The Ultimate AI Playbook: From Measuring Adoption to Delivering Impact
When the American Management Association asked if I'd pull together a few insights from my work with organizations for their latest installment of AMA Quarterly, I jumped at the opportunity to pull a few ideas together in a fresh package. Here's the article they published in the latest issue.
***
In five years, no one will care how many people logged into ChatGPT. They'll care about who used it to transform their work. The organizations that understand the difference between more use and better use are quietly outperforming their competitors-while everyone else celebrates meaningless "adoption" metrics.
"For the first time in my 40-year career, I’ve never seen anything like GenAl in our business: were seeing 100% adoption of this new technology," declared the global head of Al at one of the world's largest professional services firms during a recent fireside chat.
Everyone in the audience nodded appreciatively, marveling at this achievement.
Except I knew better.
Just weeks earlier, I had facilitated several workshops with hundreds of people up and down the ranks of that very organization. The on-the-ground reality? Most folks weren't even minimally competent with the technology they were supposedly "adopting."
Behold the great Al measurement delusion: Organizations are tracking usage metrics while missing the only thing that matters: impact.
This isn't a technology adoption curve. It's a human transformation curve. And most organizations are measuring the wrong things. They might be tracking "100% adoption," but they're creating 0% value.
THE SPARK THAT ILLUSTRATES TRUE AI VALUE
Meet "Adam." He's never written a line of code. At the time we first spoke, he didn't even have a LinkedIn account. But in just 45 minutes, he built an Al-powered tool that's on track to save the National Park Service thousands of days of work every year.
How? By solving one of the most mundane yet time-consuming challenges every public facility faces: how to properly request new carpet tiles.
Don't let the carpet tiles fool you. Behind every basic maintenance request in our National Park Service lies a labyrinth of federal specifications, OSHA requirements, ANSI standards, and building codes. Facility managers such as Adam often spend two to three days assembling paperwork for even routine repairs. And across 433 park sites spanning 85 million acres, those hours add up fast.
When the National Park Service invited me to lead a modest upskilling initiative on generative Al, Adam spoke up during our second session: "It stinks that I have to fill out so much paperwork for basic funding requests."
In just 45 minutes, Adam built his tool. It asked him a series of questions about the scope of work, OSHA requirements, ANSI standards, and other details-and then generated a polished, complete funding request document.
Instead of two or three days, Adam's next funding request was done in two hours.
A few weeks later, another facility manager spoke up: "Wait, Adam, you made that tool? Someone sent me the link last week. I had blocked out Monday through Wednesday to process a door and window replacement request. Using your tool, I was done before lunch on Monday."
Adam hadn't even realized people were sharing his tool.
I did a quick, back-of-the-envelope calculation and said: "If this tool saves just one or two days per request across the parks in the system, that's over 7,000 days of labor saved annually." Someone corrected me: "Actually, it's closer to 14,000 days-you're being too conservative."
That's the equivalent of 20 years of labor saved every year.
And it started with someone who, until recently, didn't even have a LinkedIn account.
Takeaway: Minimal overhead can yield massive returns when Al addresses a real pain point. The question isn't "How many people are using Al?" It's "How much real time and money are we saving?"
RETHINKING WHAT WE MEASURE: A 2×2 FOR MEANINGFUL Al
Here's the framework I use when evaluating Al adoption:
On the X-axis, we have how you're using Al:
Poorly: Basic prompts, accepting first answers, minimal iteration
Well: Advanced techniques, thoughtful guidance, strategic iteration
On the Y-axis, we have where you're using Al:
Trivial: Low-value, sporadic use cases with minimal impact
Valuable: Core workflows, strategic processes, leverage points
Most organizations are stuck in the bottom-left quadrant: using Al poorly for trivial tasks.
The goal isn't to get everyone logging in daily. It's to move everyone toward the top-right quadrant: using Al strategically for valuable workflows.
If your organization is serious about Al impact, stop counting logins and start measuring:
Workflow augmentation: How many critical workflows have you improved (not just made faster)?
Solution virality: Is that great tool you built being used by others? That's where you start to scale impact. Remember the National Park Service facility manager who created a tool that saved thousands of days across the park system.
Interaction depth: Are you having single "oracle-style" exchanges effectively single-request Google searches with Al), or robust back-and-forth collaborations? The key to outperforming is to treat Al like a colleague, not an oracle.
Calendar coverage: What percentage of your regular responsibilities are Al-augmented? As Brice Challamel, head of Al products and innovation at Moderna, told me, "I can't imagine doing any part of my job without incorporating layers of Al into it. It would be so lazy, so stupid, so reckless."
Beyond soft benefits like time saved, these metrics often translate directly to reduced operational costs and faster go-to-market timelines-turning Al from a novelty into a bottom-line booster.
