AI adoption for executives is a structural problem. I see it in every company that treats AI like a training event. Gregory Shepard, CEO of Startup Science, required every team member to spend three weeks researching how AI could improve their specific role. What he found: a job previously done by six people in marketing could be done by one. Executives generating results like that built a system first.
Quick Answer
AI adoption for executives starts with organizational structure, not software. Before asking anyone on your team to change how they work, assign a named AI owner with real authority, block protected time for experimentation, and appoint departmental champions who understand their workflows from the inside. Glenn Gow coaches CEOs to treat leading AI adoption across their team like any other company-wide initiative: it requires a dedicated owner, a defined process, and measurable outcomes tied to business results – not activity metrics.
Build the Container Before You Ask for Change
AJ Bruno, Co-Founder and CEO of QuotaPath, drew the line his team could not walk back from: “Quotapath will be an AI first company at the end of the year. Like, full stop. If you don’t believe that to be true, then it’s okay, raise your hand. Like this might not be a good fit for you because everything you do on a daily basis… has to involve AI by the end of this year.” He did not stop at the declaration. He appointed Brandon Smith, his head of revenue operations, as head of AI ops and attached a specific efficiency target to the role.
Steve Harmon, CEO of Spartan Logistics, used a different structural forcing mechanism. He banned his leadership team from using Google for 90 days and required AI chatbots only. “I challenged my leadership team to say, Hey, there’s some really good things going on here. I want all of you to not Google anything for the next quarter. I want you to only use the chatbot of your choice. And they’re all looking at me like, whatever Steve, it can’t be that different. And at the end, we came back 90 days later… and they’re all like, you were right. This is better,” he said.
Both approaches share the same backbone, and it is what I tell CEOs to pay attention to: a named leader made a structural decision and held the organization accountable. Neither waited for organic adoption to emerge.
Mandate vs. Empathy: Two Paths, Two Outcomes
Two camps of executives have emerged in how they lead AI adoption across their teams. One issues hard mandates and forces immersion. The other invests in psychological safety and patient education. I work with both. Here is what each approach produces for the CEO:
| Approach | Who Used It | What It Produces |
| Hard mandate with forced immersion | AJ Bruno’s year-end AI-first ultimatum at QuotaPath; Gregory Shepard’s three-week functional research mandate at Startup Science; Steve Harmon’s 90-day no-Google rule at Spartan Logistics | Fast cultural shift, short-term friction, rapid proof of capability for skeptics |
| Patient investment in education and psychological safety | Dario Markovic’s extended AI training budget at Eric Javits; Doug Merritt’s peer-to-peer learning program not tied to performance reviews at Aviatrix | Slower adoption curve, lower fear-driven resistance, stronger long-term habit formation |
Dario Markovic, CEO of Eric Javits, ran the patience approach: “We spent a lot of time and budget internally for getting everyone up to speed with technology and AI. If they have certain fear or just not sure about using it or not know how to use it, we’re very patient of getting them every tool and every education they need to get to use it… once you have someone that can lead you and show you, this is much easier than what you did before… this could open your mind even to more things.”
Doug Merritt, CEO and President of Aviatrix, named the real obstacle: “The challenge I don’t think is the technology. It rarely is. It’s human change within an organization… AI is scary for all of us. Is my job going to be replaced? What is my role going forward if AI keeps advancing? And the human element, I think, is the more important… until we hit super intelligence or some future inflection point, it is a tool.”
The mandate breaks inertia fast. The patient approach sustains momentum longer. I coach CEOs to use one approach to shift the culture and the other to hold it.
AI Adoption for Executives Starts With Naming an Owner
The first structural move I coach every CEO to make is naming a person, assigning the role officially, and attaching a real number to it. AJ Bruno, Co-Founder and CEO of QuotaPath, did exactly that. “Brandon Smith, our head of revenue operations, I made our head of AI ops… Everything from our quota to OTE is currently one and three and a half X. I want a one and 5x by the end of the year. And the only way we can do that is through automation and sales efficiency,” he said.
