A technology keynote speaker is worth the fee in 2026 only if the people in your room use the technology you've already paid for more often 90 days after the event, because that's where most technology budgets are quietly losing money right now. Companies buy the platform, run the training, send the memo, and then watch usage stall while the invoices keep arriving. Dr. Noah St. John isn't a technologist and isn't a futurist. He's spent 29 years coaching senior operators on why capable people resist a change they know is good for them, and you can request his date and fee for your event before you read another word.
A technology keynote speaker in 2026 has to change what people do with technology after the event, in terms a CIO or a CFO can measure, or the hour becomes one more line item that didn't pay back.
That's a different brief from the one most planners inherited.
For years, the job of a technology keynote was to show the room the future. Here's what's coming. Here's how fast it's moving. Here's why you should be excited, or worried, or both. That brief made sense when the technology was still on the way.
For most companies, it isn't on the way anymore. It's here. It's licensed. It's sitting in every employee's browser. The question your executives are asking has changed from "what's coming?" to "why aren't we getting the return on what we already bought?"
Those are two different questions, and they need two different kinds of speaker.
Search for a technology keynote speaker and you'll mostly find the first kind: futurists, trend forecasters, AI researchers, and former tech executives with a view of the next decade. Many of them are excellent at what they do. If your room genuinely needs a map of what's coming, one of them is the right booking.
But if your room already has the tools, and the tools aren't being used, a better map of the future won't fix it. The people in your audience don't lack information about AI. They're swimming in it. What they lack is the willingness, in the moment it matters, to change how they work.
This page is written for the planner with that second problem. It starts where your CFO starts, with the money. The mechanism comes after, because that's the order a buyer actually cares about it in.
Companies are losing money right now on technology they've already bought, because the licenses, the integration work, and the training are paid for while the behavior change that produces the return never arrives.
Start with what chief executives say about their own results.
IBM's Institute for Business Value surveyed 2,000 CEOs across 33 countries and 24 industries between February and April 2025. The CEOs reported that only 25% of AI initiatives had delivered the expected return on investment over the past few years. Only 16% had scaled enterprisewide.
Read that the other way round. By the CEOs' own account, three out of four AI initiatives hadn't delivered the return they were bought to produce.
Then the pilots that never become anything. Gartner predicted in July 2024 that at least 30% of generative AI projects would be abandoned after proof of concept by the end of 2025, citing poor data quality, inadequate risk controls, escalating costs, or unclear business value. Some of those reasons are technical. "Unclear business value" usually isn't. It's what a project looks like when the tool works in the demo and nobody changes how they work afterward.
Then the gap between the winners and everybody else. BCG's 2025 research, summarized in its 2026 report AI Transformation Is a Workforce Transformation, found only about 5% of organizations had achieved substantial financial gains from AI. Those companies showed three-year total shareholder returns roughly four times higher than AI laggards.
None of those numbers describe a shortage of technology. The companies in these studies bought the tools. The tools, for the most part, do what the vendor said they'd do.
What the numbers describe is a shortage of adoption. The spend happened. The change in behavior didn't. And every month the gap stays open, you're paying full price for a capability your people are using at a fraction of its value.
If that's the gap on your leadership team's mind, ask for a quote tied to your event and a short call to test whether it applies.
Most of the value from AI depends on people and process rather than the technology itself, which is why a technology budget that stops at the software leaves most of the return uncollected.
BCG has a simple way of saying this. In its case work across hundreds of companies, it describes a 10-20-70 rule: about 10% of the value from AI comes from the algorithms, another 20% comes from the technology needed to implement them, and the remaining 70% comes from rethinking the people component.
Now compare that with where most rollout budgets go.
In a typical rollout, the license is the biggest line. Integration and data work come next. Training gets a budget, usually a small one, and it's almost always about features: here's the button, here's the prompt, here's the workflow. Almost nothing goes to the question of why a trained, capable person still doesn't use the tool when a deadline hits.
