Dive Club Episode Summary: Sam Stevenson - The Winding Journey of Designing a Beloved AI Product

Podcast Information:

• Title: Dive Club 🤿
• Host: Rid
• Episode: Sam Stephenson - The winding journey of designing a beloved AI product
• Release Date: March 28, 2025

Introduction

In this enlightening episode of Dive Club, host Rid engages in a deep conversation with Sam Stevenson, the co-founder and designer of Granola, a rapidly rising startup in the AI-powered workspace tools sector. Granola has quickly become an indispensable tool in Rid's own workflow, highlighting its growing reputation in the design and productivity communities.

Founding Story

Sam Stevenson narrates the humble beginnings of Granola. Both Sam and his co-founder Chris quit their jobs to explore side projects, eventually converging their interests to create a product that leverages AI to enhance note-taking during meetings.

Sam Stevenson [01:14]: "We met like halfway through [quitting our jobs], started hanging out and jamming on side projects together... Chris had discovered GPT-3, thinking it was going to change everything in technology."

Their initial idea centered around creating an AI-driven note-taking app that could organize and manage meeting notes more effectively. However, early discussions revealed the pitfalls of building a product based purely on top-down design without understanding real user needs.

Sam Stevenson [02:50]: "Realistically, useful things get built by figuring out a point in someone's workflow where they're struggling... and focusing on that."

This realization led them to pivot from abstract concepts to identifying and solving specific problems faced by users in their daily workflows.

Early Product Experiments

Sam and Chris embarked on extensive user research, conducting interviews to unearth pain points in everyday work processes. They discovered that many users struggled with ineffective note-taking during meetings, leading to lost action items and increased stress.

Sam Stevenson [03:58]: "Observing what people actually do gives you very different results to just asking people what they think they do."

One of their initial prototypes featured an autocomplete function where users could type a key term from the meeting and Granola would flesh it out into a useful note. While innovative, this version faced usability challenges, such as the need for users to manage multiple interactions manually.

Sam Stevenson [08:18]: "It has to feel really effortless."

Despite various experiments, including real-time bookmarking and favoriting notes, many features didn't resonate with users, often being forgotten or underutilized during meetings.

Pivot to Granola

The turning point for Granola came when a key investor, Solea, showed support during a pivotal conversation in January [12:03]. This backing provided the necessary confidence for Sam and Chris to commit fully to Granola as a startup.

Sam Stevenson [12:03]: "He was into it... he said, I might be in. If you guys are thinking about raising, then count me in."

This support was instrumental in transitioning Granola from a side project to a focused startup endeavor.

Onboarding Challenges and Solutions

Initially, Granola lacked an effective onboarding process, making the product hard to understand for new users. Under the guidance of advice from YC partner David Leap, the team recognized the need to separate product utility from the onboarding experience.

Sam Stevenson [28:23]: "David Leap... said you're trying to solve two problems at the same time... it's kind of like doing it on hard mode."

To address this, Granola adopted a hands-on approach, conducting personalized video calls to onboard each user. This method, while resource-intensive, provided invaluable feedback and ensured users could effectively utilize the tool despite its early-stage limitations.

Sam Stevenson [29:55]: "We did, like, five a week... it wasn't a ton, but we learned a lot."

This strategy helped improve retention and fostered early advocates like investor Michael, who despite initial shortcomings, consistently used Granola and provided critical feedback.

Interface Design Philosophy

Sam emphasizes Granola's minimalist and spatial design philosophy, aiming to keep users in their "spatial brain" rather than overwhelming them with linguistic complexity. This approach is inspired by cognitive psychology theories, notably Daniel Kahneman's System 1 and System 2 thinking.

Sam Stevenson [18:20]: "We wanted granola to feel kind of spatial... keep the user in their spatial brain rather than just in the language... System 1."

Granola's interface is designed to be a single, cohesive workspace where notes and meeting summaries coexist seamlessly, avoiding the clutter of multiple panels and sidebars found in competing products.

Sam Stevenson [21:32]: "Ruthlessly prioritize the core flow of starting a meeting from the notification, jotting down a few things... happening seamlessly without you needing to Think about granola during the meeting."

Product Iterations and Refinements

Throughout Granola's development, the team engaged in continuous testing and iteration based on user feedback. One significant challenge was balancing the transcript's utility without making it the primary feature, ensuring that Granola remained focused on actionable notes rather than mere transcriptions.

Sam Stevenson [22:58]: "We put it in a 400 pixel box... people find it hard to trust."

To maintain simplicity, Granola relegates the transcript to a secondary role, accessible but not obtrusive, reinforcing the tool’s core purpose of enhancing meeting productivity through concise notes.

The team also tackled the complexity introduced by early prototypes by stripping down Granola to its essential features, ensuring a clean and intuitive user experience.

Sam Stevenson [16:55]: "We just added more stuff... made the product horrifically complex... eventually decided to simplify."

Future Directions

Looking ahead, Sam envisions Granola expanding beyond individual usage to facilitate team collaboration, integrating more deeply with other productivity tools, and automating follow-up actions based on meeting discussions.

Sam Stevenson [36:31]: "Ultimately, we want to build a tool that you're using for a lot of your day... invent the future of how people interact with software and how they work alongside this AI stuff."

He describes a future where Granola not only takes notes but also manages tasks and integrates seamlessly with platforms like Linear, allowing users to delegate actions directly from the meeting interface.

Key Lessons and Insights

Sam shares several valuable insights from his journey with Granola:

1. User-Centric Development: Building products based on observed user behavior rather than preconceived notions ensures higher utility and adoption.

2. Iterative Prototyping: Rapid experimentation and willingness to discard ineffective features are crucial for refining the product.

3. Effective Onboarding: Separating product development from onboarding allows for clearer focus and better user experiences.

4. Simplicity Over Complexity: Maintaining a minimalist design helps in creating intuitive and user-friendly interfaces.

5. Continuous Feedback: Engaging with users regularly and valuing their feedback is essential for ongoing product improvement.

Sam Stevenson [49:00]: "Really the only thing that matters is whether they pick it up again and again and does it really solve their problem."

Conclusion

This episode of Dive Club offers an in-depth look into the strategic and design decisions that shaped Granola into a beloved AI-powered tool. Sam Stevenson's candid reflections on challenges, pivots, and the relentless focus on user needs provide valuable lessons for designers and entrepreneurs alike. Granola's journey underscores the importance of adaptability, user-centric design, and the courage to iterate continuously in the pursuit of creating products that truly resonate with users.

Notable Quotes:

• Sam Stevenson [01:14]: "We met like halfway through [quitting our jobs], started hanging out and jamming on side projects together..."

• Sam Stevenson [03:58]: "Observing what people actually do gives you very different results to just asking people what they think they do."

• Sam Stevenson [12:03]: "He was into it... he said, I might be in. If you guys are thinking about raising, then count me in."

• Sam Stevenson [28:23]: "David Leap... said you're trying to solve two problems at the same time... it's kind of like doing it on hard mode."

• Sam Stevenson [49:00]: "Really the only thing that matters is whether they pick it up again and again and does it really solve their problem."

This comprehensive summary encapsulates the key discussions, insights, and conclusions from the episode, providing a clear and engaging overview for those who haven't had the chance to listen.