How I AI: Brian Greenbaum's 3-Step Playbook for Driving Company-Wide AI Adoption

So many of our episodes focus on specific, tactical workflows for building things with AI. But a question I get all the time is, “This is great for me, but how do I get my entire team on board?” It's one thing for one person to be an AI power user; it's another thing entirely to transform a whole organization.

That’s why I was so thrilled to sit down with Brian Greenbaum, a product designer at Pendo. He faced this exact challenge and developed a brilliant, step-by-step plan for getting his product and design teams to not just use AI, but to embrace it as a core part of their workflow. It’s a story that starts, fascinatingly, while he was on paternity leave and had a personal epiphany with an AI coding tool.

In this episode, Brian unpacks his entire playbook. We're not just talking about theory here; he shares the exact Slack message he sent to leadership, the structure of his interactive workshops, and the framework he used to measure the program's success. This is the ultimate guide for anyone who has raised their hand and said, “I want to lead our team’s AI transformation.” It's one of the biggest leadership opportunities available right now, and Brian shows us exactly how to seize it.

Let's dive into his three core workflows for kickstarting, structuring, and scaling AI adoption in your company.

Workflow 1: The Inception - Sparking the AI Transformation

Every major company initiative starts with a spark. For Brian, that spark came from a personal project and a powerful realization that AI could fundamentally change how his team at Pendo built products. Here’s how he turned a personal “aha” moment into the catalyst for a company-wide movement.

Step 1: Have a Profound Personal Experience

While on paternity leave, Brian, a self-described tech geek, started playing with Cursor, an AI-first code editor. He had a side-project idea for a music app that would let him scan a QR code on a physical card to play a specific album on Spotify, mimicking the tactile experience of a record player. Though not an active developer, he was stunned by what happened next.

“I pulled up cursor and like within a couple hours I had a working prototype and like, that just blew me away. I was creating QR codes, I was creating PDFs. I was like doing all this like really, really, really cool stuff.”

This wasn't just a cool hobby project; Brian immediately saw the professional application. At Pendo, he works on analytics-based features, and creating realistic, data-driven prototypes in Figma is a constant challenge. He realized he could use tools like Cursor to build high-fidelity, code-based prototypes that would communicate ideas far more effectively than static mockups.

Step 2: Craft a Compelling Message to Leadership

Brian knew he couldn't keep this to himself. While still on leave, he drafted a detailed message on Slack to his manager, his manager’s manager, the CPO, and other AI enthusiasts at the company. He didn't just say, “AI is cool”; he framed it with a clear, two-part business case:

1. Internal Efficiency: The product team could “leverage the cutting edge of AI tools to get more done in fewer hours and less resources, improve decision making and communicate and validate ideas more effectively.”
2. External Thought Leadership: By becoming proficient in AI, Pendo could better serve its customers undergoing similar transformations and position itself as a thought leader.

This message was so powerful that the CPO immediately asked him to present at the next all-hands meeting. Brian had successfully created the buy-in needed to get a formal initiative off the ground.

Workflow 2: The Playbook - Structuring Your AI Adoption Program

Getting leadership excited is one thing; translating that excitement into a tangible program that people actually engage with is another. Brian developed a brilliant two-pronged approach to foster learning and experimentation.

Step 1: Combine Synchronous and Asynchronous Learning

Brian recognized that the biggest barrier to AI adoption is time. Everyone knows it’s important, but they’re too busy to figure it out. His solution was to create both dedicated time on the calendar and a space for continuous, self-paced learning.

• Synchronous: Bi-weekly “Product AI” Sessions: These were scheduled meetings designed to be interactive and hands-on. The goal wasn't just to present information, but to get people’s hands dirty.
• Asynchronous: A Public Slack Channel: This channel became the hub for what Brian calls “radical many-to-many sharing”—a place for anyone to share interesting articles, post experiments, and ask questions.

Step 2: Make the Synchronous Sessions Interactive

For his kickoff session, Brian didn't just talk about AI. He made everyone use AI, live. He designed a simple, fun, and eye-opening exercise using an app-building tool called bolt.new.

Provide a Simple Prompt: He had everyone in the session navigate to the tool and paste in the exact same prompt to create a basic to-do list application.

Create a simple to-do list application where a user can add tasks, mark them as complete, and delete them. The interface should be clean and minimalist.