THE LEADERSHIP BLUEPRINT: MOVING FROM ADOPTION TO IMPACT
Step 1: Show, don't tell
Last week a CEO asked me the same question I hear from every executive: "How do I get my team to use Al?" My answer shocked him.
"I really want to show people how to use Al to think strategically," he said.
"That's your burning question?" I asked.
He nodded. "Yes."
"Okay, here's my approach to answering a CEO's most burning question: You might expect me to answer straightaway, but I have a strategic thinking partner I'd like to invite to the conversation, if you don't mind."
Then I opened ChatGPT and shared my screen. As we walked through the conversation, the CEO started getting attached to the tactics being suggested. "Can I just do these things?" he asked eagerly.
I stopped him. "As much as I like some of these ideas, they're not the point. I was trying to demonstrate a new behavior for you, real time. Do you see what I just did? I just placed your question-the most important question I'm going to be asked all day-into Al!"
That's when it clicked for him. "You're teaching me to fish, and not just fishing for me."
"That's what you have to do for your team. You have to share your screen and show people how you're using Al to answer your own questions."
This pattern is clear across successful Al-driven organizations.
Diarra Bousso, founder of fashion company Diarrablu, refuses to "level up privately while expecting more from her team." Instead, she creates detailed Loom videos showing her team exactly how she uses Al to enhance her work.
Brad Anderson, President of Product, Engineering, UX, and Ecosystem at Qualtrics, takes the same approach: "At least every week, maybe several times in the week, we'll have a few minutes as people are joining a meeting, or maybe we're wrapping up a little bit early. I'll be talking to the team. I'll say, 'Hey, let me show you what I've been working on in my notebook. And I'll pull up a shared notebook, show how we've imported data into it, start asking questions of it, and actually just show people how to do it."
His conclusion? "The only way that I've ever known to drive behavior change is you have to show people."
Step 2: Allocate dedicated innovation capacity
When the National Park Service invited me to lead that upskilling initiative, I taught the group how to create simple, personalized GPT tools. But the key to success wasn't just the training- it was creating space for experimentation.
Will Guidara, the restaurateur behind Eleven Madison Park-once named the world's best restaurant-understood this principle. He introduced a role he called the "Dream Weaver." This person's sole responsibility? To help staff bring their innovative ideas to life. Guidara recognized that having an insight isn't the same as implementing an insight. Often, folks on the front lines, with great ideas for improving the customer experience, lacked the bandwidth to realize their dreams.
The Portland Trail Blazers applied this same principle to Al adoption. When faced with the challenge of integrating Al across the organization, they created their own version of the Dream Weaver-a role that David Long, their VP of digital and innovation, dubbed the "great aggregator."
Long's role goes far beyond understanding Al. He became the bridge between technology and practical application, between ideas and implementation. He aggregated insights from across the organization, identified pain points, and facilitated Al experimentation.
The result? Tangible Al-driven solutions such as "Betty Budget," a custom GPT that simplified budget code navigation, and an Al assistant that analyzes customer survey results and flags negative responses for immediate action.
Step 3: Create psychological safety around Al
According to a November 2024 study by Slack, "The Workforce Index," up to 48% of employees are hiding their Al use from managers, with that number jumping to 55% among younger workers. Think about that: Half your workforce is concealing tools that could dramatically improve their work.
Right now, your best people are hiding their Al use. That brilliant analysis Jenny did? She's not sharing her process because she's afraid you'll value it less. That innovative solution Mark developed? He's keeping his Al techniques to himself.
Your organization isn't just losing efficiency-it's losing all the compound learning that comes from shared discovery.
At Asana, they've tackled this head-on. The Asana Al system automatically specifies how much of a report was written with Al assistance-and here's the kicker: Higher Al usage carries more prestige. Lower Al usage? That's almost seen as a missed opportunity to create better work.
The stories we tell about Al in our organizations become self-fulfilling prophecies. When leaders whisper about Al use or treat it as something to be embarrassed about, they create a culture of hesitation and shame. But when leaders openly share their processes, they create a culture of innovation and growth.
It's time to change your leadership soundtrack from "don't worry, I won't use Al" to "let me show you how I'm using every available tool to create exceptional work."
Step 4: Shift the goalposts for measurement
I witnessed this shift in a workshop with a Fortune 500 company. At the end, a senior leader proudly declared, "Don't worry, I won't use Al for the green-lighting process." I had to pause the meeting,
"How would you feel," I asked, "if your doctor said, 'Don't worry, I won't use any technology to help with my diagnosis'?" The room went silent. Then the leader smiled and said, "Let me rephrase that. Of course l'm going to use every tool available to mitigate my own bias and complement my thinking so we can reach the best possible path forward."
This isn't just theory. Guy Kawasaki shared how Al improved his book Think Remarkable: 9 Paths to Transform Your Life and Make a Difference (Wiley, 2024). When Al suggested an example better than anything he'd thought of, he didn't hide it-he celebrated it.
