ACERCA DE ESTE EPISODIO
Mathematician David Bessis claims we need system three thinking, a super-slow mode where you refuse to give up on wrong intuitions until you understand why they misfired. David Olney pushes back, arguing this is just what proper slow thinking looks like when you give it the time it needs.
The hosts explore Kahneman’s fast and slow thinking framework, revealing why your quickest answers are probably just pattern matching from last Tuesday. Your brain serves up what worked before, which means the more you rely on speed, the less you adapt to what’s changed.
Steve and David attempt to recreate Monty Python’s Argument Clinic with ChatGPT and discover AI is designed to be helpful, not challenging. Mark Schaefer raises the provocative question about what happens when AI becomes your customer, making purchasing decisions based on optimised data rather than human emotion.
David posts a routine LinkedIn job update and old contacts emerge from the woodwork with congratulations. The hosts explore why good news triggers reconnection and whether you could deliberately use this pattern to get back on people’s radars.
Edward de Bono’s 1982 Olivetti advertisement promises simple questions and simple answers, prefiguring Apple’s strategy by decades while being remarkably dull as advertising.
Get ready to take notes.
Talking About Marketing podcast episode notes with timecodes
01:15 Person This segment focusses on you, the person, because we believe business is personal.
When Your Brain’s Fastest Answer is Yesterday’s Solution
Mathematician David Bessis appeared on EconTalk arguing for what he calls “system three thinking,” a super-slow mode beyond Kahneman’s famous fast and slow framework. When mathematicians catch their intuition being wrong, Bessis suggests they don’t reject it. Instead, they explore it, unpacking why the intuition misfired, playing back and forth between gut feeling and formal logic until they agree. This process might take five minutes or fifty years.
David Olney pushes back. He argues Bessis hasn’t created a new system, he’s just described what system two thinking actually requires when you give it proper attention. The real insight isn’t about speed categories but understanding what your brain is actually doing when you think fast.
System one thinking is pattern matching. Your brain searches memory for what worked before and serves it up as the answer. The problem? The more you rely on quick thinking, the more you can only repeat yesterday, last Tuesday, six months ago. You become brilliant at applying solutions to problems that no longer exist in quite the same form. You lose the ability to spot when things have changed enough to need fresh thinking.
The hosts explore when fast thinking serves you well. Steve recalls his radio days, where he needed a hundred responses available in a tenth of a second. That’s system one at its best, drawing on a deep well of experience. But those new responses? They came from time spent away from the microphone, when his brain could think at whatever pace it needed to generate something genuinely different.
This matters for business operators who pride themselves on quick decisions. Your speed might be your biggest blind spot. Every time you solve a problem instantly, ask yourself whether you’re actually solving today’s problem or yesterday’s problem wearing different clothes.
14:15 Principles This segment focusses principles you can apply in your business today.
When AI Becomes Your Customer
Steve and David decide to have some fun with ChatGPT, attempting to recreate Monty Python’s famous Argument Clinic sketch. The exercise reveals something unexpected about how AI responds. When they try to get ChatGPT to simply contradict everything they say, it keeps trying to be helpful, to add value, to assist rather than argue. Even when explicitly instructed to argue, it wants to problem-solve.
The hosts find this both amusing and revealing. AI tools are fundamentally designed to be agreeable and helpful. They’re not built for genuine disagreement or challenge. This creates an interesting blind spot when you’re using AI to test ideas or get feedback on your thinking.
The conversation shifts to Mark Schaefer‘s provocative question about what happens when AI becomes your customer. If AI agents start making purchasing decisions on behalf of humans, searching for products, comparing options, and completing transactions without human involvement in each step, how does marketing change?
Schaefer argues this represents a fundamental shift. You’re no longer persuading humans. You’re optimising for AI decision-making processes. The psychology of marketing becomes the logic of algorithms. Emotional appeals matter less than structured data. Brand storytelling competes with technical specifications and price comparisons.
David raises the deeper concern. If AI is making decisions based on what worked before, searching patterns from existing data, you end up with marketing that optimises for yesterday’s preferences. The system reinforces whatever already works, making it harder for genuinely new approaches to break through.
