In this episode, Geoff Woods, bestselling author of The AI Driven Leader and founder of AI Leadership, reveals how leaders can leverage AI to tackle big decisions and strategic challenges. Geoff introduces a transformative mindset shift: stop asking, “How can I do this?” and start asking, “How can AI help me do this?” Through real-world examples, he demonstrates how AI can act as a thought partner in high-stakes scenarios like debt restructuring, product innovation, and boardroom dynamics. Geoff also shares practical tips on building daily AI habits to enhance decision-making, amplify results, and spark innovative breakthroughs. Whether you're a leader looking to rethink your approach to AI or seeking actionable strategies for smarter decision-making, this episode offers valuable insights for driving success.
In this episode, Geoff Woods, bestselling author of The AI Driven Leader and founder of AI Leadership, reveals how leaders can leverage AI to tackle big decisions and strategic challenges. Geoff introduces a transformative mindset shift: stop asking, “How can I do this?” and start asking, “How can AI help me do this?”
Through real-world examples, he demonstrates how AI can act as a thought partner in high-stakes scenarios like debt restructuring, product innovation, and boardroom dynamics. Geoff also shares practical tips on building daily AI habits to enhance decision-making, amplify results, and spark innovative breakthroughs.
Whether you're a leader looking to rethink your approach to AI or seeking actionable strategies for smarter decision-making, this episode offers valuable insights for driving success.
Key takeaways:
AI leadership website: Home | Aileadership
Geoff on LinkedIn: Geoff Woods | LinkedIn
00:00 Introduction to Geoff Woods and His Journey
00:27 Mindset Shift for Leaders Using Generative AI
01:19 The Power of AI as a Thought Partner
03:55 Strategic Use of Generative AI
05:22 Real-World Example: Restructuring Debt with AI
08:45 Daily AI Engagement for Business Growth
13:34 Building AI Proficiency and Overcoming Challenges
19:28 Creating and Leveraging AI Boards for Decision-Making
24:01 Insights from AI Advisory Boards
25:05 Strategic Thinking and Habit Formation
27:03 Practical Applications and Case Studies in AI Leadership
33:12 Case Study: Strategic Offsite with AI
37:48 Final Thoughts and Takeaways
📜 Read the transcript for this episode: Transcript of How You Can Leverage AI for Big Decisions – Insights from Geoff Woods |
[00:00:00] Geoff Woods: I'm Geoff Woods. I'm the author of the number one bestseller, the AI driven leader and founder of AI leadership. I got into AI, uh, by being in the space for a while, I built a training and consulting company based on a book called the one thing, then stepped in as the chief growth officer for a big steel company called Jindal steel ended up taking them from 750 million to 12 billion. And when I was there, I came across AI. They basically put me in charge of it. And, uh, that opened the door to a whole new future.
[00:00:27] Jeremy Utley: Would you mind to talk for a second just as we start this conversation? Um, What's the mindset shift required to get the most out of generative AI for a leader of a business?
[00:00:40] Geoff Woods: I think the simplest thing is changing one question that you're currently asking yourself, which is, how can I do this? And shifting toward asking, how can AI help me do this? Like, it's actually so simple that I will literally have people Take a sticky note and a sharpie and write, how can AI help me do this?
And literally put it on the computer. Because if you ask that question once a day, you create awareness of a potential use case. You take action by writing the prompt. And if you focus on really good communication ingredients within prompt engineering, you get a good result. And then you just want to ask the question again.
But I think what makes it real for people, Jeremy, Is when you think about, I mean, how many books have you read over the course of your life? Do you think?
[00:01:28] Jeremy Utley: It'd be too embarrassing to say because it's too large to, well, wait, are you, are you including, um, boxcar children and everything,
[00:01:40] Geoff Woods: but here, here's the realization.
What, what percent of the collective knowledge that you both have acquired over your lives do you think you can recall and apply in this second?
[00:01:51] Jeremy Utley: Of course, of course,
[00:01:52] Geoff Woods: it's like 1 percent or less, but this, this is where I
[00:01:56] Jeremy Utley: think you might know Jeff guy Kawasaki told us that his guy Kawasaki GPT is better at being guy Kawasaki than he is.
[00:02:03] Geoff Woods: I heard that interview. I thought it was awesome. And here's why I think this is so powerful. When you look at the thought partnership side, we have been wildly successful on our human processing power, right? Which is our ability to recall less than 1 percent of everything that we've ever learned, which is insignificant compared to the collective knowledge.
So if you now think of yourself as thought leader and AI is your thought partner, you can tap into 200 million books worth of data that these AI models have been trained on so far. It can recall 100 percent of it in the fraction of a second. And so if you give it the right leadership, It can take you to a whole different playing field.
[00:02:43] Jeremy Utley: So you're talking about a reframe of the user as the user is now the leader of the AI.
[00:02:49] Geoff Woods: Thought leader. AI is your thought partner. The mistake is when leaders abdicate their thought leadership and just copy and paste whatever generative AI gets them without any editorial oversight. That's just a path to mediocrity.
