1. AI Fluency Starts with Mandate + Mindset

In 2023, Zapier issued an “AI Code Red”—a company-wide challenge to get hands-on with AI tools for driving real business value. It was a clear signal that experimentation was part of the culture.

‍

“It was an invitation, really a challenge… to find practical ways to use [AI] to help build our company.” — Brandon Sammut

‍

The takeaway? Signal urgency and align mindset through leadership-led communications.

‍

Zapier’s “AI Code Red” wasn’t just a catchy headline—it was a strategic cultural move. By authoring a company-wide blog post, their CEO Wade Foster made AI a shared priority, reframing it as both a responsibility and an opportunity. Don't wait for top-down directives to "arrive" organically. Create your own urgency signal. Use leadership voices (e.g. internal blogs, town halls, kickoff messages) to publicly frame why AI fluency matters now—for your customers, your company, and your people. Clear, bold communication sets the tone for a shift in mindset and behavior.

2. Hack Weeks as Fluency Accelerators

Zapier embedded AI learning into its quarterly Hack Week, leading to guided experimentation and engagement—without creating new rituals.

‍

“It’s guided, purposeful experimentation. And it created energy while minimizing disorientation.”

‍

The takeaway? Integrate AI fluency into existing rhythms like hackathons or innovation sprints.

‍

Zapier embedded AI experimentation directly into existing company traditions—namely, their quarterly Hack Week. This minimized disruption while maximizing engagement and learning-by-doing. Leverage rituals your org already trusts—like design sprints, innovation days, or quarterly planning—as containers for applied AI learning. Structured play leads to faster fluency than standalone training. Encourage teams to bring real work problems and leave with real solutions.

3. Peer-to-Peer Learning Scales Internal Expertise

Zapier designated AI-savvy teammates as permanent go-to supports, codifying peer teaching into daily workflows.

‍

“Part of your day job is helping your peers build fluency.”

‍

The takeaway? Design systems where AI knowledge-sharing becomes part of everyone’s role—normalizing peer support.

‍

Zapier designated “power users” as internal go-tos for questions, and they actively supported AI-curious teammates in Slack—turning learning into a shared responsibility. Identify internal AI champions early and give them air cover (and recognition) to become embedded coaches. Set up always-on forums (e.g. AI channels, office hours, async Q&A docs) where questions are welcomed, not feared. Normalize asking, sharing, and celebrating experiments—especially ones that fail and teach.

4. A Flexible Fluency Framework—From Unacceptable to Transformative

Their open-source AI Fluency Rubric defines four tiers:

• Unacceptable—resists AI use
• Capable—knows how to use tools
• Adoptive—uses AI to improve performance
• Transformative—rethinks work from the ground up
  
  

“It’s not a check-box skill. It’s a capability you build—and it looks different by role.”

‍

The takeaway? Build fluency models tied to role-specific expectations—both for hiring and upskilling.

‍

Zapier’s rubric has four tiers: Unacceptable, Capable, Adoptive, and Transformative. It’s used across hiring, onboarding, and team development.

Don’t rely on vague expectations like “be AI-savvy.” Instead, co-create a rubric that’s role-specific and behavior-based. Define the mindsets, tool fluency, and outcome-driven examples you expect at each level—then embed it into recruiting, feedback, and development. When expectations are explicit, learning becomes intentional.

5. Durable Skills Matter Even More in an AI World

Zapier emphasizes critical thinking, empathy, and craft depth—knowing that technical skills (e.g., prompting) may evolve over time.

‍

“Durable skills outlast changing modalities. Prompting may change—but curiosity and agency don’t.”

‍

The takeaway: Invest equally in timeless skills and tool-specific capability.

‍

Zapier intentionally values critical thinking, problem scoping, and communication—recognizing that prompting syntax will evolve, but strategic thinking won’t. In an AI-first world, humans must become better problem definers, not just faster prompt writers. Design learning that develops judgment, collaboration, creativity, and systems thinking. These are the foundations of transformational work—no matter the toolset.

6. AI Coaching in the Flow of Work

They developed a 1:1 AI coach that analyzes meeting transcripts to surface actionable feedback—right where work happens.

‍

“We couldn’t observe one-on-ones before. Now AI embeds coaching into the workflow.”

‍

The takeaway: Explore AI tools that offer feedback in-the-moment—like a coach over your shoulder.

‍

Zapier’s “1:1 Coach” analyzes meeting transcripts and gives managers feedback on quality, balance, and behavior—automatically. Don’t treat coaching and training as separate from work. Use agents, copilots, or AI prompts to embed nudges, suggestions, and reflections directly in flow (e.g. after meetings, drafts, or tickets). The closer learning is to the moment of need, the higher the impact—and adoption.

7. Fluency Anchored in Team Goals, Not Tools

Zapier aligned all AI initiatives with team missions and KPIs, rather than stand-alone features.

‍

“AI is a medium—not the outcome. The goal is better work.”

‍

The takeaway: Anchor learning in outcomes you already care about—making it meaningful and relevant.

‍

AI fluency efforts at Zapier were always anchored in team missions and core KPIs—like “fast, accurate support” in customer care. Avoid tool-first rollouts. Instead, ask teams: What outcome are you solving for? What friction or opportunity exists? Then explore where AI can help. When fluency is tied to clear business goals, learning becomes relevant—not theoretical.

8. Adoption Depends on Trust & Transparency

The missing ingredient in many AI rollouts? Trust. Zapier emphasized the need to proactively explain how data is stored, used, and secured.

‍

“Explain data use—don’t assume people will ask.”

‍

The takeaway: Build transparency into AI experiences—especially when they involve sensitive information.

‍

Zapier found that even when AI coaching worked well, adoption stalled unless users knew exactly how their data would be used, stored, or accessed. Trust is table stakes. Clearly articulate what data is collected, who sees it, and how long it lives. Make privacy policies visible in the flow of tool usage. Invite questions early—and answer them before they're asked. Psychological safety drives experimentation.