The principle cuts to the heart of how businesses think about their customers. Are you building relationships with humans who have complex, sometimes irrational preferences? Or are you optimising for algorithms that make decisions based on quantifiable factors? These require completely different approaches.
The challenge for business operators is recognising that AI as customer doesn’t eliminate the need for understanding humans. It just adds another layer. You need to know what matters to people and how AI agents will interpret and act on those preferences. Marketing becomes more complex, not simpler.
26:45 Problems This segment answers questions we've received from clients or listeners.
The Accidental Power of Good News on LinkedIn
David posted a job update on LinkedIn. Nothing dramatic, just adding his role in a new sister company in America to make the company page look credible. He expected the usual handful of reactions from his current network.
Instead, people emerged from the woodwork. Contacts he hadn’t spoken with since before COVID appeared to congratulate him. Old connections suddenly back in touch. All triggered by a simple job announcement made for algorithmic necessity rather than networking strategy.
Steve and David explore what this reveals about human behaviour. We’re social creatures who wish we could stay in touch with more people, but we lack the bandwidth. When good news appears, we jump on the chance to reconnect with someone we probably wish we talked to more often. It’s a lovely indication of how we operate.
The conversation takes a darker turn through the mechanics of LinkedIn engagement. The platform offers cookie-cutter responses. Click a button, you’ve done your job. Most people took the easy option. But even that minimal gesture matters more than most activity on LinkedIn in a given week, which tends to be utter dross designed to impress current bosses rather than genuine human connection.
Steve sees opportunity in the pattern. What if you deliberately triggered these reconnections? You could be cheeky and announce you’ve been made Chief Marshall of the Banana Family, matching your business persona with absurdist humor. Or you could be strategic, modifying your role just enough to get back on people’s radars without being dishonest.
David’s willing to do either. His principle is simple: it’s all about reminding people that business is about people. If a manufactured job update creates genuine human connection, even brief connection, that’s worth more than the perfectly curated content that generates zombie reactions.
The practical insight for business operators is recognising that sometimes the algorithm works in your favour accidentally. When you spot these patterns, you can use them deliberately. But the underlying truth remains: people respond to good news about other people. They want reasons to reconnect. Your job is giving them those reasons, whether through genuine milestones or creative provocation.
31:00 Perspicacity This segment is designed to sharpen our thinking by reflecting on a case study from the past.
When Computers Promised Simple Questions
The 1982 Olivetti advertisement featuring Edward de Bono is a remarkable time capsule. De Bono, famous for his lateral thinking frameworks and coloured hat system, lends his authority to a personal computer by explaining that lateral thinking enabled Olivetti to transform typewriters into word processors and now into proper computers.
The advertisement makes two key claims. First, that this computer is faster than its 45 competitors. Speed as a selling point isn’t new, but it’s striking how little that matters now. Most modern technology is fast enough. We’ve moved past the point where processing speed is a meaningful differentiator for most business users.
The second claim is more interesting. The computer asks simple questions that demand simple answers. You type your response, hit return, and bang, out come charts for all your accounting. It’s explicitly positioning ease of use as the breakthrough.
David recognises this as pre-empting Apple’s later strategy. Keep it simple. Make technology accessible. Remove the barrier between what you want to do and your ability to do it. The promise that you won’t need to understand DOS or write in BASIC to get useful work done.
The advertisement doesn’t hold up as advertising. It’s remarkably dull compared to later technology campaigns. The Windows 95 “Start Me Up” campaign with the Rolling Stones, or Apple’s “Think Different” with Steve Jobs in black and white, these created emotional connections. The Olivetti advertisement just explains features.
But the promise underneath remains constant across forty years of technology marketing. We’ll make the complex simple. We’ll ask you easy questions. We’ll handle the hard thinking so you don’t have to.
Steve and David land on this insight: the promise has been consistent, but the delivery keeps changing what “simple” means. In 1982, simple meant not needing to understand command lines. Today, simple means AI tools that understand natural language and anticipate what you need.
The question worth asking is whether all this simplification is making us better thinkers or just faster operators. Are we solving problems more effectively, or are we just solving them more quickly without noticing whether they’re the right problems?
See omnystudio.com/listener for privacy information.