[00:03:02] Henrik Werdelin: When you ask a question, and you want to try to kind of double click on getting a more advanced answer, Have you found a better, um, kind of way of thinking about it than this, , what is the question that you have as you post?
What is kind of like the output that you would like? So what is the format of the output? And what is the knowledge? that you want to use as the frame for GPT to kind of, answer. So what documents or whatever, so input, output framing, and then sometimes you even need to kind of like double or triple box kind of it, because sometimes you then figure out that you have to ask it something and it has to come out with an output and then you have to give another bot that kind of input.
[00:03:51] Geoff Woods: I, I take a different approach. Okay. The vast majority of the times that I am using generative AI, I am not treating it like Google where I'm asking it a question and expecting an answer. The vast majority of the time I'm using it is for strategic thinking and decision making. So I'm describing a problem or an opportunity with lots of context.
I'm assigning it a role. I'm then turning the tables and telling it to interview me by asking one question at a time, up to three questions, up to four questions, up to five questions, depending on how deep I want it to go. It's job is now to pull the information out of my head. And then to accomplish a task.
So let me give you a real world example. And this is all rooted in the idea that I think 80 20 rule, 80 percent of your results come from 20 percent of your efforts. I think most people playing around with generative AI right now are using it for tactical 80 percent things on their to do list that drive the minority of the results in the business.
Writing a better email, helping with marketing copy. These are not leadership activities. This does not belong at the C suite yet. You've got the C suite spending their time doing it. They're doing their people's jobs. I'm just saying, let's focus on the 20 percent things that you need to focus on as a leader that'll drive 80 percent of your results, which usually comes down to strategic thinking.
So I asked a CEO, what's your biggest problem in your business? That if we could just solve that, it would make the biggest impact for you. And he goes, Oh, that's easy. I run a manufacturing company. I leased a whole bunch of equipment from this company in Japan. Things have shifted in our market and the debt structure is now killing us.
We are going to go out of business if we do not get the debt restructured. Okay, what have you done? And he starts walking me through. Oh, we tried this and we tried this and we tried this. None of it worked because this is a publicly traded company in Japan. It went all the way to the board and they are refusing to restructure the debt.
Because they are worried they will lose face in Japanese society. We have no next step. We're probably going out of business. This is a perfect AI use case. Big problem makes an impact in the business. You could use a thought partner. So as thought leader, we wrote the prompt, everything I just described to you, word for word, I typed in as the context, then assigned the role of acting as an investment banker with deep expertise and restructuring debt interview us by asking one question at a time, up to three questions to gain additional context so that you can then generate five non obvious strategies we could deploy to get the board to restructure the debt.
AI turns the tables and asks, Do you have relationships with any other influential executives in Japan that this board would respect? I looked at the CEO and he was floored. He goes, I would have never asked that question, and oh my gosh, I do! And so we fed it back to the model and it asked two more questions about how he had navigated Japanese society in the past.
And it came back and said, here's your five non obvious strategies. Number one, the saving face consortium. You have enough relationships with other influential executives in Japan, approach him to acquire the debt, give him really favorable terms. Your debt will get restructured. Your board will save face.
I looked at him, guys, his whole body language had shifted and I can tell he's holding tears back. And he looks at me and he goes, Jeff, I haven't slept in 90 days. I have literally had no next step. And I've been making peace with the fact that this business that I had built, my baby is going out of business.
And in less than 10 minutes, I see hope we texted recently. And he said, the ball's moving. And I think this is actually going to get done.
[00:07:49] Henrik Werdelin: That's incredible.
[00:07:52] Jeremy Utley: I think hope in less than 10 minutes Is a great value prop right there. So
[00:07:55] Henrik Werdelin: well, could you use that?
[00:07:57] Jeremy Utley: Yeah. Yeah.
[00:07:58] Geoff Woods: I mean, I've got so many examples like this, where again, if you are as a leader, if you recognize you now have this new thought partner at your fingertips and problems and the things that you can solve in a fraction of the time are mind blowing.
[00:08:15] Jeremy Utley: How do you think about the problems? Is it, for example, A lot of the CEOs that I deal with are about to hire a consultancy for blank. Are those the kinds of problems? You know, we need to know where the white space in our market is. We're going to hire BCG for to do a 5 million study over the course of five months and get back to us. That's a conversation I had last week.
[00:08:39] Geoff Woods: Yeah. Yeah. And that could be something you look to generative AI for. And I still think, I don't believe that consultants are out of business. I think what consultants do is going to change. I literally saw a technology yesterday that all I did was gave it my website URL and gave it some copy from my webpage and it processed a quarter of a million tokens to do a full competitive landscape and basically walk through.
Here's your competition. Here's all the things you need to think about. I mean it was probably six months worth of consulting scope that I literally saw generated in less than 120 seconds. Well, that's, that's my, that's my
[00:09:19] Jeremy Utley: question. The, of course, yeah. There's still like that big job of like, I guess when you know BCG has done it, you assume they fact checked.