That's the 70% sitting unaddressed.
BCG's own data shows what happens at the manager level when it's addressed. In the same 2026 report, 88% of managers at companies BCG calls "future-built" actively use AI in decision-making, compared with 25% of managers at laggard companies. Future-built firms were also four times more likely to have structured AI learning programs with protected learning time.
Those are behaviors, not purchases. Protected learning time is a decision a leader makes about the calendar. A manager using AI in a decision is a person choosing to do something new in front of their team, with the risk of looking unsure while they figure it out.
That's the part of the problem a keynote can actually reach. Not the algorithm. Not the data pipeline. The human being standing between the tool and the result.
The adoption gap is the distance between what your technology can do and what your people actually do with it, and in most companies it's the single biggest reason a technology investment underperforms.
BCG's AI at Work 2025 survey put a number on one version of it. Among leaders and managers, more than three-quarters said they used generative AI several times a week. Among frontline employees, regular use had stalled at 51%, a figure BCG described as a "silicon ceiling."
Stalled is the right word. The tools didn't get worse. Access didn't shrink. Half the workforce simply didn't make the tool part of how they work.
You've probably seen the same pattern inside your own company, even without a survey.
A new platform launches with a town hall and a training week. Usage spikes. Then, over the following month, it slides back toward the people who were already enthusiastic. Everyone else goes back to the spreadsheet, the email chain, the way they did it last year.
Ask them why and you'll get reasonable-sounding answers. "I'm too busy to learn it right now." "The old way is faster for this." "I'll get to it after the quarter closes." "I don't trust what it produces yet."
Some of those answers are true. Many of them are cover.
Underneath a lot of "too busy" is something people don't say out loud in a meeting: the new tool makes them feel slow, exposed, or replaceable, and the old way makes them feel competent. Given a choice between feeling competent and feeling exposed, most people choose competent every time, even when the exposed path is better for them and the company.
That's not a technology problem. It's a human one, and it's the gap Dr. Noah's work is built for.
Many employees who use AI at work hide it, because they worry that being seen using it makes them look replaceable, and that secrecy keeps good practice from spreading across the company.
Microsoft and LinkedIn's 2024 Work Trend Index, based on a survey of 31,000 people across 31 countries, found three in four knowledge workers were using AI at work. It also found 78% of AI users were bringing their own AI tools to work.
Then the finding that should worry every executive who signed a technology contract. Among people who use AI at work, 52% said they were reluctant to admit to using it for their most important tasks. And 53% worried that using it on important work tasks made them look replaceable.
Sit with what that means inside your company.
Your most effective AI users may be hiding exactly what's working. The analyst who's figured out how to cut a three-day report to three hours isn't teaching the team. She's quietly doing it and hoping nobody asks why she's suddenly so fast. The account manager who drafts proposals in half the time doesn't mention it in the pipeline review, because saying so might invite the question of why the company needs as many account managers.
That's the Caveman at work. The fear isn't irrational from the inside. It's a very old survival calculation: don't let the tribe see anything that might make you expendable.
And look at the other side of the same survey. Microsoft found 60% of leaders worried their organization's leadership lacked a plan and vision to implement AI. So leaders are unsure, and employees are hiding. Both sides are holding back the same information from each other, for the same underlying reason.
No software update fixes that. The fix is making it safe, and normal, to be seen learning in public. That's a leadership behavior, and leadership behaviors are what a keynote aimed at the right room can change.
Training alone doesn't close the adoption gap because training teaches people how a tool works, while the gap is about whether they'll change their behavior under pressure once the training is over.
Training still matters, and the research says so. Microsoft's 2024 Work Trend Index found only 39% of global AI users had received AI training from their employer. BCG's AI at Work 2025 found only 36% of employees felt their training was enough, and 18% of regular AI users said they'd received no training at all. BCG also found 79% of respondents with more than five hours of training were regular AI users, compared with 67% of those with less.
