TL;DR - Dave Edelman has been working on personalization since 1986, back when databases were the new thing in marketing, and he watched it grow into what he and his co-author call a $2 trillion prize. Then last week he published a warning that we’re about to drown our customers in a tidal wave of hyper-personalized garbage. We got into the piece of junk mail that made him buy solar panels without ever looking at a competitor, the cheapest move almost nobody makes, and what separates personalization that earns trust from personalization that just makes somebody feel watched.

What You Can Expect

* A test you can run on your own next campaign: are you orchestrating what reaches a customer, or letting four departments hit the same person independently?

* The one question Dave asks every company he walks into, and why most of us can’t answer it honestly.

🎥 Catch all future interviews, subscribe!

🎧 Prefer audio? Listen on Apple Podcasts

Outline

(00:35) – Marketing since 1986, and riding every wave since

(03:55) – Why he wrote the book: slop was the danger

(07:35) – The solar mailer: Google Earth, Zillow, and 21.3%

(11:55) – Test and learn, and how Sungevity actually got there

(13:52) – Zero-party data, and the companies that never just ask

(17:15) – The front desk secretary who tested every new tool

(18:28) – Morgan Stanley, and ten hours a week back

(22:31) – One customer, four departments, all bombarding

(25:45) – Where trust turns into feeling watched

(28:26) – Why this is a strategy decision before it’s a marketing one

(31:15) – Call center transcripts, and catching a bug in fifteen minutes

(32:11) – The one thing: the high road

My Takeaways

The junk mail that worked. Dave’s town offered a property tax discount for solar panels, which set off an avalanche of mail. One envelope did something the others didn’t. He typed in a personalized URL and got a Google Earth image of his own roof with panels drawn onto it, sized to his roof’s orientation and latitude, with his likely annual usage estimated from the square footage on Zillow.

“So they had the numerator from Google Earth, the denominator from Zillow. It came to 21.3%. They said they would guarantee 20%.” — Dave Edelman

He bought without looking at a competitor, stayed seven years, and cleared 20% every year but one, when they sent a rebate check instead. And you don’t get there without 25 iterations, but it started with thinking, how can we truly delight our customers?

Delight is where almost everyone stops. The book runs on five promises: empower me, know me, reach me, show me, delight me. I don’t think we normally get to the delight area, it usually stops way before that, and the delight-me ones are exactly the ones that hold customer trust for years. What I hadn’t thought about is that AI is what finally makes that last promise affordable, because the testing that gets you there also feeds the thing doing the testing.

“It also feeds your AI. You get more data from the tests that make that AI smarter.” — Dave Edelman

Orchestrate, or bombard. One retailer ran every department on its own marketing budget, and the best customers of one category turned out to be the best customers of most of the others, so the same people got hit from every direction at once. Response rates fell.

“The customer isn’t a customer of a category, a department. Customer is a customer of our brand overall.” — Dave Edelman

They pooled the data, used AI to rank the top three things any given person would actually care about, and governed every message down to those three. Loyalty memberships up, sales up over 20% a year, cited at their investor day. To me that tells the customer we value your attention, which is the same line I’ve drawn before on when personalization crosses it.

It’s a strategy decision before it’s a marketing one.

“The first step is, this is a strategy decision. This is about how you want to compete. And that’s a real strategic decision, to restrain yourself in order to get a higher response rate.” — Dave Edelman

I coach CEOs on AI strategy, and time and time again I see the left hand of an organization with no idea what the right hand is doing, both of them touching the same customer, sometimes to the detriment of the other. And the customers just don’t feel known. Now, every company Dave named is enormous, so the fair question for the rest of us is what this looks like at forty people. His cheapest fix scales all the way down: poll your own employees first. Which reminded me of a CEO I worked with who had his front desk secretary, about 55 and not remotely techie, walk every new sign-up flow before launch and report back to marketing. It’s not hard, but it’s so rare. Same gap I keep running into when AI initiatives stall.

One Question Before You Go

Dave’s own test is the one I’d carry out of this hour.

“What is your intent? Is your intent to add value to the customer and be the brand that’s known for that, or are you just manipulating them?”

He calls the first one the high road, and he’s watched it build brands since 1986.

So put yourself in the shoes of your customer, of your audience. Would they like what you’re doing, would they trust you, and would they be delighted?

Watch the full conversation above, and go follow Dave on LinkedIn, where he publishes on this constantly.

About David Edelman

David C. Edelman teaches marketing at Harvard Business School and advises CEOs and CMOs through Edelman Advisory Services. He was Chief Marketing Officer at Aetna, co-led McKinsey’s global digital marketing and sales practice for eight years, and was an EVP at Digitas and a partner at BCG before that. Personalized: Customer Strategy in the Age of AI, co-authored with BCG’s Mark Abraham, came out from Harvard Business Review Press in October 2024. More than a million people follow his writing on LinkedIn.

About me

Joel Salinas is an AI Strategy Coach for founders and leaders, from solopreneurs to teams. AI is everywhere; judgment is scarce. Joel helps leaders adopt AI without outsourcing their judgment to it, through the AI Judgment Workshop and the 90-Day Judgment Engagement. Creator of the AI Leadership Triad. He writes Leadership in Change.

Written by a human, for humans.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

TL;DR - Nir Eyal wrote Hooked, the book a generation of product teams used to build habit-forming apps, and then wrote Indistractable to hand the rest of us the counter-move. In this Substack Live, he made a case I wasn’t expecting: only about 10% of our distraction comes from pings and notifications, the other 90% comes from inside us, and the fix isn’t more willpower but a calendar. We got into why to-do lists are broken, why checking email is the most dangerous distraction there is, and why he thinks self-regulation is the skill of the century.

What to Remember

* The one-sentence test Nir uses to tell traction from distraction, which has nothing to do with what the activity actually is

* Why he calls the to-do list a broken tool, and the timeboxing move he replaced it with

🎥 Catch all future interviews, subscribe!