Now you have this weird thing of like, is this true? But setting aside the question of consultants jobs, cause I'm actually not that concerned about them for the CEO. What I'm trying to figure out is how do they identify the thing that they should be thought partnering on? And to me, my heuristic is. The thing, historically, I've hired a consultancy to answer a month. are there other kind of shorthand for what is the thing?
[00:09:52] Geoff Woods: So I view it differently. There is, if you're like most leaders who know AI is the future, but they don't know where to start and they think they're falling behind. I actually care less what the use cases and more that once a day you ask, how can AI help me do this? And step up to the plate and swing the bat with really good communication ingredients and see if you can get on base.
[00:10:14] Jeremy Utley: How does somebody know? I actually, that was a question I was going to ask you later, but you brought it back, so I'm excited, but you started with that question of how can AI help me do this?
Um, for somebody who's just getting started. How did they know the answer to that question? So I think a lot of people, the problem actually is they can't imagine how AI can help them do it.
[00:10:33] Geoff Woods: Right. Well, let's, let's turn the tables. Jeremy, what's something on your plate today that's going to require you to think strategically?
[00:10:40] Jeremy Utley: Uh, how to structure a new relationship.
[00:10:44] Geoff Woods: How to structure a new relationship. This is a perfect use case for AI. So when I talk about prompt engineering, I think of it like cooking with ingredients in a recipe. Describing a task is an ingredient. Giving context is an ingredient. Assigning a persona.
Asking AI to interview you. These are all ingredients in your recipe of your prompt. So, , my favorite structure is context, role, Interview task with balloons. The balloons pop up when the hands do that. So for context, I'm thinking about how to structure a new relationship and then feed it lots of context.
Your role is to act as an expert in structuring these types of deals. Interview me by asking four questions. One at a time to gain additional context. And then I want you to walk me through the high level executive summary of how you would structure this and also create a section that would outline the risks that I need to mitigate and one step I could take for each.
[00:11:45] Jeremy Utley: So the answer to the question of how can AI help me do this, just to make sure I'm clear, because you said we want a leader to ask that question once a day. Yeah. It sounds like the answer is. Even if you don't know, if you have an instinct to engage AI in even collaborating on how AI can help you with it, which is super meta, but you don't actually need to know how AI can help you do this. And maybe, which is to say perhaps a different question is actually, what can AI help me do today? Right? Or something like that, right? Because how can AI help me do this makes me think that there's a bunch of different ways. It actually sounds like, no, there's one way. It's. It's have a conversation with a thought partner that you never knew you could have before.
And the bigger thing is putting something in that conversation every single day.
[00:12:32] Geoff Woods: Yeah. I mean, I did a session a few days ago for a bunch of CEOs and one guy missed the session and he runs into his roommate and goes, how was the session? I miss. He goes, Oh, I can't believe you missed it. And he goes, yeah, I'm dealing with this HR nightmare and I don't know what to do.
And the guy literally had just left my session and he goes, well, have you thought about using AI to help you solve that? And he goes, It can't do that. And he goes, watch this. And he literally took the framework that I taught him in the class that I just shared with you guys. Like give it context, describe the problem.
I'm a CEO. I'm dealing with this HR nightmare for context. Here's everything that has happened with this individual. Your role is to act as an HR professional with deep expertise in resolving these types of situations. Interview me by asking three questions, ask them one at a time to gain additional context.
And then I want you to walk me through the top five things I need to do to resolve this issue. Context, role, interview, task. If you just remember that recipe, I mean, my gosh, , you'll start light years ahead of most people.
[00:13:33] Henrik Werdelin: Have you ever found a problem that it, we're not able to solve?
[00:13:38] Geoff Woods: Sure. Um, and this is where I think it's so important to stay in the driver's seat as thought leader.
When I approach a use case, I genuinely have curiosity. Will it work? Rather than expectation of, I expect it to work, I was facilitating a strategic offsite for a company and they wanted, they literally walked in the room when I heard the CEO say, we're at a stalemate. And he looks at me and he goes, perfect timing.
And what they were trying to do is they were trying to use their last three years of performance and financials to project what the next three years would look like for the board. And they just could not get the math to work. And they'd been hammering at it for over an hour. And they said, can I help?
And I said, maybe it started to get the ball moving, but then the data wasn't structured the right way. And it imploded all over the place. And the CEO even said, I made a mistake. I asked AI to be the thought leader and I allowed myself to become the thought partner. It was too big of a use case to swing for based on our current prompting knowledge.
[00:14:43] Henrik Werdelin: That makes a lot of sense. I'm still stuck with the premise of what you're telling, which I think is so fascinating, which is most of the time when I talk to people who think about AI, , they very much think of this as a way to reduce the time that it takes to solve the 80 percent of the value that they provide in a given day.
