I’ve always dreamed big about travel, imagining epic adventures that would create lifelong memories for my family. I dream of taking my kids on an outdoor adventure to Utah — hiking through canyons, camping under the stars — but I hesitate. What if they hated it? What if I spent months planning, and a lot of money, only to realize they weren’t ready for it?

I see other families taking these big trips, making memories — do I really need to wait till my kids are more mature to get started?

Enter a weekend canoe trip on the Delaware Water Gap. We rented the gear, bought a few essentials, and tested the waters — literally. It wasn’t the grand vision. We learned what we loved, what we didn’t, and what we needed to change for next time, all while creating great family memories. That was when I realized: getting started doesn’t mean giving up on the big vision — it means prioritizing learning, momentum, and stacking wins today. We might have scaled back the vision to get started, but we are becoming the adventurous family I want to be in the future.

This is especially true when it comes to AI. The promise is enormous, but so is the uncertainty. How do we push forward without taking on more risk than the business can stomach? How can we get some return today for our efforts rather than waiting?

Enter the stacking ROI framework — your new best friend for navigating risk, reward, and the right level of investment.

Stacking ROI: How to Make Big Bets Without Going Bust

Here’s the truth: every big innovation starts as a small, tangible win. 
The key is stacking ROI — breaking down ambitious ideas into smaller, launchable experiments that de-risk investment while still delivering returns.

This visual maps this thinking beautifully:

Our initial MVP de-risks the investment and ensures value is created and built upon. Stacking ROI ensures we ground in and sell a bold vision, capturing learning and returns now and de-risking future investments.

• No Regrets Bets (Incremental Innovation) → These are the low-risk, quick-return moves. Think workflow automation, internal tools, or AI copilots that reduce repetitive tasks. They won’t change the world, but they make an immediate impact. Often this look like trying tools out internally before launching them externally.

• Calculated Bets (Adjacent Innovation) → This is where you start stretching. Maybe it’s introducing AI-driven recommendations to a small customer segment or using predictive analytics to improve operations. The risk is higher, but so is the reward.

• Big Bets (Disruptive Innovation) → These are the moonshots — the AI-powered product transformations that could redefine your business. They take the longest to pay off, but when they do, the ROI is game-changing.

If we only play in the No Regrets zone, we’ll never leap ahead. If we go all-in on Big Bets without careful experimentation, we risk blowing resources on ideas that don’t land. The trick? Build in stackable wins — momentum that funds and informs bigger moves.

De-Risking the Leap: What We Must Learn Before We Build

To de-risk AI innovation, we need to understand the riskiest assumptions behind our ideas. (I am sure you’ve heard that before here, here, and here). The best way to do that though? Get close to the work.

Too often, AI projects start with a high-level vision but fail because they don’t account for the gritty realities of day-to-day workflows. Before we write a single line of code, we need to go beyond a macro understanding of the problem and deep-dive into the actual workflows of our end users. What are they trying to accomplish? What slows them down? Where does human judgment matter most? These insights tell us what AI can and should automate — and what still needs a human touch.

So, Where Do You Start?

1. Describe Your Big Bold Vision. What are your hopes for how AI can transform your business, customer relationships, products? What trends are you seeing that make this future vision seem plausible? 
   Don’t just go big with no grounding, a vision is strategy and strategy requires consideration and planning.

2. Identify your “No Regrets” bet. What’s the smallest AI-driven experiment you can launch that delivers immediate business value related to that vision?
   We look for ways to start internally and small. This might feel like efficiency gains today but the goal and return is linked to bigger metrics that matter.

3. Map out your Calculated Bet. If your first experiment works, what’s the natural next step that pushes innovation without taking on too much risk?
   Segmenting your users internally or externally into smaller cohorts (or identifying your High Expectations Customer — HXC) enables a slow release and learn pattern.

4. Define the return on your Big Bet vision. What is the size of the prize we are after if we succeed — focus on the lead metrics that help you know you are progressing to the big lag metrics.

The businesses that win in AI aren’t the ones that wait until the path is clear. They’re the ones that experiment, learn, and stack their ROI strategically.

The best time to start? Yesterday. 
The second-best time? Right now.

Need help getting started? Let’s chat.