🎧 Prefer audio? Listen on Apple Podcasts

Outline

(1:19) – Hooked and Indistractable are complements, not opposites

(5:07) – Internal triggers, and the month I deleted every news app

(6:24) – The book about focus Nir can’t stand

(7:22) – Only 10% of distraction comes from the pings

(8:56) – Traction, distraction, and the Latin root of both

(10:49) – The email trap: work that isn’t the work

(12:27) – Name one product built to help you use it less

(14:20) – Effort pacts, and Viktor Frankl’s space

(15:43) – Why to-do lists are broken

(17:22) – Timeboxing, and the real metric of success

(22:30) – Self-regulation is the skill of the century

(28:36) – Persistence, belief, and Beyond Belief

My Takeaways

Ninety percent of it is coming from inside the house

So this is the number I had to go back and check twice. We blame the pings, the dings, and the rings, and Nir puts those at roughly 10% of the problem. The other 90% is an internal trigger, which is really just a feeling:

“Loneliness, boredom, fatigue, uncertainty, anxiety.”

Then he closes the escape hatch.

“It’s not your fault. You didn’t create these digital technologies. But it is your responsibility because who else is responsible?”

Traction and distraction are the same word

Here’s the reframe I’ll actually use. Both come from the Latin root trahere, to pull, and both end in the same six letters, action. So the test is simple:

“anything that is not what you said you were going to do in advance is by definition a distraction.”

Watching football isn’t a distraction if you planned it, and answering email is one if you didn’t. That second half is what got me, because Nir says the work-shaped distractions are the worst kind, “because you don’t even realize you’re distracted.” I said on the recording that this was hitting close to home, and I meant it.

The to-do list is a register of things you failed to do

Honestly, this is the part I didn’t want to hear because I tend to live by my to do list. A to-do list has no constraints, so you always add more and you always pick the easy item over the important one, and then you read back the leftovers night after night until you’ve built a fairly convincing case against yourself.

“There’s nothing wrong with you. You’re not broken. It’s this stupid time management technique we call to-do list that’s broken.”

His replacement is timeboxing, and it moves the finish line: success isn’t the checked box but “did I work on what I said I was going to work on for as long as I said I would without distractions?”

Be stingy with your time and generous with your money

This is the line I’d put on a wall. We clip coupons, we split checks at lunch, we hunt for deals, and then we hand our hours to whoever asks. “You can’t make more time. I don’t care if you’re Bill Gates or Elon Musk.” He carries it into parenting, calls self-regulation “the skill of the century,” and describes sitting down with his six-year-old to build her calendar, video games included. My version is a Bloom card that locks my work apps at 8 p.m., so getting back in means walking to my bedroom and tapping my phone against it. That walk is the whole point.

One Question to Sit With

What’s on your calendar this week that’s still only on your to-do list?

Then watch the full conversation above, and go subscribe to Nir And Far, grab Indistractable, and pick up his new one, Beyond Belief.

About Nir Eyal

Nir Eyal is the author of Hooked, Indistractable, and Beyond Belief, which are built on his work in behavioral design and human motivation. Hooked and Indistractable have sold more than a million copies in over 30 languages, and Beyond Belief is an instant New York Times bestseller. He’s a former lecturer at the Stanford Graduate School of Business and the Hasso Plattner Institute of Design at Stanford, and his writing has appeared in The New York Times, Harvard Business Review, and The Atlantic. He writes Nir And Far on Substack and gives away a deep library of free tools at nirandfar.com, including a schedule maker and a habit tracker.

About me

Joel Salinas is an AI Strategy Coach for founders and leaders, from solopreneurs to teams. AI is everywhere; judgment is scarce. Joel helps leaders adopt AI without outsourcing their judgment to it, through the AI Judgment Workshop and the 90-Day Judgment Engagement. Creator of the AI Leadership Triad. He writes Leadership in Change.

Written by a human, for humans.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

TL;DR: Charlene Li has been standing at the front of every big technology shift since newspapers went online in the early nineties, and she just published Winning with AI with Dr. Katia Walsh. We spent most of this conversation on the thing most of us won’t say out loud, which is the fear that we already missed it. Her answer is that we are all absurdly early, that fluency is the whole game, and that the only way to get it is to be bad at this in public for a while. Watch this one if you’ve been measuring yourself against everybody on LinkedIn who looks like they have AI figured out already.

🎥 Catch all future interviews, subscribe!

🎧 Prefer audio? Listen on Apple Podcasts

Outline

(00:25) – Who Charlene Li is, and why it’s never about the technology

(01:39) – From the newsroom to Forrester to Altimeter Group

(03:41) – Electricity, steam-powered factories, and why AI moves faster

(06:24) – The biggest misconception, and “you’re not even close to beginning”

(08:44) – Fluency vs. literacy, and why the only way through is through

(11:36) – The trust wall, and how a leader destigmatizes AI in one meeting

(13:48) – Why write a book on AI when everything keeps changing

(16:42) – Step one, and the one line to add to the end of every prompt

(19:27) – Pick your lane, and the board member who owned ethical AI

(25:35) – “You’re not behind. You’re just getting started.”

(33:09) – Curiosity, imagination, and using AI to find your blind spots

My Takeaways

You’re not behind. So this is the piece I keep coming back to, because I hear that fear from almost every leader I sit with and I’ve felt it in myself. Charlene has been in this since she helped a newspaper go online in 1993, she wrote the book on it, and she still took three weeks off in January to care for loved ones, came back to Claude Cowork, and had no idea what had happened while she was gone. She felt behind too. Then she took a breath and turned it around on the rest of us.

“You’re not behind. Just look around you. How many people haven’t even started yet? So if you start today and just do one thing, you’ll be ahead of the vast majority of people.”

One of my clients is a CEO with over 45 years of practice who started sending me links to artifacts he’d built in Claude, and he could not believe he made them himself with nobody helping him. That’s what starting looks like.

Fluency is a practice, and practice is the only door. Honestly, this is where most AI advice falls apart, because it tells you what the thing is and then leaves you standing there. Charlene splits AI literacy, knowing what it is, from AI fluency, which is what happens once you’ve used it enough that you stop thinking about it. There’s one way to get there and she doesn’t pretend otherwise.

“It will feel messy. It will feel like you don’t know what you’re doing... But there is no way around it. You have to go through it. The only way through it is through it.”