I just think it's like such a fascinating concept to, to kind of change that mindset and saying, yeah, sure. I like you can get it to do that, but that's kind of like almost bread and butter by now. Uh, but what are the 20 percent when you introduce people to this. Is there any other tools that help them do this?
Because I think as you're alluding to, there's this weird disconnect between having been to a class, having heard this podcast, and then going on your day to day business, and then probably forgetting that they should just try that. What is your hot trigger that kind of makes you go back into this?
Or is it just habit?
[00:15:44] Geoff Woods: Honestly, that sticky note is insanely powerful. And the only reason that I bring it up here is because I've had enough CEOs put it on their computer and it changed their life in less than 30 days. I did 45 minutes with the CEO and his COO just time, like high level stuff from my book, they are driven leader and they put the sticky notes down and within two weeks, because every day they saw it and they said, , let's try this.
Like they were really committed to going through the learning curve. It started to ignite a part of their brain. that saw things in their business that they would have never seen before. They looked at one guy, they do hard money loans for people who want to do fix and flips of houses. They have a guy whose full time job is literally a budget comes in from a general contractor and he has to look up the zip code of the house, the square footage of the house.
And he knows based on zip code and square footage, what the budget should be plus or minus 5%. And if a contractor is asking for more than they're stealing, and if they ask for less, they're incompetent. And his job is to review the budgets manually. And within two weeks of them just looking at a sticky note every day and swinging for the fences, they built proficiency talking to AI that they went to him and said, Hey, I think we can help you here.
And they built a custom GPT, trained it on all the data and reports that they had. Within two weeks, they automated 90 percent of this guy's job, which was amazing because there was way more important things that the business could use him for. They moved him to higher impact priorities. It dropped a hundred thousand dollars to the bottom line in two weeks, because they started once a day asking, how can AI help me do this?
And then they attempted to talk to AI with really good communication ingredients.
[00:17:28] Jeremy Utley: Is that? Um, I love it. To me, maybe because of I'm a teacher, I immediately go to, so what's the assignment, so to speak. And in my, mind, I'm imagining something like a 30 day challenge where, I mean, I have a sticky on my desk.
You know, have you tried chat GPT, which is some version of how can I help me do it? Right. And it's been there forever. I would actually say part of my problem is I don't look at it enough. It's kind of out of my line of sight. And I wonder whether something like a 30 day challenge where you say, I could imagine a scenario where you say every day for 30 days, , you put a 15 minute block on your calendar saying, have you tried AI at, you know, important 20 percent thing, right?
But it's 15 minutes a day and you just commit to this conversation every day. , what are ways to, um, reinforce. That instinct, because I'm not, convinced that a posted on the computer is sufficient for people who are busy.
[00:18:20] Geoff Woods: Well, and here's where, um, I have made a conscious choice that that was not a problem I wanted to solve in my business.
I felt like there were so many people that are trying to teach people how to prompt, trying to get you proficient with using chat GPT. That's like, that's not differentiated. And core to everything I do is showing companies how to accelerate growth and build a competitive advantage. So I kind of have to practice what I teach.
And I said, I'm not, I'm not for the people. I can absolutely show you how to go from zero to one using this at a very strategic level as a leader. But I'm not the one that's going to have to motivate you to do this. I'm not for those people. If you need motivation, I'm not your guy. I'm for the guy, I'm for the people that are running into walls trying to use this to accelerate growth and build a competitive advantage.
I'm just taking the wall out of the way so they can go faster.
[00:19:14] Henrik Werdelin: Have you found, uh, have you found it to be useful to use the voice version or the text
[00:19:18] Geoff Woods: version? Both. And it depends on the use case. So I, I have assembled many AI boards now and it started with a client that had a hostile board and the CEO was told, you've got six months to turn this around or you're all gone.
And he came to me, he's like, Jeff, I literally don't know what to do. And so I sat his executive team down in front of chat GPT in this instance, turned it into an HR professional with deep expertise in creating personality profiles and how to conduct an interview of the executive team. About one of the directors named Susan and it just asked one question at a time until it knew Susan well enough that it spat out a personality profile and I looked at them and said, you're the thought leader.
That's your thought partner. What do you think? What do you like about it? What don't you like about it? Tune it. They did it. We saved it. We repeated that process for every board member. I then created a custom AI model on the spot, fed it the personality profiles of every board member, and told it your job is to be our AI board.
We're gonna upload our deck to you before the board meeting, and as each board member, you'll review every page, and give us feedback on where there's landmines that we've missed that will cause the bloodshed in the next meeting. And we took a past deck that we already knew the answers of, and fed it to it to test it.
And it literally said on slide eight, Susan's going to get distracted by the granularity of the details on the slide, which is going to lead to a 30 minute detour and blow up your agenda. Instead of all the details, save these three points because it's what she cares about most. And I looked at the CEO and he goes, holy crap, that's exactly what happened.
So we use it to prep for the meetings. We take the transcript from the real meeting and feed it back to the model and have it improve itself based on reality. Within six months, the board came back and said, these are the best means we've ever had in the best decks we've ever seen. Great job.