Observe the Results: The most powerful moment came when everyone saw their results. Even with the identical prompt, the AI generated a wide variety of different applications. Some even produced errors, which became a teachable moment about iterating with AI.

Encourage Creative Exploration: Finally, Brian encouraged everyone to “go wild” and experiment with creative modifiers. He wanted to break the mental model of only building the “Minimum Viable Product” and reignite the muscle for imagining what’s possible.

Prompts like “add a retro 8-bit pixel art theme,” or “make it look like MySpace from 2007” got people laughing and experimenting, showing them the creative potential of these tools.

This hands-on approach demystified the technology and made it feel accessible and fun, while the Slack channel ensured the conversation and learning continued long after the meeting ended. We even saw examples of designers using Midjourney to create delightful animated characters for UI, something that would have been too time-consuming before.

Workflow 3: The Flywheel - Measuring and Scaling Adoption with a "Golden Path"

To ensure the initiative had a lasting impact, Brian knew he needed to measure its effectiveness and create a safe, scalable framework for AI usage across the company. This involved moving from grassroots enthusiasm to a structured, company-supported program.

Step 1: Measure Success with Sentiment Surveys

As part of a company-wide OKR to improve AI leverage, Brian’s group started by getting a baseline. They sent out a “sentiment survey” to understand how employees felt about AI and how much they knew about the company's policies.

The survey asked five key questions about:

• Their personal sentiment toward AI's impact.
• Their feeling about AI's potential positive impact on employees.
• Their familiarity with the company's AI usage policy.
• Their awareness of which AI tools were available to them.

They ran the survey at the beginning and end of the quarter. The results were dramatic: after implementing their programs, they saw a significant positive increase across all metrics, especially in awareness of the usage policy and available tools. This proved their efforts were working.

Step 2: Create a “Golden Path” for Tool Usage

The biggest gap the survey revealed was a lack of clarity. People were using their personal ChatGPT accounts and didn't know what data was safe to use, which tools were approved, or how to get licenses. To solve this, the team collaborated with legal, security, IT, and finance to create an AI Knowledge Center on Confluence.

This centralized document became the single source of truth and included:

• An alphabetized table of all approved AI tools.
• Clear data-sharing guidelines for each application (e.g., “Internal Data Only,” “No PII”).
• The security and legal status of each tool.
• A straightforward process for requesting access or new tool evaluations.

This “golden path” eliminated confusion and risk, empowering employees to experiment safely. It replaced “shadow IT” with a culture of transparent, enabled innovation.

The Result: From Experimentation to Roadmap Influence

The ultimate proof of this program's success is how it empowered employees like Brian to influence the company's direction. He used his newfound skills to build a prototype MCP (Model-Context-Protocol) server that connected to Pendo's public APIs. He then recorded a demo showing how he could use a natural language prompt in Claude to generate a beautiful, interactive dashboard of product usage data.

This wasn't just a cool demo; it was a powerful illustration of AI's potential. It caught the attention of the CTO and directly impacted Pendo's product roadmap, accelerating the development of internal AI agents.

Conclusion: Your Playbook for AI Leadership

Brian Greenbaum’s journey at Pendo provides a powerful and practical playbook for anyone looking to become an AI change agent. It’s a masterclass in moving from a personal passion to a structured, measurable, and impactful company-wide initiative.

By following these three workflows—sparking the initial transformation, structuring a program for learning and experimentation, and creating a scalable “golden path” for adoption—you can build a true culture of AI innovation. As Brian’s story shows, this isn't just about making your teams more efficient; it's about unlocking new levels of creativity and directly influencing the future of your products.

I can't emphasize this enough: if you have the initiative and the energy, raising your hand to lead this charge is one of the most significant career-building opportunities available today. So take this playbook, adapt it to your organization, and start building the future.

Brought to you by our sponsors:

Thank you to our amazing sponsors for supporting the How I AI podcast:

• Google Gemini: Your everyday AI assistant
• Lovable: Build apps by simply chatting with AI

Episode Links

• YouTube: Watch the full episode here
• Spotify: Listen on Spotify
• Apple Podcasts: Listen on Apple Podcasts

Find Brian Greenbaum on LinkedIn and check out Claire Vo and ChatPRD for more AI product resources.