I said something close to this on the live about electricity, that plugging AI into an organization nobody wired for it does nothing, but hers is the harder version, because wiring a house is a project you finish and this is a thing you only get by being bad at it for a while. She also gave the best beginner tip of the night: add “before you start, ask me any clarifying questions you may have” to the end of your prompt, and the whole thing turns from a vending machine into a conversation.

Mindset comes before tools, and before governance. Look, most of us hand AI to the CIO, buy everyone a license, and call that a rollout, and I keep watching it play out the same way. Her line on it was the sharpest thing she said all evening.

“This is not something you delegate. And what I’ve seen now is that leaders are abdicating this role.”

The fix she gave is almost embarrassingly small. Use AI to prep for your next meeting, tell the room you did it, then ask them how they used it, which sets the expectation and takes the shame out of it in the same breath. If your people think using AI counts as cheating, no tool budget on earth saves the rollout. That’s the same wall I wrote about in the empathy gap in AI implementation.

The thing I’m stealing from her. Here’s the part I hadn’t thought of, and I said so on air. Charlene feeds her own meeting recordings to AI and asks it what went wrong in the room.

“Take a transcript of my last meeting because something in it didn’t feel right. Something went off the rails and I’m not sure what’s happening. Diagnose that and give me suggestions.”

It comes back and tells her she never set an agenda, or that she talked for 70% of a call she should have been listening through. I’ve fed transcripts into Claude plenty of times to build a post out of them, and it never once occurred to me to point that back at myself and ask where I got in my own way. I use Granola to capture the calls, so the raw material has been sitting there the whole time, which is really the argument for why none of us is behind. The thing that moves you forward this week is probably already open on your screen, and you just haven’t pointed it at anything that matters yet.

The mindset work Charlene is describing, getting your own hands dirty before you ask your team to, is exactly what I walk leaders through in the AI Judgment Workshop, which is open for registration now. If you’d rather go deeper inside your organization, that’s the 90-Day Judgment Engagement.

One Question to Sit With

Charlene left this one hanging and I haven’t been able to put it down:

What could you actually do if everybody in your company was fluent with AI, not licensed, fluent? What would you build that you’ve already talked yourself out of?

Watch the full conversation above for her answer, and for the line I keep repeating to myself, that you’ve never been in control anyway. When you’re done, go subscribe to Charlene Li on Substack.

About Charlene Li

Charlene Li is a New York Times bestselling author whose books include Groundswell, Open Leadership, and The Disruption Mindset. Her latest, Winning with AI: The 90-Day Blueprint for Success, is co-written with Dr. Katia Walsh, one of the world’s first chief AI officers, and it’s a leadership and strategy book rather than a technology book. She founded Altimeter Group in 2008, spent years as a Principal Analyst at Forrester Research covering the rise of social media, served as Chief Research Officer at PA Consulting, and now leads her own firm, Quantum Networks Group. She’s just moved her newsletter to Substack, and the book is at winningwithaibook.com.

About me

Joel Salinas is an AI Strategy Coach for leaders at small and mid-sized businesses and nonprofits. AI is everywhere; judgment is scarce. Joel helps leaders adopt AI without outsourcing their judgment to it, through a monthly live workshop and 90-day engagements. Creator of the AI Leadership Triad. He writes Leadership in Change.

Affiliate links included.

Written by a human, for humans.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

TL;DR: Phil Powis ❤️⚡️ and I have been going back and forth on Substack for over a year, and his writing lands in my feed like a breath of fresh air, so this one felt less like an interview and more like two friends comparing notes. We got into why AI is nobody’s goal, why your team’s trust matters more than the tool, and the one question he keeps coming back to that can reorganize a whole business. Worth the watch if work has started to feel hollow.

🎥 Catch all future interviews, subscribe!

🎧 Prefer audio? Listen on Apple Podcasts

Outline

(00:30) – Friends across Substack, a breath of fresh air

(01:23) – Burnout, a tumor, and the business he walked away from

(07:35) – Quality of life vs. lifestyle

(09:07) – Sacred business vs. opportunistic business

(11:15) – Why AI is nobody’s goal

(16:48) – The AI Judgment Workshop and reading your team

(19:18) – The Harmony Map, and slowing down in the AI era

(27:40) – The eco rat race: old identity, new context

(29:46) – We’re all toddlers with AI

(31:48) – “What do you truly want?”

(35:39) – The Golden Sequence: connection, growth, belonging

My Takeaways

Sacred business vs. opportunistic business. So Phil opened with the story I keep thinking about. He built a boutique advertising business doing over $100,000 a month with basically him and a virtual assistant, and it nearly cost him everything, autoimmune problems, a health scare, time essentially bedridden. He walked away because his heart was never in it, and out of that came the distinction the whole conversation runs on.

“When we talk about sacred business, it’s honoring all of who you are, your gifts, your talents, your failures, your victories, and it’s tied to service. The opportunistic business is chasing the trend. It’s not really looking at who am I and what wants to come through.”

When you chase the trend, you spend your life playing catch-up on somebody else’s dream. I’ve watched leaders do it with AI too, grab the shiny thing because everyone else did, and then wonder why it drains them.

AI is nobody’s goal. Here’s where I jumped in, because I see the same thing coaching leaders on AI. I’ll ask why they’re bringing it in, and a lot of the time the honest answer is “we just want to be an AI-forward business.” That’s the trap Phil is describing, and it’s why I keep saying AI is nobody’s goal.

“So much of what I’m doing is just following my own threads of curiosity and wonder, and asking the questions I’m looking to answer for myself, and then allowing people a window into that.”

AI is the means to whatever your goal always was. Marry that goal to something you actually care about and the work stops feeling like burnout, which is the whole discipline behind the AI Leadership Triad. The tool moves you toward what you wanted, it doesn’t get to become the thing you want.

The eco rat race. Phil lived in Costa Rica for three years, and he had the funniest, truest line of the whole talk, about the founders and VCs who burn out, move there to build a wellness center, and change absolutely nothing.

“They bring the same New York energy to the picture. They’re sitting in the most beautiful locations in the world and they’re still the most stressed out people you’ll ever meet. They’re bringing that old identity to a new context.”