[00:21:05] Henrik Werdelin: I think the CEO of section one at the same experience.
He also Ran, uh, the Dex through that, correct? Yeah. We had on the podcast and he, he did the same thing. He ran the decks through it.
[00:21:17] Geoff Woods: Oh, yeah.
[00:21:17] Henrik Werdelin: And then same experience. And so it seemed to be such a good piece of advice,
[00:21:22] Jeremy Utley: Not to dork out too much, but just one, specific question on that, because as he mentioned that's somewhat of a use case , that we're somewhat familiar with.
Um, , which I love. It's one of my favorite things. You said take the transcript from the actual meeting and improve the board, . Can you talk more about that in part? Yeah, sure. And I'm a huge believer in iterating A GPT and refining A GPT. How do you use the transcript to actually make the GPT better for the next quarterly meeting or
[00:21:49] Geoff Woods: So we pull it back in and we say, compare what you simulated would happen. Versus what actually happened. What were some of the things that happened that were missed? And this is where we start to guide it. And we ask questions. It was like, well, you know, we didn't predict that Susan was going to ask this question or that Jake was going to ask this question.
Well, that's actually a question that Jake asks a lot of the time. Check the personality profile for Jake. What can we do to enhance the personality profile so we can account for the fact that his mind works this way? Great update the personality profile.
[00:22:25] Jeremy Utley: these just documents that are fed into the GPT as knowledge documents?
Like the personality profile of each board member?
[00:22:30] Geoff Woods: So I've done this, and this goes back to your question about voice. I use voice often times if I'm driving and I want to talk to my board, or if I'm walking and I want to talk to my board. I fed AI, what I'm consciously aware are my strengths and weaknesses as a founder and a CEO.
And I asked it to conduct an interview of me to help me see things about myself that I may not be seeing that are strengths and weaknesses that I also need to account for. Then, based on that, if I were to assemble a real board in real life, what are the skills I would want to recruit to the table to help me achieve my goal?
And I walked it through the goal of my company over 10 years. And it called out lots of things and then I had it research famous people who exemplify those things. So like Steve Jobs is on my board for visionary thinking and product design. Tony Hsieh is on my board for customer experience. Jeff Bezos is on my board for operational scalability.
Warren Buffett's on my board for long term thinking and risk mitigation. But there are like, I'm also Steve Jobs. I'm going to tell you what Steve Jobs is not on my board for. Being a father and a leader. I specifically said, omit that. And I don't want his advice on that. And so I have conversations with my AI board on a weekly basis about the really challenging things that are occupying my mindshare, that I would love to have somebody who could just help me clarify my thinking and take the next step.
[00:23:55] Jeremy Utley: What's one insight that you've gotten from your AI board in the last week?
[00:24:01] Geoff Woods: Um, one of the things that it called out is that, um, it sees me as such a strong visionary and my goals are so insanely big that I can allow my ambition , to overshadow the risk.
And in my pursuit of global impact. I may be easily influenced by people who sell a big vision to me, but might be Um, all talk and no walk, which is totally true. And it was exactly because I'm literally having some very important conversations with specific people, about scaling this thing. And I needed that gut check.
[00:24:44] Jeremy Utley: Are there individuals or relationships that you've then kind of flag , and then how do you guard against being overly influenced?
[00:24:51] Geoff Woods: So I literally did this with, I'm considering a relationship with a person. Yeah. And I sat down with pen and paper to do my own thinking to start.
, I have a dedicated chair for strategic thinking and a playlist. I take this insanely seriously. Um, I did my thinking. I then took a picture of my notes, fed it to ChatGPT and then said, or fed it to the board and said, here was my thinking time session from this morning, recap what I wrote. And so I had it OCR my image and turn my handwriting into text so I could make sure it was working with the right data.
It was. And then I said, great, now let's take this to the next level. Act as my board. This is the thing I'm evaluating. I want you to ask up to five questions one at a time to really stress test my thinking. And then I want you to come back and tell me what you see as the top opportunities or the top positives and the top risks that I need to be aware of.
And based on this, what would your recommendation be on how I proceed and why? Now I already had my gut instinct of what I thought I should do, but I'm The whole reason the subtitle of my book is Harnessing AI to Make Faster, Smarter Decisions is the only way you make the best decisions is if you have the best information and data in front of you.
I'm clear of what I see, but man, I'm so optimistic that we're rowing in the right direction that I won't acknowledge that we're rowing toward the cliff. And so, I will need other people to challenge my thinking and help me see what I don't see to shine a light on my blind spots so I can make the best decisions.
[00:26:22] Henrik Werdelin: I like the driving and walking. We talked about like what are the triggers that could remind you to do this. I noticed when you were saying it that lately, when I walk my dog in the morning, I now use that as a session to contemplate, you know, like a big strategic problem.