"Of course I included it," he said. "My responsibility to the reader is to deliver the best examples possible."
WHY AI IS ONLY AS GOOD AS THE PEOPLE BEHIND IT
Everyone's asking the wrong questions about Al: "What can it do?" "How should we use it?" "Where will it take us?"
Steve Jobs understood something about technology that matters now more than ever: When Apple's iPod dominated the MP3 player market, Microsoft launched the Zune to compete. On paper, the features were comparable. Yet the Zune failed.
Jobs explained why: "...The people at Microsoft don't really love music or art the way we do. We won because we personally love music... If you don't love something, you're not going to go the extra mile, work the extra weekend, challenge the status quo as much."
That truth applies to Al just as much as it did to portable music players. In an age when large language models can generate "average" knowledge instantly, the only thing that sets you apart is the conviction you bring.
Russ Somers, VP of marketing at Quantified.ai, got tired of hearing about corporate workflows. Instead, he built an Al-powered guitar tutor because he actually cares about guitar. While others asked "What's the best Al use case?" Russ turned to the guitars on his wall for inspiration-and ended up creating something far more personal, and therefore far more powerful.
Nicholas Thorne, co-founder of Audos and partner at Prehype, put it bluntly: If you aren't an authentic participant in the community you hope to serve, you'll get "competed away by the low-cost version of average knowledge." And in an Al world, "average knowledge" is practically free.
CULTIVATE A CULTURE OF EXPERIMENTATION
During a recent Beyond the Prompt podcast interview, Advertising Legend Jenny Nicholson highlighted something crucial: Bringing humanity into Al interactions isn't just nice to have-it's a requirement. She astutely pointed out that the only truly new element in any Al interaction is your experiences, thoughts, and personality.
As Nicholson put it, "Helping people realize that (they have to bring their humanity to the conversation) is a nonnegotiable. What is in the model is what is in the model... The only thing that is quote unquote new is anything that you bring." To foster this experimental mindset, encourage your team to:
Interrupt like a human. Don't be afraid to interrupt Al mid-generation. Just like in human conversation, excitement and spontaneity often lead to better outcomes,
Babble. Don't worry about sounding unintelligent when interacting with Al. You don't need to have all your thoughts perfectly in order. The process of thinking out loud can lead to surprising insights.
Keep the conversation going. Treat Al like a conversation partner. Don't settle for the first response-engage, question, critique, and refine until you get something truly valuable. The back-and-forth is where the magic happens.
QUICK WINS FOR ORGANIZATIONS
Here is a checklist of things organizations can do to achieve "wins" with Al:
Start an "Al Innovation Channel" in Slack/Teams Create a dedicated space where people share successful prompts and lessons learned. Celebrate both wins and instructive failures equally.
Identify three to five repetitive tasks Help teams build their first GPT-based solutions for these pain points, following Adam's example from the National Park Service.
Audit one high-value process per department Where's the biggest ROl potential for a 45-minute Al fix? Look for processes that currently take days but could be compressed into hours.
Create your Al manifesto Document your POV as a leader on Al's place in the company. This isn't a policy document-it's your philosophy on Al and how it will shape the kind of work you do. Make it a living document that evolves with your organization's Al journey.
Put a sticky note on your monitor This note-reading "Have you tried Al?"-isn't just for you. It's a visible signal, to everyone who comes to your desk, about what matters to you.
Run an "Al recess" Schedule a monthly hour where teams log in and experiment together, guided by a few curated "prompt challenges." Make it fun, low-pressure, and focused on discovery rather than immediate business outcomes.
Identify your dream weaver Who in your organization could serve as the bridge between ideas and implementation? This person should have the authority to allocate resources and the skills to facilitate experimentation.
Track actual cost savings or new revenue Gather real data on how much time or budget each Al pilot saves-or how it opens up new market opportunities. Use these numbers to validate your next Al investments.
LOOKING AHEAD: THE COMPETITIVE EDGE OF QUALITY AI USE
There will be two kinds of people: those who've used Al to amplify what they actually love, and those who've used Al to paper over the fact that they don't. There will also be two kinds of organizations: those that create capacity for Al innovation, and those that don't.
Al used well reshapes our capacity for innovation. Al used blindly is just a vanity metric. Your job isn't to protect your team from Al. Your job is to protect them from falling behind because they thought they needed to hide their Al use.
The real metric of success is not whether your entire company "touches Al," but whether Al meaningfully elevates the company's impact. And that starts with you, today, identifying a single, critical workflow and piloting a deeper Al approach-and then sharing those results across your organization. Because when algorithms generate anything "average" with ease, only above-average passion—combined with human disagreement—will create real value.
Don't ask: "What can Al do?"
Ask: "What do I love enough to push Al beyond 'average'?"
***
This article originally appeared in AMA Quarterly
Join over 24,147 creators & leaders who read Methods of the Masters each week