Carolina Wilke nailed it in the chat, you can build a new business with an old identity and hit the same walls.

You can’t relocate your way out of who you are, and you can’t AI your way out of it either.

We’re all toddlers with AI. The thing I keep coming back to is how humble Phil is about all of this. He refuses to look down on anyone’s messy AI use, and he put the reason plainly.

“With AI specifically, we’re all kind of like toddlers. And I include myself in that, and even you, because how far are we really into this? We’re a couple years into this journey.”

I study this for a living and I still bring in a new expert every Thursday, because none of us has it figured out. Phil’s antidote to the noise is one question he and Carolina ask their clients: what do you truly want. Ask it enough times and it gets dangerous in the best way, because it’ll take you down a rabbit hole if you let it. That question is the human work AI can’t do for you, and it’s the same reason I keep telling leaders not to outsource their thinking to it.

Reading your team before you buy the tool is exactly what I run leaders through in my AI Judgment Workshop, which is open for registration now. If you’d rather go deeper one-on-one or with your organization, just let me know!

One Question to Sit With

Phil ended on the question I want to leave you with too: what’s the why behind where you’re putting your hours, and does it marry where your passions lie?

Watch the full conversation above for how Phil and Carolina help people sit with that. And when you’re done, go subscribe to Sacred Business Flow.

About Phil Powis

Phil Powis is the creator of Sacred Business Flow, a Substack with more than 33,000 subscribers that he writes with his partner Carolina. He teaches founders and creators to build what he calls a sacred business, one that honors who you are and stays profitable, instead of an opportunistic business that just chases the trend. His work pairs business and marketing strategy with inner work, including the Harmony Map assessment, which reads what he calls your nine fundamental frequencies and points you to the one “breakthrough frequency” worth building around. Subscribe to his writing at Sacred Business Flow, and take the free Harmony Map assessment at sacredbusiness.com.

About me

Joel Salinas is an AI Strategy Coach for leaders at small and mid-sized businesses and nonprofits. AI is everywhere; judgment is scarce. Joel helps leaders adopt AI without outsourcing their judgment to it, through a monthly live workshop and 90-day engagements. Creator of the AI Leadership Triad. He writes Leadership in Change.

Written by a human, for humans.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

Polina Pompliano has published a deeply reported profile of an extraordinary person every Sunday since 2017, and before that she ran a daily VC newsletter at Fortune. Her book, Hidden Genius, argues that the truly exceptional don’t run on morning routines or five daily habits, but on mental models that change how they see.

We talked about how she gets past the polished version people hand her, why she thinks villains are freer than heroes, and the one thing she refuses to hand over to AI even though she uses it every day. If you write, lead, or read people for a living, this one is worth your half hour.

🎥 Catch all future interviews, subscribe!

🎧 Prefer audio? Listen on Apple Podcasts

Outline

(01:13) – A cold read: what ten minutes reveals

(03:10) – We are how we move through the world

(03:44) – The mask, access, and Logan Paul’s lollipop

(04:45) – Kevin O’Leary and the villain he plays

(10:54) – Why villains get to be human

(11:21) – Scaramucci’s Superman, everyone’s Batman

(17:42) – Understanding who you are not

(19:48) – Long-form in a short-form, AI-flooded world

(22:02) – “When I write, I think”

(25:20) – Reinvention and the Martha Stewart comeback

My Takeaways

We are how we move through the world.

So the line of hers I opened with is the one I keep coming back to, because it flips how most of us try to know somebody. We listen to what people say about themselves and take it as the truth, when the real signal is in how they actually move. Polina puts an enormous emphasis on access for exactly this reason; she wants to shadow a person through their day and catch the moment the mask slips.

“You just really have to observe. The mask, we know the mask. These are public figures, they’re very well media trained. My goal is to destabilize the talk track a little bit, or see a contradiction that everybody has.”

She told a story about Logan Paul that made it concrete: in the car, he told her he likes anime and dinosaur documentaries. Then he admitted the loud persona is a performance he switches on the second he walks onto a set. That gap between the private person and the public character is the whole game.

Villains are allowed to be human.

Here’s the part of the conversation that really made me think. She was talking about Kevin O’Leary, Mr. Wonderful from Shark Tank, and how he actually likes being cast as the villain.

“Heroes are very limited because they’re not free. They’re expected to be perfect. Villains are allowed to be human. They can have flaws, they can get angry, they can make mistakes. People are more forgiving.”

I hadn’t thought about it that way, that the villain gets more room to be a full person than the hero does. For anyone leading, that’s a real trap worth naming. The more polished and heroic the image you protect, the less permission you leave yourself to be human in front of your team.

The tell is the internal monologue.

What she’s really hunting for is the story people tell themselves, not the polished one they tell her, and she gets at it by asking how they imagined their life would turn out versus how it actually did, then listening to how they bridge the gap.

“You’re your life’s main character, you’re the protagonist, everybody else is a supporting character. So the way you tell that story tells me a lot.”

She gave the example of Anthony Scaramucci and his six-foot Superman statue, where he told her Superman is who we strive to be, but realistically we’re all just Batman, flawed and human and able to break. How someone narrates their own life explains most of what they let themselves do.

“When I write, I think.”

This is the one that matters most for the leaders reading this. Polina is not anti-AI; she said it helped catch a misdiagnosis when she was pregnant, and she uses it to pressure-test her thinking constantly. But she draws a hard line at the writing itself.

“When I write, I think. That’s how I think. So if I just offload everything to AI, then I’m not thinking at all. It’s thinking for me, and that’s really dangerous.”

She uses AI the way I’d argue every leader should, as a brainstorming partner and a sparring partner. Feed it a draft and ask what you’re missing; tell it your position and ask it to argue the other side. What she won’t do is let it do the thinking for her, which is the whole discipline behind the AI Leadership Triad. The tool sharpens your judgment; it doesn’t get to replace it.

One Question to Sit With

If AI can now imitate almost anyone’s output, the writing, the style, the answer, what’s the part of your work that only you can actually do?