So if your people haven't been trained, train them. That comes first, and nothing on this page replaces it.
But notice what the BCG numbers also show. Even with more than five hours of training, about one in five people still weren't regular users. They knew how. They didn't.
That's the group this page is about.
Here's why training can't reach them. A training session happens in a safe room with no deadline, no audience, and no consequence for getting it wrong. The real decision to use the tool happens somewhere else: at 4:40 on a Thursday, with a deliverable due, a manager waiting, and a familiar old method that's guaranteed to work.
In that moment, the brain isn't asking "do I know how to use this?" It's asking "what happens to me if I try this and it goes badly?" Training answers the first question. It rarely touches the second.
Change management research makes the same point from another angle. Prosci's Best Practices in Change Management research, drawn from more than 2,600 change practitioners, found 88% of projects with excellent change management met or exceeded their objectives, compared with 13% of those with poor change management. Prosci describes that as roughly seven times more likely to meet objectives.
Seven times. Same technology, same project plan, different handling of the people going through the change. If you want to talk through where your own rollout sits on that spectrum, start a conversation about your date.
Book a futurist or an AI expert when your audience needs to understand what's coming; book an adoption-focused speaker like Dr. Noah St. John when your audience already has the technology and isn't using it.
Being fair about this matters, because the wrong booking costs you the slot either way.
There's a real need for the "what's coming" speaker. When a leadership team is deciding where to place bets over the next five years, they need someone who studies the signals for a living. That's a genuine discipline, and some people are very good at it.
Amy Webb, for example, is the founder and CEO of Future Today Strategy Group and teaches strategic foresight as an adjunct professor at NYU's Stern School of Business. Kevin Kelly was the founding executive editor of Wired, and his 2016 book The Inevitable forecasts the technological forces he expects to shape the following decades. There are also AI researchers and technical leaders who can explain what the models can and can't do with an authority no generalist can match.
If your brief is "help our leaders understand where technology is headed," one of those people is the right call. Dr. Noah isn't. He doesn't forecast technology, and he won't pretend to.
Here's the honest test for which gap you have.
You have a "what's coming" gap if your leaders genuinely don't know which technologies matter for your industry, haven't decided what to invest in, and need a framework for making those bets.
You have an adoption gap if the decisions are made, the tools are bought, the training happened, and usage is still flat. If your CIO can show you the license count and the login numbers, and the second one is disappointing, you don't need a better forecast. You need people to act on the present.
Many companies have both. If you do, the order matters. A futurist in the morning followed by an adoption session in the afternoon can work well, because the first raises the stakes and the second addresses why people will still hesitate. Two futurists in a row, for a room that already has the tools, usually produces a great day and no change.
Capable people resist new technology because the brain treats looking incompetent, losing status, and being replaced as threats to survival, and that reflex, built over roughly 200,000 years of human history, fires faster than any rational argument for the tool.
Dr. Noah calls this the Caveman in your brain. It's the 200,000-year-old wiring that kept our ancestors alive by steering them away from anything that might get them pushed out of the group. For almost all of human history, being cast out was a survival problem. So the brain learned to treat social threat as real danger, and to move away from it fast.
That isn't just a metaphor. In a 2003 study published in Science, Naomi Eisenberger, Matthew Lieberman, and Kipling Williams scanned people's brains while they were excluded from a simple virtual ball-tossing game. Activity rose in the anterior cingulate cortex, a region previously linked to the distress of physical pain, and that activity correlated with how distressed people said they felt. Social exclusion registered in the brain in a way that resembled physical pain.
Now put that brain in front of a new AI tool.
Trying the tool in front of colleagues means risking looking slow or clumsy. Admitting you don't understand it means risking status. Using it well might mean the company decides it needs fewer people like you. Every one of those is a social threat, and the Caveman reads social threat as danger.