And I have this conversation with Chatgpt about it. And so in a weird way, the walking with the dog becomes the trigger of making sure that I do this. Um, but I, I like this idea of like thinking about what could be every time that you do this, because obviously if you think about that you do it all the time, then you don't get it done.
But it was like. Well, I always have 20 minutes when I drive to this specific client, or I always have 20 minutes when I walk the dog, or when I wait for picking up the kids at baseball, whatever. Uh, that's kind of a fun way of thinking about it.
[00:27:08] Geoff Woods: So triggers is important if you study habit formation, uh, which I did when I was building the company based on the one thing.
You look at something in your life that already exists that you can build it on. And so B. J. Fogg, . You know, he told me he taught 10, 000 people to floss their teeth. And it's like the number of adults that do not floss their teeth every way.
That's actually a pretty remarkable thing. And how did he do it? He said, after I brush my teeth, I will floss one tooth because everybody had the habit of brushing their teeth. And if you can just add this onto something and one tooth that's so small, it's a tiny habit as he would call it, or James Clear would call it an atomic habit.
It's so small that you can do it. And if you wrap the floss around your fingers, are you really going to stop? And so strategic thinking. Was not always a habit for me. I used to be living in operational overwhelm. I was on the hamster wheel every day, running, running, running from meetings to emails to meetings, and then if somebody says, Hey, do you got a minute?
I'd say yes, because I was a team player, but at the end of every day, I knew I was busy, but a deep down, I questioned what I got done and I said, okay, I want to slow down and start moving from operational overwhelm towards strategic clarity, what is one thing I can do? And the answer was, I can have. Five minutes of thinking time.
I can sit down with a pen and paper for five minutes. And so I made a rule. Before I even open my computer, I have to grab my journal. That was the trigger. Before I can touch the computer, I have to grab my journal. Now if I grab my journal and put it back down, still a win. But if I pick up the journal, am I going to open it?
Yeah. And if I open, am I, am I going to grab the pen? Yeah. And if I write a question at the top of the paper, will I at least start journaling? Yeah. And now it's such a part of who I am that on the days when I don't have strategic thinking, like I can actually feel it in my body. And so using that for AI suddenly became a lot easier.
[00:29:09] Jeremy Utley: I think just on the, uh, habit formation, you know, geekery for one moment, I think the most important thing that even BJ fog misses all respect to BJ fog, uh, is the trigger has to be sensible , Like, for example, he gives the example of every time he goes to the bathroom, he does two push ups. And that sounds like, oh yeah, I go to the bathroom all the time, therefore I do push ups.
But it's like, do you really want to do push ups by the toilet? Like, do you want, like what? And I love that example because it's such a terrible example. It's like, you gotta be kidding me. You should not do push ups after you go to the bathroom. But my point is, the way that my dentist got me to floss my teeth is by telling me to put my floss in the shower.
Um, because I love the shower and if my floss is there, it's an excuse to stay longer in the shower. That's actually an elegant, beautiful behavior design. But it requires a knowledge of the person, right? So to me, your habit insofar as it works, it works in part obviously because of habit formation and triggers
but it also works because it's sensical. You probably put your notebook by your computer. You probably are sitting, like, there's a number of environmental factors there. I love your idea of, I'm walking the dog. I've already got my phone. I probably already have headphones.
Ordinarily. I'm going to podcast. You could make the trigger like, because rewards are an important part of the habit loop, right? The reward could be, I get to listen to a podcast, but , the behavior is before I listen to a podcast, I need to do a five minute strategic, you know, thought exercise with my AI board, right?
[00:30:42] Henrik Werdelin: Yeah.
[00:30:42] Jeremy Utley: But all that say. If folks want to actually make this practical and to me, I think it's a privilege and congratulations to you. If you're only working with people who are highly motivated, I am a deep cynic having taught millions of people over the last 20 years. Most people aren't highly motivated, and so it's a great disqualifying factor to only work with highly motivated people.
If Listeners are like most people, which is they're not very highly motivated. What BJ Fogg would say is ease is the primary lever you've got to play with and you want to make it as easy as possible. Right.
[00:31:15] Henrik Werdelin: Ability and motivation is those other two vectors.
[00:31:19] Jeremy Utley: Yeah. Ability. And the more the ability required, the less easy it is. Right. So it's kind of inversely proportional to ease. Yeah. Yeah.
[00:31:29] Henrik Werdelin: I was just bragging that I actually, for once, have read one of the books that you refer to. Which is like, uh, not very common.
[00:31:36] Geoff Woods: Or did you just look it up on AI? Oh dude,
[00:31:39] Henrik Werdelin: I just looked it up.
Like, how can you help me with sounding smarter and more academic when talking to Jeremy and Jeff?
[00:31:48] Geoff Woods: You could do that.
[00:31:49] Jeremy Utley: Yeah. You could have a GPT just running on the side of the riverside window here. That's right.
[00:31:53] Henrik Werdelin: Here's a random question from the site. At Bark the other day, , one of my colleagues, he interviewed a person and then the person completely nailed the interview, but the person interviewing was convinced that basically. , the person was like running everything through GPT, right?