Polina’s bet is that it’s the human things, the noticing, the taste, the paying attention that took her years to sharpen. Watch the full conversation above for how she gets there, and then go subscribe to The Profile.

About Polina Pompliano

Polina Pompliano is the founder of The Profile, a newsletter of deeply reported profiles of the world’s most extraordinary people, published every Sunday since 2017. She previously ran the daily VC newsletter Term Sheet at Fortune, where she wrote more than 1,300 articles. She’s the author of Hidden Genius, a book about the mental models the world’s most successful people actually use. Subscribe to her work at The Profile.

About me

Joel Salinas is an AI Strategy Coach for leaders at small and mid-sized businesses and nonprofits. AI is everywhere; judgment is scarce. Joel helps leaders adopt AI without outsourcing their judgment to it, through a monthly live workshop and 90-day engagements. Creator of the AI Leadership Triad. He writes Leadership in Change.

Written by a human, for humans.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

TL;DR - Michael Bungay Stanier has spent a decade teaching leaders to be more coach-like, and the core of it, “ask more, tell less,” matters more now than when he wrote The Coaching Habit ten years ago. In this Substack Live, he gave me four principles for leading in the age of AI: clarity over certainty, context over content, connection over cocooning, and courage over collusion. We got into why AI’s confidence is a trap, why the real work of a leader is helping people find the right question instead of handing them the answer, and the three coaching questions that still hold up in 2026. Watch the full conversation above.

🎥 Catch all future interviews, subscribe on YouTube

🎧 Prefer audio? Listen on Apple Podcasts

🤝 Want to chat one-on-one? Book a free intro call.

Outline

(0:43) – Leadership WTF: what it even means to lead now

(1:34) – Does coaching still matter when everyone has ChatGPT

(3:12) – Ask more, tell less, and why it’s harder than it sounds

(5:44) – Daniel Goleman’s six styles and the one leaders skip

(8:35) – The four Cs, laid out

(9:05) – Clarity over certainty

(11:08) – Courage over collusion

(12:45) – The sycophancy trap: six months chasing a fake theorem

(14:31) – Connection, and why humans rubbing up together still matters

(19:01) – Why big change plans fail: capacity and agency

(23:11) – The productivity fear: save 25%, produce 25% more?

(26:09) – Coaching a team in 2026: the three questions

(30:27) – Three books Michael recommends

My Takeaways

The four Cs. This was the spine of the whole conversation, and it’s the thing I’m still going over. Michael has been sitting with a simple question, what do we actually need to do as leaders in a time of AI, and he came back with four moves. “Clarity over certainty, context over content, connection over cocooning, and then courage over collusion.” I made a chart of it because I want to keep it in front of me, and honestly, once you hear all four together, you can’t unsee the tension in each one. Every C is a choice between the easy thing AI is great at and the harder thing that’s still ours to do.

Clarity over certainty.

So here’s where it got real for me. Michael’s point is that AI is fantastic at sounding sure, and that confidence is exactly the trap. “AI is brilliant at being certain,” he said, and then he told a story from a conversation with Dan Ariely (the behavioral economist behind Predictably Irrational) about a researcher who wasted six months chasing a new math theorem that ChatGPT told him he’d found, and that turned out to be complete nonsense. Certainty is cheap now. The leader’s job is clarity, which is the slower work of asking what the real problem even is before you go solve the confident-sounding one.

Connection over cocooning.

This is the one I keep chewing on. Michael admitted something I think a lot of us feel and don’t say out loud, that one of the strangest things about AI is he almost doesn’t need to talk to other human beings anymore, because it’s just him and his agent going back and forth. “It feels like it’s human, but it’s not human.” And I sat with that, because I could open Claude right now and ask it how Michael would answer a question and get a close approximation, which is exactly the thing that should scare us a little. His answer is that leaders create more than the sum of their parts when there’s real connection, and there’s something about humans rubbing up together that no agent replaces.

Capacity before change.

The most practical thing he said is that big change plans almost always fail, because leaders keep pouring water into a full glass. If your people are already at capacity, a shiny new AI rollout has nowhere to land. And this landed against a real moment for me, when someone on my team heard “AI will save you 25% of your time” and asked me straight up whether that meant I’d now require 25% more output. That fear is the whole game. Michael’s take is that the promise of AI isn’t just doing more stuff, it’s creating the capacity and the agency for people to run small experiments and actually manage the change instead of drowning in it.

The three questions that still work

So if you take one thing you can use Monday morning, take these. Near the end I asked Michael what coaching a team actually looks like in 2026, and he gave me three questions from The Coaching Habit that still hold up when AI can answer almost anything. The strategic question, “what are we going to say yes to, and if we say yes to this, what must we say no to?” The real-challenge question, “what’s the real challenge here?” And the learning question he closes every conversation with, “what was most useful or valuable here?” I’ve started using that last one, because it makes sure everyone walks away a little smarter, including me. I made a second card of these so you can keep them in front of you.

If AI can generate every answer in your organization, what’s the one thing your people still need a human leader for? Sit with that, then watch the full conversation above, and if you want the real depth on any of this, go grab The Coaching Habit and subscribe to Michael’s Change Signal podcast.

About Michael Bungay Stanier

Michael Bungay Stanier is the author of The Coaching Habit, one of the best-selling coaching books of all time, now celebrating its ten-year anniversary. He also wrote How to Work with Almost Anyone and hosts the Change Signal podcast, where he cuts through the blather to find what actually works in change. His whole mission is helping leaders be more coach-like: stay curious a little longer, rush to advice a little more slowly. Find his books and free resources at thecoachinghabit.com.

About me

Joel Salinas is an AI Strategy Coach for leaders at small and mid-sized businesses and nonprofits. AI is everywhere; judgment is scarce. Joel helps leaders adopt AI without outsourcing their judgment to it, through a monthly live workshop and 90-day engagements. Creator of the AI Leadership Triad. He writes Leadership in Change.

Written by a human, for humans.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

Things have changed.