Then add the math of loss. In their 1992 paper on cumulative prospect theory, Amos Tversky and Daniel Kahneman estimated that losses weigh roughly 2.25 times as heavily as equivalent gains in people's decisions. The competence someone already has feels like something they could lose. The productivity the tool might bring feels like a gain they might get. The possible loss wins, at more than two to one.
That's the whole story of a stalled rollout in two research findings. The new tool asks people to give up a feeling of competence they already own in exchange for a benefit they haven't experienced yet. The Caveman votes no, quietly, and then the rational mind supplies a reasonable-sounding excuse.
This is why the Caveman vs AI is the real contest in your company. Not your people versus the machines. Your people's 200,000-year-old threat detection versus a tool that arrived last quarter. The technology moves at the speed of software. The Caveman moves at the speed of evolution. Guess which one wins by default.
The Invisible Brake™ is Dr. Noah's name for the subconscious pattern that holds a capable person just below the level of exposure that feels safe, and in a technology rollout it shows up as smart people finding reasons not to use tools they know would help.
After 29 years of coaching senior operators, Dr. Noah kept meeting the same person. Trained. Experienced. Completely clear on what needed doing. And somehow not doing it.
He calls the thing in the way the Invisible Brake™. It isn't laziness, and it isn't a character problem. It's a protective mechanism doing its job with out-of-date information.
In a technology rollout, it looks specific and recognizable.
A senior manager keeps asking for "one more pilot" before committing, because committing means owning the result. A sales leader lets the team keep working the old way because pushing the new process means a tense conversation with a top performer who doesn't want to change. A finance director prints the dashboard and marks it up by hand, because the new tool makes her feel like a beginner in front of the analysts she's supposed to lead. An engineer keeps using AI privately and never shares what works, because sharing invites the question of what his job is for.
Every one of those people is capable. Every one of them could use the tool. The behavior is blocked, not missing, and you don't unblock a behavior by adding another training module on top of it.
That distinction is the whole approach. Most adoption programs assume something's missing and add more: more features, more training, more communication. This approach assumes something's in the way and removes it. When it's removed, the capability that was already there starts showing up on its own.
For a company that's already spent the money, that's good news. You don't need a new platform. You're recovering value from technology you already own, through people you've already trained.
If you want the broader leadership picture behind this, Dr. Noah's article on the AI leadership gap covers how leaders close it in their own behavior first. This page is narrower: what a keynote can do about it in one room, on one day, and how you'll know if it worked.
People experiment with new technology only when they believe making a visible mistake won't be held against them, and research ties that belief directly to how much a team learns.
Amy Edmondson's 1999 study in Administrative Science Quarterly, "Psychological Safety and Learning Behavior in Work Teams," looked at 51 work teams in a manufacturing company. She defined team psychological safety as a shared belief that the team is safe for interpersonal risk-taking. Teams with higher psychological safety showed more learning behavior, and learning behavior was linked to better team performance.
Adopting AI is learning behavior in its purest form.
It's asking a question that shows you don't know something. It's trying a prompt that doesn't work, in a shared document where colleagues can see it. It's admitting that the first draft the tool produced was wrong and you didn't catch it. It's telling your manager, "I tried this and it saved me a day, and I'm not sure what to do with the extra day."
Every one of those carries social exposure. Every one of them is what the Caveman suppresses.
And here's the part that matters most to the executive booking the keynote. Psychological safety isn't created by employees. It's set by how leaders react the first time somebody takes the risk.
If a manager rolls her eyes the first time someone shares a clumsy AI experiment, nobody shares again. If a vice president asks, in a meeting, "so what do we need you for, then?" after someone shows how much time the tool saved, the whole floor hears it by lunch, and every hidden AI user stays hidden.
If the same leaders go first instead, sharing their own clumsy attempts, thanking people for experiments that failed, and making it clear the saved time gets reinvested rather than cut, the hiding stops.
