How to Unlock High-Impact AI Use Cases: Start With Your Organization, Not Your Tech Stack

Most AI initiatives stall not because the technology fails, but because organizations skip the foundational work that makes AI actually useful.

In the Domopalooza 2026 breakout session, AI Bootcamp: Unlock Your High-Impact AI Use Cases, Cody Irwin, AI adoption director, walks through a practical framework for finding, prioritizing, and prototyping AI use cases that deliver tangible business value. The approach centers on a simple but often overlooked truth: discovering how to unlock high-impact AI use cases requires organizational clarity first, technology second.

The session introduces a structured workflow built around four Ps—processes, plans, prioritization, and prototypes—designed to help teams move from scattered experimentation to focused execution. Whether you're just starting your AI journey or trying to escape an endless cycle of proofs of concept, the insights here offer a clear path forward.

Wire your organization for AI success

Treating AI adoption as purely a technology problem is the fastest way to end up stuck. Cody describes a common failure mode where companies spread AI tools across employees like peanut butter, expecting magic to happen. The result is often scattered effort, unclear returns, and what he calls the endless POC (proof of concept) phase.

"I'm finding more and more that the actual driver is the organization itself being wired in a way that AI works well," Irwin explains. He referenced a McKinsey framing that puts this in stark terms: "AI is 20 percent algorithms and 80 percent organizational rewiring."

The implication is significant. High-impact use cases emerge when three elements are in place: a clear purpose (the why), aligned people, and documented processes. Without these, AI becomes a magnifier of existing confusion rather than a driver of efficiency.

"AI is a magnifier," Cody notes. "If we don't have those things, AI hallucinates. And we blame it for it."

To set your organization up for AI success, consider these foundational steps:

• Establish your why: Document why your company exists and what it's trying to achieve. AI won't infer this for you.
• Align your people: Address fears and resistance by showing how AI helps rather than threatens. Start with work people genuinely dislike.
• Document your processes: Capture how work actually happens today, not just what's in people's heads. This becomes the foundation for reliable AI plans.

The shift from experimentation to transformation happens when teams move inside out—starting with organizational clarity—rather than outside in, chasing the latest technology.

Build an anti-to-do list first

Once organizational foundations are in place, the next challenge is figuring out where to start. With a thousand potential AI applications, picking the right ones matters more than picking many.

Cody introduces a concept he credits to a colleague in Europe: the anti-to-do list. Instead of asking what AI can do, ask what work people really do not want to do anymore. These monotonous, repetitive tasks become your first wave of AI opportunities.

"If you had a thousand interns today, what would you have them do?" Irwin asks, sharing a prompt that helps teams surface high-volume, low-value tasks ideal for early wins.

This approach reduces internal friction because it positions AI as a helper rather than a threat. When AI takes over the busy work—data entry, repetitive reporting, manual monitoring—people become more willing to experiment with broader applications.

The anti-to-do list method works because it aligns AI adoption with immediate pain points. Here's how to apply it:

• Survey your team: Ask each person to list three to five tasks they find monotonous or frustrating.
• Identify patterns: Look for common themes across roles and departments. These represent high-impact opportunities with built-in organizational buy-in.
• Start small: Pick one or two tasks to automate first. Early wins build momentum for larger transformation efforts.
• Measure relief, not just efficiency: Track not only time saved but also team sentiment. Adoption accelerates when people feel the benefit.

This strategy also helps reframe the conversation around AI. Rather than asking teams to learn new technology, you're offering to remove work they never wanted to do in the first place.

Prioritize for marginal gains, not moonshots

Even with a solid anti-to-do list, the temptation to chase big, innovative AI projects can derail progress. Irwin emphasizes that priority is king, and the best starting point isn't so much incrementation.

"We want to find things that are low effort, high impact. That's what we're going after," Cody explains.

The framework he recommends comes from McKinsey and BCG research on innovation investment. It suggests a deliberate mix: Most early investment should go toward incrementation (doing what you already do a little better), with smaller allocations to improvement (reinventing existing work) and innovation (doing entirely new things).

This approach mirrors the marginal gains philosophy made famous by British Cycling in the early 2003s; James Clear wrote about them in "Atomic Habits." Small improvements—better pillows, better food, better clothing—compounded into dramatic results. For AI adoption, the same principle applies: stacking 1% improvements builds confidence, capability, and measurable outcomes faster than chasing moonshots.

To apply this prioritization discipline, use an effort vs. business impact scoring model:

• Score each use case: Rate potential AI applications on two axes: expected business impact and implementation effort.
• Target the sweet spot: Focus on use cases that land in the low effort, high impact quadrant. These deliver quick wins without draining resources.
• Avoid the traps: Some ideas sound exciting but are hard to implement with limited return. The scoring model helps you spot these before investing.
• Review your distribution: Check whether your portfolio leans too heavily toward innovation. If so, rebalance toward incrementation until your team builds fluency with AI.

Cody points to Domo's catalog of documented AI use cases from customers, organized by department and integrations, as a resource for identifying opportunities that others have already validated. Starting with proven patterns reduces risk and accelerates time to value.

Key takeaways about finding AI use cases

The path to high-impact AI use cases runs through organizational readiness, not technology adoption. Here's what to keep in mind:

• Treat AI success as a transformation problem. Clear purpose, aligned people, and documented processes matter more than the tools you choose.
• Start with the anti-to-do list. Focus initial efforts on work people dislike, the busy work and monotonous tasks, to reduce friction and build momentum.
• Prioritize marginal gains over moonshots. Use an effort vs impact scoring model to identify low-effort, high-impact opportunities that deliver quick wins.
• Document current-state processes. Capturing how work happens today creates the foundation for reliable AI plans and reduces hallucination risk.

You now have a practical framework for moving from scattered AI experimentation to focused, high-impact execution. The tools and specific platforms matter less than the discipline of starting with your organization's real needs and building from there.

Ready to go deeper? Watch the full session from Domo to see the complete workflow in action, including live prototyping examples and additional resources for identifying your first AI use cases.