The bar used to be produce content, produce content, produce content, and most people didn’t have the time or the persistence to keep going, so just showing up was the whole game. That bar is gone now. Anybody can type a prompt and get ten decent posts before lunch. Which means the thing that actually keeps a reader with you is the part AI can’t fake. David McIlroy has built a solopreneur business coaching creative entrepreneurs, with 34,000+ subscribers on Substack, and we spent this conversation on exactly that tension: how you grow when the tools that used to separate you from everybody else are now free to everybody else. We got into the 1:5 engagement ratio, why community became the whole point after the pandemic, and where he draws the line on letting AI touch his fiction. Watch the full thing above.

🎥 Catch all future interviews, subscribe on YouTube

🎧 Prefer audio? Listen on Apple Podcasts

🤝 Want to chat one-on-one? Book a free intro call.

Outline

(1:25) – From Medium to Substack, and starting with one publication

(3:37) – Youth worker to hiking blog to solopreneur

(5:36) – Why the digital-course boom cooled and community took over

(9:18) – Finding your why and your message

(10:49) – The 1:5 ratio: using other people’s audiences

(13:34) – The content bar collapsed with AI

(18:07) – AI should amplify your strengths, not replace your expertise

(22:27) – Using AI as an editor for fiction

(23:28) – The sensitivity check: AI on a humanitarian content team

(26:57) – David’s new group coaching program

(29:17) – Who should reach out to David

My Takeaways

The content bar hit zero. David and I got to the same place from different sides. The old advantage was volume, and volume is now free. So the question flips. If a post can come out 100% AI-created, what actually separates you from the hundred other people running the same prompt? David put it this way: “You can create. It’s the kind of content, it’s the depth that now you can go to, which makes the difference.” The depth is yours. The prompt isn’t.

Borrow other people’s audiences. When your own list is small, David’s move is a rough 1:5 ratio, one thing you post, five thoughtful comments on bigger creators’ work. And he was firm that the comment has to earn its place. “It needs to be something thoughtful that adds to the conversation and demonstrates that you have some level of expertise to share as well.” Not “great idea, happy for you.” That’s the version an AI bot posts, and readers can smell it. (I get those on LinkedIn constantly, three perfectly structured paragraphs from someone who clearly never read the post.)

Share the messy stuff. This is the one I’d underline for any leader building a presence online. David’s point was that people don’t connect with polished text, they connect with a real person behind it. “You’ve got to share some of the messy stuff as well, the struggles, the good times, the family stuff, the pets, the nature.” AI can generate the frameworks. It can’t hand you your kid’s drawing or your dog running through the grass, and that’s the part a reader actually stays for. It’s the same reason I keep coming back to community as the thing that carries you.

AI as editor, not author. David keeps AI completely out of his fiction as a writer, but he runs finished manuscripts through it to catch plot holes and inconsistencies. “It’s working like an editor on some level. It doesn’t produce any ideas or guide the story in any way, but it helps keep me right if I’m making mistakes.” That’s exactly how I use it too. On the humanitarian side, my content team of about 25 people runs work through Claude to check whether a message written by a non-Latino would read wrong to a Latino audience. Not to write it. To catch what a human would miss and then keep going. The line David draws is the right one: the moment AI starts replacing what makes you you, you’ve stopped growing.

One Question to Sit With

Where do you draw the line, what do you let AI do, and what stays yours? The bar for producing content just hit zero, and the part that’s actually worth following is the part AI can’t fake.

About David McIlroy

David McIlroy is a strategy coach for creative entrepreneurs, a solopreneur, fiction author, and podcaster based in Northern Ireland, with 34,000+ subscribers across his Substack publications. He helps creators build a personal brand, strategy, and system for clarity and growth. His new weekly group coaching program opens Tuesday, July 28, 2026. Subscribe and reach out to him on Substack.

About me

Joel Salinas is an Executive AI Coach for leaders at small and mid-sized businesses and nonprofits. 1:1 coaching, team workshops, and AI strategy work built around amplifying what your team is already good at. Creator of the AI Leadership Triad. He writes Leadership in Change. If you want help thinking through your own AI strategy or online presence, start here.

Written by a human, for humans.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

Travis Rosser co-founded Kajabi, the platform creators have used to sell more than $10 billion in courses, coaching, and knowledge, and he joined me on Substack Live to trace the whole arc.

We went from a Shark Lagoon PDF he sold as a kid to a company that put publishing tools in everyone’s hands. Along the way he unpacked why he tells builders to fall in love with their customer’s pain instead of their idea, his Four P’s of knowledge capital, and why AI is turning all of us from creators into builders. If you’re sitting on expertise and wondering what to do with it, this is the one to watch.

🎥 Catch all future interviews, subscribe on YouTube

🎧 Prefer audio? Listen on Apple Podcasts

🤝 Want to chat one-on-one? Book a free intro call.

Outline

(00:04) – Welcome, and going live on Substack

(00:53) – A sprinkler toy, a taco, and a lifelong side-hustle habit

(02:39) – Shark Lagoon: the first thing he ever sold online

(04:36) – Kajabi’s first customer

(05:28) – Launch day, and a million dollars in 24 hours

(08:44) – The Saturday the support tickets caught fire

(09:56) – The survey that revealed what Kajabi really was

(10:32) – From $10 billion to “we’re all becoming builders”

(12:13) – “Fall in love with your customer’s pain, not your idea”

(15:24) – Knowledge capital and the Four P’s

(22:57) – Skills as the new digital product, and taste as the new edge

(36:43) – What Travis is building now: Wazi and PageSumo

My Takeaways

Fall in love with your customer’s pain, not your idea. Here’s the line I keep coming back to from the whole conversation. Travis has been building things his entire life, and the thing he kept circling back to is that being good at creating stuff is exactly what makes you fall in love with the wrong part of it.

“When you’re good at creating things, you fall in love with the idea too much,” he said. “But if you’re not helping someone, if you don’t provide value to them, it doesn’t matter.”

What I pushed on with him is how much sharper that line cuts right now, because once anyone can build almost anything over a weekend, a lot of us are going to mistake “I can make this” for “someone actually needs this,” and those two things were never the same.

Everybody is sitting on knowledge capital, and most of us take it for granted. So here’s the framing from his book You, Inc. that I want more leaders to steal. Travis calls it the Four P’s: Profession, Passion, Pain, and Problems. His claim is that all four are knowledge somebody else would pay for, and we discount them because they come easy to us.