An AI initiative without a named owner is a directive. A named owner without a specific target is a title. I coach every CEO I work with to start here: name the person, name the number, connect the two. How executives accelerate AI adoption past the announcement stage comes down to that combination.
Block Protected Time or Adoption Stays Optional
Hilary Dubin, Co-CEO of Jones, built what most executives only talk about: mandatory, protected time for AI experimentation. “What we have recently implemented and I’ve kind of been the driver behind this at Jones is a Friday AI innovation time for the whole team. So every Friday, the whole afternoon is dedicated to AI exploration, sandbox, like don’t worry about your usual tasks… I think having that system and that accountability of the demos where it’s like, you come and you show off something cool, makes it more fun and actually gets you to take the time to explore,” she said.
I tell CEOs: if you don’t build in the accountability loop, you don’t have a time block – you have an open calendar item that will disappear by week three.
Chris Rolls, CEO of TTC Global, named the same bottleneck in his AI rollout strategy: “The challenge that many organizations have that we of course go through as well is making sure we create time for innovation and making sure we create time for everyone within the organization to innovate.” Unblocked time on the calendar is a structural requirement. Any AI rollout strategy that skips it will stall within a quarter.
Departmental Champions Scale What Central Teams Cannot
Julie Szudarek, CEO of Self Financial, built the model I point most CEOs toward first: a centralized AI ops function and center of excellence paired with named departmental champions. “We have a centralized function that does our sort of AI ops and AI center of excellence… We also have champions from each department. Those champions have put forth, like these are three biggest opportunities we think from an AI perspective. And then our ops group then goes and helps the champions kind of figure out how they could go about implementing those top three opportunities,” she said.
The model solves the gap that kills most AI rollout strategies. The central AI ops team does not know the workflows. The departmental teams do not know how to implement. Departmental champions bridge both: they have the workflow knowledge the central team lacks, and the center of excellence carries the implementation muscle the champion lacks. I see this structure produce faster adoption with less central-team burnout than any other model I have worked with.
Build Cross-Functional Education Into the Calendar
Fran Brzyski, CEO of Hark, built peer-led AI upskilling without a formal training budget. His engineering team leads it. “We have different Slack channels for AI education that aren’t just sales related, aren’t just marketing related. It’s for the whole company. And a lot of that comes from our engineering team to teach and explain things to us… We do a lunch and learn. We started where somebody will pick a topic where AI is unlocked a ton of efficiency for them or creativity for them. And they’ll explain to the rest of the company,” he said.
Engineering as the teaching team is counterintuitive for most CEOs I coach. Most organizations push AI information down from leadership. Fran inverted it, and cross-functional education built that way compounds on its own without ongoing budget.
Fix the Foundation Before You Add AI
Steven Monterroso, CEO of ShareVault, named the infrastructure condition that most executives skip: “You have to first take your inner operations and make sure you have good data coming in so that good data could come out, right?… If you don’t have that, then all you’re going to get is a mess. You’re only going to 3x the mess you’ve already had, right?”
Talbot Gee, CEO of Heating, Air-conditioning & Refrigeration Distributors International, describes what happened when the organization’s custom AI chatbot for complex regulatory documents hit the limits of scale: “It worked kind of until it didn’t and when it didn’t it’s because it got literally too complicated… With that thing, the most expensive issue would have been success. Like the more users you have, the more expensive the tool gets because the hosting fees go up, right? Who would have thought? We had no idea about the economics of some of these AI tools.”
Before any AI rollout strategy, I tell CEOs to audit their internal data quality and understand the infrastructure economics of whatever they plan to deploy. AI does not repair bad inputs. It scales them.
Measure Outcomes, Manage the Culture Shift, and Plan for Capacity
AI upskilling that produces activity metrics without shifting outcomes is failing, and I see it in most companies that have been running AI tools for six months or longer. Ken Jisser, Founder and CEO of The Pipeline Group, tracks it with a single question: “We use AI to make humans more efficient and we measure the productivity of AI by the outcomes it produces, not the activity increases… if it’s not increasing outcomes, it’s not being used effectively. And most organizations aren’t using it effectively.”