That's why this content lands hardest with leadership in the room. Your employees already know the tools exist. What they're watching is how their leaders react.
Afformations® are question-based self-talk, a technique Dr. Noah coined in 1997, and they give people something short and specific to do in the seconds when the old habit pulls them away from the new tool.
The technique is small. Its research base is why it survives a room full of engineers and analysts who don't like being told what to think.
A 2010 study published in Psychological Science by Ibrahim Senay, Dolores Albarracín, and Kenji Noguchi tested something narrow. Participants were more likely to solve anagrams when they prepared by asking themselves whether they would work on them, rather than declaring that they would. In follow-up experiments, simply writing "Will I" instead of "I will" produced better anagram performance and stronger intentions to exercise. The researchers' explanation: a question invites you to come up with your own reasons, and reasons you generate yourself carry more weight.
That's exactly why this works with technical audiences instead of against them. A skeptical engineer rejects "I embrace change and love new tools" instantly. He knows it's a slogan. The same engineer will engage with "Why am I the person on this team who figures out new tools first?" because the question sends his mind looking for evidence, and he'll find some.
Dr. Noah's name for the technique is Afformations®. An affirmation is a statement you assert. An Afformation® is a question that assumes the thing is already true and sends your mind hunting for why.
In a technology rollout it's this concrete.
A manager about to open the new dashboard instead of asking for the usual spreadsheet asks herself why she's good at learning things in front of her team. Her mind produces real examples, because there are real examples.
An analyst about to share an AI shortcut in the team channel asks why his colleagues are glad when he shows them something useful. The question surfaces the reasons, and the reasons make sharing feel generous instead of risky.
It takes a few seconds, and it happens in the exact window where the old habit would otherwise win. That timing is what makes it a working tool rather than a journaling exercise.
Dr. Noah St. John fits technology audiences because he works on the human side of adoption that technical speakers leave untouched, and because he's plain about what he isn't: a technologist or a futurist.
Start with the honest part. Dr. Noah won't explain how large language models work. He won't forecast which AI vendors will win. He won't give your engineers a better architecture or your data team a better pipeline. If your room needs any of that, book someone who does it for a living.
What he does is the other 70%.
The credentials, stated plainly. Dr. Noah has spent 29 years coaching senior operators, since launching SuccessClinic.com in 1997. He's written 27 books, with publishers including HarperCollins, Hay House, Simon & Schuster, and Mindvalley. His clients have produced more than $3 billion in cumulative results across 150-plus countries. He coined Afformations® in 1997, created the Power Habits® System, and gave the TEDx talk "Done with Head Trash."
The fee, stated just as plainly: Dr. Noah's signature keynotes run $25,000 to $50,000, scaled to your event and audience. You shouldn't have to guess at that number, and you shouldn't have to sit through a sales call to learn it.
Credentials aren't the main reason he works in a technology room, though. Three other things are.
First, he leads with mechanism rather than hype. Technical audiences have sat through years of breathless AI keynotes. They extend credit to a described process, one specific enough to be wrong, and they withdraw it fast from anything that sounds like a pep rally.
Second, he doesn't ask the audience to feel excited about AI. He asks them to notice one specific moment, the hesitation before using a tool they already know would help, and gives them something to do in it. That's a behavioral instruction with a clear trigger, which is the format technical people already trust.
Third, he's explicit about limits. The approach doesn't fix a bad tool, a broken integration, missing training, or a legitimate fear of layoffs that leadership hasn't addressed. It works on the gap between knowing and doing. Saying so up front is what buys credibility with an audience that's heard oversized promises before.
If that matches what your audience needs, send the event details and get a quote.
Dr. Noah's technology keynotes are built from four of his frameworks, adapted to your audience and the specific adoption problem you want to move, rather than delivered from a fixed deck.
Tame the Caveman in Your Brain™ is the foundational talk. It explains the 200,000-year-old threat-detection wiring that treats looking incompetent and being replaced as danger, why it fires faster than deliberate thought, and what to do in the moment it fires. It's the strongest general-session fit for a mixed room of executives, managers, and individual contributors going through a digital transformation together.