Because the way that the person was looking kind of like to the side of the camera. And so then the question kind of like emerged in the management team was basically, Is this something that's just awesome because they make really good use of generative AI? Or is this a person that is so kind of like, I guess with your terminology is, is turning the GPT to the thought leader instead of the thought partner.
Uh, so therefore we shouldn't hire that person. And so, uh,
[00:32:34] Geoff Woods: I wouldn't make a decision just based on that, but I also think it's, what are you looking for in a person? I hear that and I go, okay. Here's somebody who's innovative. Here's somebody who's scrappy. Here's somebody who's using their resources to gain an edge.
Great. I can put that aside. Now I want to interview him for the core competencies of the job. Next interview is going to be in person.
[00:32:56] Henrik Werdelin: That's a good point. , uh, , we love case stories and you are so good at telling them, I mean, , . We'll take another one. If you have one
[00:33:05] Geoff Woods: shirt.
So this happened three weeks ago, I was facilitating a strategic offsite for a company. They do about 100. Not about they do. Last year they did 150 million in A. R. R. Software company and they missed their number and the board was not happy. Venture backed. And they said, You've got to grow top line 30 percent this year.
And the general consensus when we walked into the room was, there's just no frickin way. The numbers are just too big. And my job was to facilitate. And so, we had a conversation about it. I firmly believe that the purpose of a goal is actually not to achieve a result, but to inform who you can become.
It's a compass for your actions. , I want you to suspend your judgment of the goal, cause the goal was given to you. You can complain all you want, it's not gonna change. So if we want to hit 30 percent this year, we actually need a plan, not for 30 percent growth. We actually need a plan for 50 percent growth because plans don't go according to plan.
And the plan has to have sufficiency beyond the target so that when things go wrong, we still hit it. People lost their marbles, but I said, okay, just set your belief aside. What would have to be true? In order for us to hit 50 percent and somebody said, well, we'd have to drive at least 20 million in new revenue from products that don't even exist yet.
And I went, great. I wrote it on the board. And before you know it, they had a roadmap for what they would have to do to get there. It was just outside of their comfort zone. It was outside of the current trajectory. Fine. We got to work on each of the priorities and one of them, the 20 million one, these guys, they've got their sticky notes and their Sharpies.
I'm waiting for Jeremy to pop up to be like, let me help you with your design. Yeah. When it all added up best case scenario, it was only going to drive 12 million, which really means best case scenario. It's actually six because plans never go according to plan. And the CEO just looks at the room and he goes, guys,
You got anything else? And they're like, there's nothing. He goes, man, we might have to lower the goal. And I looked at him. I went, don't you dare. And you could have seen all the heads just snap around the room. Like, what's he going to do next? And I just pulled up chat GPT. I asked for permission to copy and paste their strategic plan into my team's account.
So it wouldn't train on the data. They said, that's fine. I pasted the strategic plan. I fed it tons of context about the current product offerings, the problem that's solving the competition, the competitive advantage we want to own in the future. And then I said, act as an aggressive growth minded board member with deep expertise in our industry.
Interview us by asking one question at a time, up to five questions to gain really deep context so you can then generate 10 non obvious product solutions or combinations that each one of them could drive 2 million in recognized revenue conservatively in the next calendar year. That was the prompt, and I hit enter.
And it started asking one question at a time. And the questions were so specific to the industry that the C suites going, Oh my gosh, this is just chat GPT. After five questions, it goes great. Here's your 10 non obvious product solutions or combinations. And I just watched the eyes in the room get huge.
Cause some of the stuff it recommended, they go, Oh my gosh. I can't believe we missed that. That wouldn't even be that hard to do. We could probably have that live in the next 30 days.
And something that was just reinforced for me is that it's really tough to read the label when you're inside the box. And every single one of us is in our own box, which is our perspective. And there are things that we think are impossible, which they might be based on your own human processing power.
It's why we've always valued relationships. Cause if you can get in front of a wiper with a thought partner, you share an idea and they go, well, yeah, that's a great idea. And what about this? And all of a sudden one plus one is equaling 11. When you do this with AI, with yourself as the thought leader and AI as the thought partner, it's a powerful synergy you can use to your advantage.
And you gotta start.
[00:37:23] Henrik Werdelin: I mean, like, I love that.
[00:37:24] Jeremy Utley: Drop the mic.
. You're amazing. You're amazing.
You're amazing. We can't wait to be of service to you. We can't wait for the audience to hear this stuff and hopefully just drive bajillions of dollars of revenue. That's
[00:37:35] Geoff Woods: right. Um, call my mom next and tell her what you just said.
[00:37:39] Jeremy Utley: I will. You got it.
Henrik Werdelin, what do you take away from that conversation?
[00:37:46] Henrik Werdelin: I mean, like I said, I was mentioned during the pod. I do think the idea of thinking about.