Steven Collens, CEO of Matter, named the structural trap that produces bad metrics: “We could use AI to do this or that or this or that or this. The possibilities are almost, I mean, they’re unlimited almost at this point… we can’t have, in a big organization, we can’t have a thousand pilots running and expect to get a lot of value out of that. It’s like, what are the things that are really going to drive value and how do we get people focused on those?” I tell CEOs to pick three bets and measure them against outcome baselines, not activity counts.
There is also a cultural risk that spreads quietly once adoption picks up. Chris Crowe, Founder and CEO of CMBYND, observed this pattern inside his own teams: “I think it was a stat we had run, which was 75% of people just take what’s produced from AI and agree with it. Say, that’s right. And what that’s doing is stopping the thinking in your organization, slowing it down and shifting your culture of your organization inadvertently.” I coach CEOs to treat this as a leadership accountability: an AI-first company that does not train employees to question AI output is building passivity into its culture.
And when adoption succeeds at scale, a new question arrives. Avanish Sahai, Board Member at HubSpot, named it directly: “I would argue the biggest change or the biggest challenge that AI projects are having or going to have… I think the biggest issue is the people. It’s the change management… What do I do if I have 20, 30, 40% extra capacity, am I going to repurpose them? Am I going to retrain them? Am I going to lay them off?” I tell every CEO I coach to build the answer to that question before the capacity arrives. Executives who wait until AI has freed up 30% of their workforce spend the next quarter reacting instead of leading.
The measure of a successful AI rollout is not how many tools your team uses. It is whether those tools are producing better outcomes per person than you were getting before.
FAQ
What should executives do first to lead AI adoption across their team?
Glenn Gow recommends three structural decisions before rolling out any tool: name a dedicated AI owner with a specific efficiency target, surface two or three high-value use cases per department through champion conversations, and block recurring time on the calendar for experimentation. Announcing a tool without those three elements in place produces short adoption curves and fast regression to old habits.
Who should own AI adoption in my company?
Glenn Gow recommends looking for the AI owner in a function that already tracks operational efficiency – revenue operations, business operations, or a chief of staff with cross-functional reach. AJ Bruno of QuotaPath made his head of revenue operations the AI ops lead because that person already owned the metrics the efficiency targets needed to move. The role must carry a specific outcome tied to a number the owner did not have before, and the authority to hold departments accountable to it.
What is an AI center of excellence and does my company need one?
An AI center of excellence is a centralized function that handles implementation support, vendor evaluation, and ROI measurement across departments. Glenn Gow’s view: most companies under 500 people do not need a standalone center of excellence. They need one person with the authority to prioritize AI investments and a documented decision process for vetting new tools. Julie Szudarek built the full hub-and-spoke model at Self Financial because she had departmental scale that required it. Smaller teams get the same value from a simpler owner-plus-champions structure.
How do I get employees to stop reverting to old habits after an AI rollout?
Reversion happens when AI use is optional and goes unmeasured. Two mechanisms prevent it: accountability loops (demos, shared metrics, weekly check-ins from the AI owner) and structured time that makes AI the default during work hours rather than an add-on. Hilary Dubin at Jones made Friday afternoons mandatory AI exploration time with a show-and-tell demo attached, because without the demo, the time disappears into routine tasks. Structure creates the habit. Measurement proves it is working.
What is an AI rollout strategy for a mid-size company?
Glenn Gow’s recommended sequence: audit data quality first, appoint an AI owner, identify the three highest-value use cases through departmental champion conversations, run a time-boxed pilot on each, measure outcomes against a pre-rollout baseline, and set a decision checkpoint at 90 days. Do not run more than three parallel pilots at once. Steven Collens of Matter found that organizations running unlimited pilots simultaneously produce almost no concentrated value from any of them.