Caveman Leadership™ is aimed at the leaders who set the tone for adoption. Same mechanism, applied to the decisions they put off: committing past the pilot, having the direct conversation with a resistant top performer, going first with a new tool in front of the team, and reacting well when somebody shares a failed experiment.
Caveman Selling™ and The Caveman Conversion Code™ fit sales teams at technology companies. Dr. Noah is known as The Caveman Conversion King™, and this is his commercial work: the hesitation that shows up when a seller needs to hold the price, ask for the decision, or tell a prospect the truth about what implementation will take. For a software or services company, that's often the room where an hour moves revenue fastest.
The Power Habits® System is the follow-on structure for companies that want the keynote to become a program rather than a one-time event.
Common settings where this content fits:
Formats run from a 45- to 60-minute general session up to half-day workshops for a leadership group. For companies rolling out a tool across many teams, a live general session for leaders followed by a short, repeatable version managers can run in their own team meetings usually works better than one recorded video sent to everyone.
At a technology company's own customer conference, an adoption-focused keynote is aimed at the thing your renewals depend on: whether customers actually use what they bought from you.
This is a use case many planners at software companies miss.
Your customers have the same adoption gap described on this page, except it's your product sitting unused in their organization. Their champion bought it. Their team was trained. Usage is flat. And at renewal time, flat usage turns into a hard conversation about value, a discount request, or a quiet non-renewal.
Most customer conference agendas answer that with product roadmap sessions and feature training. Both are useful. Neither addresses why a trained customer team keeps working the old way.
A keynote that names the Caveman in the room does something different for your customers' champions. It gives them language to explain internally why adoption stalled that doesn't blame them, doesn't blame the product, and points to something they can actually do about it. A champion who walks out with that language goes home better equipped to drive usage, and usage is what renews.
It also changes the tone of your event. Customers are used to vendors telling them how great the product is. A session that says, honestly, "the tool isn't the hard part, your people are, and here's why" earns a kind of trust that a feature demo can't.
If that's the event you're planning, reach out about your customer conference date.
Demand one named adoption behavior from any technology keynote speaker, measured 30 to 90 days after the event with data you already collect, and walk away from any speaker who can't give you one.
Most planners judge a keynote by the applause and the evaluation forms. Both measure the hour. Neither measures the result. Here's what to ask for instead. It applies to every speaker you're considering, including Dr. Noah.
A named behavior. Not "embrace innovation" or "a culture of curiosity." A specific action a specific group of people will take more often. For example: managers use the new analytics tool in their weekly review instead of asking for a spreadsheet. Sellers log activity in the new platform the same day. Team leads share one AI workflow a month in their team channel.
A measure you already collect. Weekly active users by team. Feature usage in the platform's own admin console. Share of reports generated through the new tool versus the old one. Number of workflows shared internally. If the speaker's result can't be read from data you already track, it'll never be read at all.
A time frame. Thirty days for early signals, 90 days for a real read. Anything longer and too many other things change to know what caused what.
An honest limit. Ask what the talk won't do. A speaker with no limit to name is either inexperienced or overselling. Both cost you the room's trust in the first ten minutes, and a technical room is less forgiving than most.
Clarity about which gap they fit. Ask directly: are you a "what's coming" speaker or a "why aren't we using it" speaker? A good speaker answers without hesitating. One who claims to be both at full depth is probably neither.
Preparation. Ask what they need from you in advance. A speaker who needs nothing is delivering a canned hour. Dr. Noah's team asks for a short call with the person who owns the adoption outcome, plus whatever you're willing to share about where usage is stuck.
A plan for after. Ask what reinforces the talk in the following month. A speaker with no view on that hasn't thought about your result.
That list is the standard. You can use it to request a fee quote for your event and hold any speaker, including this one, to it.
