March 25, 2024
"A shared understanding of how you got here creates a sense of purpose, not just urgency."
Alastair SimpsonDebunking the Myths of Cross-Disciplinary Collaboration
October 24, 2019
"Constant communication is necessary; agreement emails can’t be one-and-done."
Alicia MootyDesign Staffing Models
September 30, 2021
"Don’t write a book expecting to retire on royalties; this is a specialized field with modest sales."
Louis RosenfeldCoffee with Lou: Should You Write a (UX) Book?
March 7, 2024
"Community advisory boards were elected by their peers to review findings and ensure accuracy."
Sarah FathallahLessening the Research Burden on Vulnerable Communities
March 30, 2020
"Without buy-in, you can’t get budgets approved or tools put in place to grow research practices."
Clemens Janssen Jane DavisEfficiently Scaling Research as a Team of One
March 28, 2023
"True sustainability work requires cross-functional collaboration among product, design, engineering, and sustainability offices."
Chris HammondEmbedding sustainability into enterprise design and development: A journey towards "sustainability consciousness"
April 2, 2025
"Some AI companies charge for services even when their datasets are limited and can misrepresent certain populations."
Tom Armitage Carla Diana Kanene Ayo HolderDay 2 Panel: Looking ahead: Designing with AI in 2026
June 11, 2025