How do I measure whether AI adoption is actually working?
Measure adoption by outcomes, not activity. If AI tools are producing more emails sent or more calls logged without a corresponding lift in revenue, closed pipeline, or cost per outcome, the adoption is misaligned. Glenn Gow recommends setting a 90-day outcome baseline before any tool goes live so you have a real comparison point. Ken Jisser of The Pipeline Group built his entire AI measurement framework around one question: is this producing better outcomes per person, or just more activity?
What happens to my team when AI creates 20-30% extra capacity?
This is the question most executives fail to prepare for. Avanish Sahai of HubSpot identifies the hardest challenge in AI adoption as the HR decision that comes after efficiency gains arrive: whether to repurpose, retrain, or reduce headcount. Glenn Gow’s recommendation: define the redeployment policy before the AI rollout begins and communicate it to leadership explicitly. Tie efficiency gains to growth targets rather than cost reduction. When employees know what happens to freed capacity, fear of adoption drops.
How do departmental AI champions work in practice?
Departmental AI champions are employees selected from within each function to identify the highest-value AI opportunities for their team, pilot solutions with central support, and share results across the organization. The champion role works because it pairs insider workflow knowledge with organizational permission to experiment. At Self Financial, Julie Szudarek’s champions each submitted their top three AI priorities, which the central AI ops team then helped them scope and implement. Champions are not AI experts. They are workflow experts who have been given permission to think about AI.
How do executives overcome employee resistance to AI?
Glenn Gow sees two distinct drivers of resistance: fear of job replacement and habit inertia. Fear responds to transparency – a clear statement of what will and will not change, and a defined policy on what happens to freed capacity. Habit inertia responds to forced exposure. Steve Harmon of Spartan Logistics broke his leadership team’s habit inertia with a 90-day no-Google rule that forced them to prove the tools’ value to themselves. Both forms of resistance require a structural response, not a persuasion campaign.
What is the biggest mistake CEOs make when rolling out AI company-wide?
The biggest mistake is treating AI adoption as a tool rollout rather than an organizational change initiative. Tool rollouts live in IT. Organizational change lives with the CEO. Glenn Gow coaches executives to own the mandate personally, assign a named owner, and measure it like any other revenue initiative. The second most common mistake: deploying AI on top of bad data and operations. Steven Monterroso of ShareVault puts the outcome plainly – bad internal data does not get fixed by AI, it gets tripled.
What does it mean to be an AI-first company?
An AI-first company is one where every employee’s daily workflow involves AI by design, not by choice. Glenn Gow coaches CEOs to treat the AI-first declaration as a structural commitment, not a culture statement – which means naming an owner, blocking time for experimentation, and measuring outcomes before the declaration is made public. AJ Bruno set the standard at QuotaPath by declaring the company would be AI-first by year-end and immediately appointing a named AI ops lead with a specific efficiency target attached. The declaration without the structure is a slogan. The structure is what makes it a company posture.
Should I mandate AI adoption or let it happen organically?
Glenn Gow’s answer: the choice depends on what you are trying to break. A hard mandate – banning old tools, issuing a year-end AI-first declaration, requiring employees to research AI applications in their specific role – breaks habit inertia fast and produces rapid proof of capability. Organic adoption preserves psychological safety but rarely generates the pace executives need. The executives Glenn Gow coaches who got both results used a top-down mandate to create the initial shift, then invested in patient education and protected time to sustain it. Gregory Shepard’s three-week functional research mandate at Startup Science produced a finding that a marketing job previously requiring six people could be done by one. Organic adoption would not have surfaced that finding on the same timeline.
What You Do Next
If you are ready to stop announcing AI and start building the structure that makes it stick – the ownership model, the time blocks, the champion framework – Glenn Gow works with CEOs to design exactly that. Schedule a conversation with Glenn and walk away with a clear picture of what your adoption structure must look like before you ask anyone on your team to change how they work.