Put an adoption-focused technology keynote before the working sessions and product training, not as the feel-good closer, so the room has shared language for the resistance that comes up in every breakout.
The most common mistake isn't picking the wrong speaker. It's putting the right speaker in the wrong slot.
Adoption content gets booked as the closing session of a two-day event, after the roadmap, the demos, and the technical tracks, when everyone's thinking about their flight. It plays well. It changes nothing, because there's no working session behind it.
The same hour placed at the start does something different. It gives the room a name for what's about to come up in every session: the pilot nobody wants to end, the team that won't switch, the tool that works in the demo and sits unused afterward. Once the mechanism has a name, those conversations get more honest, because the failure stops being about who's resisting and starts being about a pattern everybody shares.
The second mistake is aiming at the wrong layer. An inspiring session for individual contributors, with no follow-through from their managers, produces employees who want to try the tool and managers who still ask for the old spreadsheet. That's worse than saying nothing, because it teaches people that the new behavior isn't actually wanted.
The third mistake is treating it as a single event. A keynote is a starting point. If nothing reinforces the behavior in the following weeks, the old habit wins, because the old habit has years of practice behind it and the new tool has a week.
If you only have one slot, the best placement is usually the opening general session of a leadership meeting, followed by a breakout where each leader picks the one adoption behavior they'll model and measure over the next 90 days. If you're planning that kind of agenda now, talk through the slot and the audience with his team.
In the first 30 days the visible change is language and more open sharing; by 60 to 90 days, the measurable change should show up in active use of the tools you already own.
Concrete expectations, so you can judge whether it worked.
Weeks one and two. People start naming the hesitation out loud, often with some humor. "That's the Caveman talking" shows up in stand-ups and team channels without prompting. That isn't the result, but if it doesn't happen, the talk didn't land, and you should know early.
Around 30 days. If leaders have been given something to reinforce it, the first measurable signal is usually sharing. More people post what they tried, including what didn't work. Hidden AI use comes into the open. That can look messy, because you're now seeing experiments and mistakes that used to stay private. Plan for that, or you'll misread it as a setback.
60 to 90 days. Watch weekly active users by team, feature usage in the tools you care about, and whether the old workflow is actually being retired. All of these are behaviors that carry social exposure, which means all of them are what the brake suppresses. If they move, the mechanism moved.
Why 90 days and not one week? Behavior takes repetition to stick. In a 2010 study in the European Journal of Social Psychology, Phillippa Lally and colleagues at University College London followed 96 volunteers building a new daily behavior. The median time to reach automaticity was 66 days, with a range from 18 to 254 days. A single hour on a stage starts the process. The weeks after it decide whether the new behavior becomes the default. That's why companies that want the 90-day result usually ask about follow-up options when they request a quote.
What won't have moved in 90 days. Enterprise-level return on the technology, productivity figures, and headcount decisions are lagging measures with too many other inputs to pin on one keynote. Anyone who promises you an ROI number in one quarter from one talk is guessing.
The most common mistake planners make booking a technology keynote speaker is choosing someone who describes the future when the audience's actual problem is using the present.
Here are the others worth avoiding.
Booking on the demo reel. A speaker's highlight video shows how the room felt. It says nothing about what the room did afterward. Ask for the behavior and the measure, not the clip.
Letting the vendor pick. If your technology vendor offers to sponsor or supply the keynote, you'll likely get a talk about the vendor's product. That's fine for a product session. It won't address why your people aren't using the product you already have.
Assuming the audience is excited. Planners at technology events often assume everyone in the room is an enthusiast. Many aren't. Pew Research Center's survey of 5,273 employed US adults, published in February 2025, found 52% of workers were worried about the future impact of AI use in the workplace, and 36% felt hopeful. A talk pitched only at enthusiasts loses half the room in the first five minutes.