“Everybody is an expert in their own life,” he said, and the example that got me was his kids yelling across the house about how to beat a level in Minecraft. In that one small section of life, his oldest son was the expert. So the thing he leaves you with is simple and a little uncomfortable: you already have something worth teaching, and the only real work is looking back at your own trail long enough to notice what it is.

We’re shifting from a creator economy to a builder economy. This is the part I was leaning in for, because it’s exactly what I’ve been living. Travis’s take is that AI has handed everyone the capacity to build, not just to make content.

“AI has given us the capacity to build, where anybody can build software, anybody can build agents, workflows,” he said, “and I really encourage anybody that watches this to go start doing that just for fun on the side.”

I told him about the workout app my wife and I both wanted but couldn’t find, so one weekend I sat down in Claude Code (Anthropic’s coding tool you run by talking to it), took screenshots of the two apps we were each paying fifteen dollars a month for, listed the features we actually wanted, and built our own. I’ve used it almost every day for six months. That used to take a team and a budget, and now it takes a curious Saturday.

In the age of infinite output, taste is the edge. So if anybody can build anything, what’s actually scarce now? Travis’s answer is taste, and taste comes from being an expert who knows what to leave out. He pointed to In-N-Out here in Southern California, how the whole brand is a curated menu, a few things done well instead of a wall of options.

“It’s really good at producing a lot of stuff quickly,” he said about AI, “but how do you have taste? How do you curate the good? That’s the next real expert, the person who knows how to navigate that and still produce high-quality stuff.”

His advice for getting started is to build in a vacuum for a while and be your own first customer, because the fastest way to develop taste is to make something you’d actually use. That’s the same thread of adaptability and creativity I keep writing about in the AI Leadership Triad. The tools got cheap, so judgment is the job now.

One Question to Sit With

What’s the one pain, yours or a customer’s, that you’d build for first now that the tools aren’t the hard part anymore?

Watch the full conversation above, and then go follow Travis Rosser and check out what he’s building at Wazi and PageSumo.

About Travis Rosser

Travis Rosser is the co-founder of Kajabi, the platform creators have used to sell more than $10 billion in courses, coaching, and digital products. He’s the author of You, Inc., a book about turning your own knowledge into a business through what he calls the Four P’s. These days he’s building two new tools with AI: Wazi (wazi.ai) and PageSumo (pagesumo.com). Follow his work on LinkedIn.

About me

Joel Salinas is an Executive AI Coach for leaders at small and mid-sized businesses and nonprofits. 1:1 coaching, team workshops, and AI strategy work built around amplifying what your team is already good at. Creator of the AI Leadership Triad. He writes Leadership in Change. If you want help thinking through your own AI strategy or online presence, start here.

Written by a human, for humans.

Thank you Colette Molteni, Claire Machado, Duncan The Sage, Tiffany Farley, Chris Winter, and many others for tuning into my live video with Travis Rosser! Join me for my next live video in the app.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

Most of us start a business to get our time back, and somehow end up the one person it can’t run without. In this Substack Live I sat down with Cory Blumenfeld, a five-time founder who built a virtual assistant company called BlueMoso, on how to stop being the bottleneck in your own company. We get into why the system usually only exists in the founder’s head, how to hire for the match instead of just the skill, and the line I keep coming back to: AI is for tasks, people are for outcomes.

Watch the full conversation above.

Also, click below to subscribe on Apple Podcasts! 👇

Outline

(00:00) – Intro: who Cory is and how he got to Substack

(01:20) – Why human content is becoming a premium tier

(01:56) – Five businesses, zero experience, and learning by building

(08:42) – The founder bottleneck: leading from the backseat

(10:06) – If you got sick tomorrow, would the business survive?

(14:28) – Type-A founders, trust, and hiring for the match

(16:46) – The hiring framework: skill, work style, communication, personality

(20:03) – VAs are for outcomes, AI is for tasks

(22:10) – Humans with AI, not AI versus humans

(23:25) – Everyone is starting at zero with AI

(33:49) – Why he named it Taking My Time Back

My Takeaways

The founder is the bottleneck. So the whole conversation really starts here. The minute a company can’t survive without the founder in every decision, the founder has stopped leading and started blocking. Cory said it about as bluntly as it gets:

“Any founder, any CEO trying to lead from the backseat drives the car off the road.”

I’ve watched this one up close, and the test I keep coming back to is simpler than people want it to be. If it’s all hinging on you, you’re not leading well, you’re just busy.

The system only exists in your head. Here’s the part founders hate to hear. When delegation fails, we blame the hire. Cory’s take is that the real problem is upstream, because the playbook never left your head. He fixed it by writing his whole content process down so completely that his assistant couldn’t fail.

“I documented my strategy, made it super clear so they couldn’t fail. I set them up for success, and it started working.”

That reframed it for me. The bottleneck usually isn’t your team’s ability, it’s the stuff you’ve never bothered to get out of your own head.

Hire for the match, not just the skill. When I asked Cory how he actually trusts people enough to step back, he didn’t start with skill. He started with fit, work style, communication style, personality, and then skill.

“Hiring someone is not just skill alone. What’s their work style? What’s their communication style? The match is everything.”

And he was honest about why most of us struggle with it. A lot of founders are type-A with trust issues, me included on some days. If you hired someone for a skill you don’t have, the worst thing you can do is hover. Give them the room to be good at the thing you brought them in for.

AI is for tasks, people are for outcomes. This is the line I’ll be repeating for a while. Cory draws a clean line between what you hand a person and what you hand a machine.

“You don’t bring in a virtual assistant to work on a task. You bring them on to deal with an outcome. You bring on AI to deal with individual tasks.”

I told him on the spot that’s a quote CEOs could put on their walls, and I meant it. It also points at where this is going. When everyone has AI doing the repetitive work, the human part, the judgment, the relationship, the actual experience, stops being the cheap part and becomes the premium one.

Figuring out where AI fits and where your people have to stay is most of what I work on with the leaders I coach. If you want to talk it through, my calendar is here.

