Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

AI success starts at the top

Leadership needs to build an environment where experimentation thrives.

[Photo: Getty Images]

Profile photo of Andy MacMillan

CEO · Alteryx

The pressure to take meaningful action on AI is real, and it’s growing. It represents a once-in-a-generation opportunity to rethink how we work, solve problems, and drive growth. Yet, alongside this opportunity comes significant noise and uncertainty. 

Amid the rush to launch pilots, buy or build tools, and keep up with competitors, it’s easy to lose sight of what matters: building an environment where experimentation thrives. 

And that kind of culture doesn’t emerge by chance. It starts at the top. 

As CEO, I don’t pretend to be the company’s top AI expert. But I am accountable for building a company where AI innovation can take shape and scale. 

Innovation starts with curiosity 

Throughout my career, I’ve found that innovation doesn’t start with technology; it begins with a question. At Alteryx, we didn’t mandate that teams use AI. Instead, we asked: Where are we spending time on repetitive tasks? What slows us down? What could we do differently?  

The answers came from the front lines. 

  • Our sales reps were spending hours crafting outreach emails. We trained a GPT model on our content and built an agent that drafts personalized messages in seconds. This freed up our team to focus on relationships rather than templates. 
  • Dashboards that once required team huddles now use AI to flag anomalies and suggest actions automatically. The endless meetings that could have been emails don’t need to happen because the team already has the insight it needs. Now, when we meet, it’s purposeful and informed. 

These weren’t sweeping, months-long initiatives. They were quick, practical pilots created by the people doing the work, using the tools they already knew. That’s the kind of innovation that sticks. 

Fuel curiosity with access and agency 

Telling people to think differently simply doesn’t work if they don’t have access to the right tools or the freedom to use them. 

We’ve made it a priority to democratize access to AI. That means providing employees with enterprise-grade GPT models, cloud data, and low-code tools they’re already familiar with. For teams who need more hands-on support, we offer practical training grounded in real-world use cases, like using generative AI to summarize customer calls or to spot trends in data. Because investing in technology is only part of the equation; investing in your people’s ability to explore and experiment is just as essential. 

Fail forward, together 

Let’s talk about failure. 

If your culture punishes risk, don’t expect innovation with AI or otherwise. At Alteryx, we’ve worked hard to normalize experimentation and even celebrate the learning that comes from getting it wrong. 

At our all-hands meetings, we spotlight bold AI use cases. We ask teams to share what they tried, what worked, and what didn’t. This transparency normalizes the fact that some projects simply don’t work out and may even spark another team’s breakthrough. 

Allowing employees to share their stories of AI experimentation makes others more likely to try something themselves. The audience learns what missteps not to take in their own projects. But even more than that, they know it’s okay to make mistakes and that people who take calculated risks are, in fact, rewarded. 

Guardrails that empower 

Of course, none of this means being reckless. Responsible AI innovation still requires structure. However, that structure should empower experimentation, not shut it down. 

That’s why we created an internal AI Data Clearinghouse concept. It’s a mechanism for employees to propose new use cases that leverage company data, with clear review paths for legal, security, and compliance teams. We don’t expect every employee to become an expert in data regulation, but we do expect them to explore ideas and surface them responsibly. 

Think of it as a safety net, not a cage. The goal is to reduce risk without killing momentum. If you’re sitting on the sidelines while competitors deploy AI-generated code or autonomous agents, you’re going to hit a wall. And when you do, your board will ask, Why didn’t we explore this sooner? 

You don’t have to go all-in on day one. Start small. Focus on low-risk, internal use cases: sandboxed environments with no exposure to core IP. Learn by doing, not by waiting. 

I also recommend establishing a subcommittee, or at least a dedicated working group, to evaluate both sides of the equation: the risks of mismanaging AI and the risks of inaction. And don’t wait for the next quarterly cycle to reassess. The technology is moving faster than that; your strategy needs to keep pace. 

AI is everyone’s job 

You might feel that you need to wait for the “right” team to take the lead. 

The truth? The right team is already here. It’s your people, from sales to support to operations, asking better questions, sharing early wins, and rethinking their roles. 

Yes, it helps to have a small center of excellence: a group that can support governance, training, and policy. But the real breakthroughs happen when teams don’t wait for permission, when an operations lead pairs with an analyst, when a support rep learns how to build a bot, when people take initiative and collaborate across silos. 

AI isn’t a side project. It’s how we work now. 

That’s why we value high-agency employees who not only take direction but also take ownership. And it’s why we’re building a culture where that mindset is rewarded, not reprimanded. 

Lead by example 

True innovation happens when leaders remove blockers, show curiosity, and give their teams the clarity and tools to think differently. 

This culture change doesn’t come from catchy slogans. It’s about what leaders do day in and day out. If your teams don’t feel empowered to try, share, and improve, AI won’t transform your business. 

And if you’re waiting for certainty, you’re already behind. Innovation belongs to the bold, and even more to those who feel empowered. 

Andy MacMillan is CEO of Alteryx.