Skipping the leaders. Booking an adoption keynote for employees while senior leaders skip the session sends a message nobody misses. The people in the room conclude the change is for them, not for the people above them.
Treating fear as ignorance. Some resistance to AI is a reasonable fear about jobs. Answering it with another explanation of how the tool works feels dismissive, because the question was never how it works. It was what happens to me.
This approach doesn't fix a bad tool, a broken integration, missing training, unaddressed layoff fears, or a strategy nobody has decided on, and a speaker who won't say so shouldn't be trusted with your room.
Here are the real limits.
It doesn't fix a bad tool. If the technology genuinely doesn't do the job, people are right not to use it. Fix or replace the tool first.
It doesn't fix a broken integration. If the new platform doesn't connect to the systems people rely on, using it means double work. That's an engineering problem, not a hesitation problem.
It doesn't fix missing training. If people haven't been shown how to use the tool, show them. This work applies to the gap between trained and using.
It doesn't fix a real threat. If your company is planning to cut the roles of the people being asked to adopt the tool, their resistance isn't a Caveman reflex. It's accurate. A keynote delivered into that environment can make things worse, because it looks like a substitute for honesty that's actually owed.
It doesn't replace a strategy. If leadership hasn't decided what the technology is for, no amount of individual adoption will add up to a result. Decide first.
It doesn't produce an ROI number in one quarter. Return on technology lags, and it has too many inputs to attribute to an hour on stage.
What's left after all that still adds up to a lot: the capable, trained, reasonably secure person who knows the tool would help and keeps finding reasons not to use it. In most companies going through a digital transformation, that isn't a small group. It's most of the workforce, and it's who this work is built for. If that's the gap costing your company the most, tell his team about your event and your audience.
Dr. Noah St. John is a strong choice for a technology keynote speaker when the goal is getting people to actually use technology the company has already bought, because his work targets why capable people resist change. He has 29 years of coaching senior operators and 27 books in print. If your audience instead needs a forecast of what's coming, a futurist or AI expert is the better fit.
Dr. Noah's signature keynotes run $25,000 to $50,000, scaled to your event and audience. His guide to keynote speaker fees breaks down what to budget when comparing speakers.
No, Dr. Noah St. John isn't a technologist or a futurist, and he doesn't forecast technology trends. He speaks to the human side of technology adoption: why trained, capable people avoid tools that would help them, and what leaders can do about it. He's the right booking when the tools are already in place and usage is the problem.
A keynote about AI adoption should cover the specific behavior your people need to do more often, such as using a new tool in their weekly work or sharing what works, and why capable people hesitate to do it. It should name a measure you already track and a 30- to 90-day time frame. General excitement about AI fades within days.
A keynote speaker can help digital transformation when the talk targets the human resistance that stalls adoption after the technology is in place. It can't replace training, integration work, or a clear strategy. The early signal that it's working is usually more open sharing of experiments within the first 30 days, followed by movement in active usage.
Yes, technology company sales kickoffs are a strong fit, especially for Caveman Selling™ and The Caveman Conversion Code™. That content targets the hesitation sellers show when they need to hold the price, ask for the decision, or tell a prospect the truth about implementation.
To book Dr. Noah, send your event date, location, audience, and the adoption problem you want to move through his speaking request page. The first step is a short call with the person who owns the adoption outcome, so the keynote is built around your rollout rather than a canned deck.
A technology keynote typically runs 45 to 60 minutes as a general session. Dr. Noah also delivers half-day workshops for leadership teams that want to turn the keynote into a working session, with each leader choosing one adoption behavior to model and measure over the next 90 days.
Expect shared language within two weeks, more open sharing of AI use and experiments by 30 days, and measurable movement in active usage of your existing tools by 60 to 90 days if leaders reinforce it. Don't expect an enterprise ROI figure to change in one quarter, since that's a lagging measure with many inputs.
Dr. Noah St. John, The Caveman Conversion King
Founder of NoahMentor.com