How do you get to be GPT to, uh, help you on the 20 percent problems, which often tend to be more little bit more aloof and, and kind of like less about tactics and more about, you know, thinking is something that I don't think a lot of people are using it for because, they might ask it in a simplistic way and then it gives you a simplistic answer.
So I think the framework of context, role, interview, and task is kind of like an interesting way of. Forcing it to be a little bit more sophisticated in , the way that it can act as a, thought partner. I think that's the main thing. I mean, like the second one is for people who would like to use it more, but I'm not using it as much as they would like it to, what is the habit stacking that they can do?
Like, is there a part of their day where they have idle time where they could put in. , this as an experiment, um, those two things seems to be good takeaways.
[00:38:54] Jeremy Utley: Yeah. I love that, uh, the habit stacking thing. I think finding your trigger for a different kind of conversation, your example of walking the dog, I think that's super cool.
The other thing that I really resonated with was the whole Pareto principle. It reminded me of our conversation with Ginny Nicholson and her saying, everybody's talking about efficiency and that's not the really special thing. I feel like Jeff kind of recast that statement, but in a business context, because in a business context, for sure, most of the people are looking for the 80 percent of things that don't actually drive the majority of the value.
And they're trying to make them 40 percent rather than 80 percent or whatever. And to just flip the script and say, no, no, no. What are the 20 percent of activities that actually drive results? , I liked his frame also of how can I be the thought leader and AI be the thought partner in this collaboration?
I thought that was a really helpful frame. And then the question to me is really just how do you remember to do that? As you know, I'm, I'm kind of obsessed with. Making it actionable. I think it relates to Steven Costlin's thing, right? From a recent episode where he talks about active learning.
It's one thing for people to passively hear this content and go, Oh wow, that's a really good idea. But to use the knowledge before it evaporates, otherwise it's basically a sugar high and you don't get any benefit from it. To me, That's my obsession is how do we codify it and codify the practice in a way where folks who are listening actually derive value and maybe it's as simple as a calendar, you know, events.
Maybe it's as simple as identifying that one time where you do have downtime and it's kind of, you know, Empty calorie time, you know, where you're listening to a podcast, maybe you stop this podcast and actually start doing something productive rather than listening. But I think there's something about that instinct. There's something about the reminder. Yeah, I know.
[00:40:38] Henrik Werdelin: Not productive time. Henrik,
[00:40:40] Jeremy Utley: you should have seen Henry. All
[00:40:42] Henrik Werdelin: of them. He's like, you can't stop
[00:40:44] Jeremy Utley: listening to this show. Do it after. But that's the problem, right? Is Folks want to, they want to have time for it rather than make time for it. And sometimes it means actually stop the pod.
[00:40:52] Henrik Werdelin: I mean, it's like putting something in the calendar a few weeks out. I think somebody else mentioned that as just, you know, a very. Obvious thing just put in the calendar. I like this idea of using it for the heavy, heavy problems. I do also like , to use AI to reduce the amount of time I use on like , the bread and butter stuff.
Like, do you subscribe to this idea that people should mainly use it to the 20%? Oh, is it kind of like, use it for a bit of both?
[00:41:23] Jeremy Utley: I think it's actually more nuanced than a choice. I think that most people aren't good partners to AI and they aren't good AI conversationalists and probably they can build their chops in the 80 percent stuff and it earns them the right to deal with the 20 percent stuff.
I
think that unless you have somebody like Jeff or you or me sitting beside you to actually be kind of a, a Sherpa of sorts to help you with the 20 percent high leverage, high strategic value. , task, , undertaking and thought partnership. Your better bet is to start with some of the 80 percent stuff and start to learn what does it take to get good output out of gen AI?
What does it take to get to a point where I feel like I'm having a really dynamic back and forth? So to me, it's not an either or it's actually about building skills and then using skills. Maybe you want to build skills in the less important, but you know, higher volume stuff. And then you want to use skills in the more important, , lower volume stuff.
That being said to the other question that you kind of asked underneath that question. Do I buy that? There's value there. I have seen in my interactions with management teams, some of the biggest value actually comes from, we've talked about this before, but role playing difficult conversations, right?
You've got a performance conversation , with a senior vice president or somebody needs to be terminated. That to me is a, 20 percent activity for sure. I have seen unequivocally leader after leader, after leader, after leader go. I am so much more prepared for a human, meaningful, intentional conversation that I would have been prior to using gen AI in that way. So I've for sure seen the benefits. of the strategic thought partnership. I think that probably that's not the place to start is my guess.
[00:43:10] Henrik Werdelin: Yeah. I mean, I, I guess the thing is to use it, where you can. Um, so I agree with that.
[00:43:17] Jeremy Utley: Until next time, folks. Give us your feedback. Let us know what you think. Leave a very honest five star review of this episode. A very honest,
[00:43:26] Henrik Werdelin: gloating kind of commentary.
[00:43:28] Jeremy Utley: Very honest, obsessively enthusiastic review. Thank you.