One Question to Sit With

If you got sick tomorrow and unplugged for two weeks, would your business still be standing when you got back? Sit with the honest answer.

Watch the full conversation above, and go subscribe to Taking My Time Back, Cory’s Substack on helping founders win their time back.

About Cory Blumenfeld

Cory Blumenfeld is the Founder and CEO of BlueMoso, a managed virtual assistant service for founders, executives, and agencies, with a vetted talent pool of more than 70 assistants and specialists. He’s a five-time founder with two exits across health tech, fintech, and outsourcing, and he runs his companies remotely from Playa del Carmen. He writes Taking My Time Back on Substack.

About me

Joel Salinas is an Executive AI Coach for leaders at small and mid-sized businesses and nonprofits. 1:1 coaching, team workshops, and AI strategy work built around amplifying what your team is already good at. Creator of the AI Leadership Triad. He writes Leadership in Change. If you want help thinking through your own AI strategy or online presence, start here.

Written by a human, for humans.

Thank you Ankita Chatrath, Janet Macaluso, and many others for tuning into my live video with Cory Blumenfeld! Join me for my next live video in the app.

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe

Much of the AI conversation in the business world, for those who have not used AI much, is some version of: kick your feet up, drink the piña colada, watch your revenue shoot up.

I’ve seen that promise up close on the fundraising side of the humanitarian work I do, and on a lot of teams AI is the enemy in the room before it’s anything else. Mallory Erickson started exactly where most of her sector started, nervous, protective, sure AI was going to hollow out the most human work there is, and then she built something that does the opposite. She’s spent her whole career on connection, she’s trained over 100,000 fundraisers, and instead of letting AI replace the fundraiser, she flipped the entire thing and made the donor the robot so the human gets to practice. Watch the full conversation above.

Also, click below to subscribe on Apple Podcasts! 👇

Outline

(0:31) – Bringing AI into deeply human work without hollowing it out

(1:03) – From accidental fundraiser to founder of Practivated

(2:30) – Why she avoided tech, and the fear that started it

(4:16) – Personalization at scale, but for fundraisers

(9:26) – The rise of “robot fundraisers” and why it broke her

(10:28) – The flip: what if the donor is the robot?

(15:50) – Start with the problem, not the AI

(21:55) – Rolling AI out to a hesitant team: validate the fear

(25:26) – Give your team a stipend to play

(31:42) – The results: 280% more human donor touchpoints

(32:39) – The LeBron test: the best still put in the reps

(38:11) – A charge to practitioners: build what you need

My Takeaways

The flip. Most AI products aimed at fundraisers are built to replace the fundraiser, the autonomous “robot fundraiser” that writes the appeals and works the donors so you don’t have to. Mallory watched that trend and reacted the way I wish more people would, by refusing it:

“If our answer to fundraiser burnout is to replace human fundraisers with robots, we can do so much better than that.” So she flipped it. “What if the fundraiser isn’t the robot? What if the donor is?”

Inside Practivated the AI plays the donor, the human role plays the hard conversation, gets scored, and builds the muscle memory before the real ask. I keep coming back to that flip, because it’s the same move every leader should be making, which is to point AI at the practice and the reps instead of at the people doing the work. It’s the version of AI I argued for in Can AI Make You a More Human Leader?, and she actually built it.

Start with the problem, not the AI. This is the part of the conversation I keep coming back to, because it’s exactly what I tell the executives I coach. Mallory has closed deals lost on purpose, walked away from organizations that came to her wanting “more AI” with no real problem behind it. Her test is simple: what have you been trying to solve that you’ve thrown everything at and still can’t crack, and is there an emerging tool that could finally move it? As she put it,

“AI is not the point. The goal is to solve your challenges.”

Validate the fear before you roll anything out. I asked her how she’d introduce AI to a team that’s all over the map on it, and her first move surprised me, because it had nothing to do with the tool. She said there shouldn’t be AI adoption without first being ready to acknowledge and validate people’s fear, not to agree that AI is scary, just to give the fear a place to exist before it hardens into resistance.

“It’s really important that we’re acknowledging and validating how people feel.”

Then she does something I love, she gives her whole team a monthly stipend to just play, with two rules: have a hypothesis going in, and come back and tell the team what you learned.

The reps don’t stop because you’re good. The skeptic line Mallory hears most is “I’ve been fundraising for twenty years, I don’t need to practice.” Her answer is the one I’m going to start stealing in my own coaching:

“Can you imagine if we said LeBron James is such a good basketball player, he doesn’t need to practice anymore? Simone Biles? Absolutely not. The people at the top of their game are putting in the reps.”

And the numbers are backing her up, a 280% increase in human-driven donor touchpoints, onboarding cut from eight weeks to two, ask effectiveness up 33% in the first 30 days. AI didn’t replace the human in any of that, it gave the human a place to practice so they could show up more present when it actually counted.

One Question to Sit With

Where in your own work are you reaching for AI before you’ve named the problem it’s actually supposed to solve? Mallory’s whole practice is built on refusing to skip that question, and I haven’t stopped chewing on how often the rest of us do.

Watch the full conversation above, and then go follow Mallory’s work at Practivated and her What the Fundraising podcast.

About Mallory Erickson

Mallory Erickson is an executive coach, fundraising consultant, and the host of the What the Fundraising podcast. She has trained over 100,000 fundraisers, co-founded the Fundraising AI initiative, and is the founder of Practivated, an AI practice platform where fundraisers role play donor conversations and get coached on what to do differently next time. Her North Star, in her own words, is improving the lives of fundraisers. You can find her work at practivated.com.

About me

Joel Salinas is an Executive AI Coach for leaders at small and mid-sized businesses and nonprofits. 1:1 coaching, team workshops, and AI strategy work built around amplifying what your team is already good at. Creator of the AI Leadership Triad. He writes Leadership in Change. If you want help thinking through your own AI strategy or online presence, start here.

Written by a human, for humans.

Sidenote… do you write on Substack? Build smarter, not just harder, with NewsletterCompass.com. As a co-founder, I’m giving you 50% off for life… 👇

This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit leadershipinchange.com/subscribe