Previously on Giuseppe’s Glimpse: In the last issue we discussed the difference between retrieval and understanding, and why handing AI the first one completely can come at the cost of the second. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I spend a fair amount of time going through trend data for client work, mostly the kind of thing that’s useful but forgettable, numbers you glance at and move past.

This week one thing in the Nextatlas feed, the AI-powered trend platform that tracks early adopters and weak signals across digital culture, made me stop scrolling. It named something I’d been noticing anecdotally for a while without quite putting words for it.

Belonging is being decoupled from ownership.

Buying a home has moved from difficult to completely out of reach for a large slice of the people who would have bought one a generation ago.

I’m not going to bore you with yet another rehash of the affordability numbers, the widening gap between what a place costs and what a salary covers is something we’ve all heard plenty about by now.

What I found more interesting is what people are doing instead of sitting around waiting for that gap to close. Rather than holding out for the same milestone their parents hit, pausing their sense of home until the math works out, people are building it now, with whatever they’ve got.

Anchoring somewhere else

People are attaching identity to a favorite café, an adopted city, a stretch of landscape, instead of to property they hold. Nextatlas tracked this conversation over time and it’s climbed steadily, dipped for a stretch in the middle, then picked back up.

It’s gone up 38% over the past twelve months. Millennials are driving most of it, with Gen Z not far behind, which tracks given who’s actually been priced out the longest.

For a long time the working assumption, mine included, was that building identity around a place instead of a property was a placeholder, something you did while your savings account tried to catch up to the market. You’d delay the real sense of home until the mortgage cleared, and everything before that counted as temporary.

I think what’s actually happening is bigger than people giving up on that particular wait. The goal itself is shifting. Owning a home was the milestone that used to define whether you’d made it, and for a lot of people that milestone just isn’t on the table the way it was for their parents.

Instead of chasing the same marker under worse conditions, people are deciding what actually matters to them and building toward that instead. It’s less a consolation prize and more just a different definition of having arrived somewhere.

Mistaking motion for rootlessness

There’s an easy way to misread this if you’re a brand. You could look at the data and conclude people don’t care about roots anymore, then go build a nomad-core campaign around it, something about freedom and no fixed address.

That would miss what’s actually happening. What people are chasing isn’t novelty, it’s a sense of belonging, somewhere that tells them who they are the way a home used to. A café where the person behind the counter knows your order, or a city that isn’t home on paper but still feels like it.

That’s still roots, just not planted where the paperwork says they should be.

For a lot of people, home ownership used to be that solid thing. Now it looks more like it might be other people, the ones who show up at the same place for the same reasons, which changes what these spaces are actually being asked to provide.

How the market is changing

Co-living is probably the clearest early signal of this shift.

What used to be pitched almost entirely to students and digital nomads passing through a city for a few months is now expanding toward people who could technically afford a mortgage eventually, and are choosing something flexible and community-based instead.

One estimate puts the global co-living market at roughly $7.8 billion in 2024, on track to more than double by 2030.

Hospitality groups, F&B brands, and cultural spaces are moving in a similar direction.

Second Home, a workspace brand that shows up strongly in the Nextatlas data, is a good example. The base pitch for a coworking space is a desk, an ergonomic chair, a screen waiting for you every morning. That’s not really what they’re selling.

Their offer is the feeling of having somewhere that’s yours in a structural sense. Somewhere you go back to and get recognized at, somewhere you belong, without ownership really playing a role.

Where the emotional equity actually sits

It’s clear that what people are after in these spaces goes past the physical place itself.

A lot of the certainty people used to organize a life around has gotten shakier: owning a home, a stable job, a family that follows a fairly predictable timeline.

At the same time, everyday life has gotten more individualistic, partly through how social media reorganizes who we actually spend time with and how little of it happens in person.

Put those two things together and what people seem to want from a café or a co-living space or a neighborhood isn’t just a nice place to be. It’s other people who’ve had the same experience and recognize it in each other, a kind of belonging that doesn’t need a deed, a job title, or a family structure to feel legitimate.

And belonging to something, being recognized by it, ends up saying something about who you are. That’s what people might actually be after.

No deed required

I think most people have a version of this. A café, a gym, a stretch of coastline you drive to every summer without fail. You don’t go because you own any part of it (or maybe you do, but that’s a different story), but because it’s tied up with how you think of yourself.

Whatever yours is, it probably has less to do with the amenities than with a kind of belonging that crept in over time, the sense that this place has some claim on who you are, and you have some claim on it.

That’s hard to build on purpose. Maybe that’s why so many attempts at it come off hollow.

The ones that work aren’t manufacturing anything. They’re just there, consistently, long enough for that sense of belonging to settle in on its own.

Stay curious 🙌

gs.

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe’s Glimpse: In the last episode, we explored what happens when a company hands a decision to AI for so long that nobody remembers how to make it without help. Missed it? Catch up here! ✨

Buongiorno everyone 👋

Do you remember the last time you got lost in a library? Not looking for a specific book, just wandering around the shelves led by the titles.

I don’t think I remember mine. It’s been a while since I searched for something without knowing, more or less exactly, what I’d find.

That’s not a complaint about libraries closing or habits changing. It’s something narrower. Something that, once again, I started thinking about while noticing how I use AI.

I think by now most of us share the experience of asking an AI tool a question to get a quick answer instead of looking for it ourselves. Knowledge is starting to feel quick and targeted by default, almost regardless of the subject.

If I think back to the way I used to learn, in school, at university, later on my own, it looked a lot different from this.

The efficiency trade

AI-powered search is remarkably good at one thing: getting you to an answer. Ask it something and it synthesizes a handful of sources into a clean paragraph, usually in a few seconds.

The old way made you do that synthesizing yourself. On paper, it meant pulling three or four books off a shelf, checking their indexes, reading the relevant chapter in each, and noticing when two authors disagreed with each other. Online, it meant something faster but not fundamentally different: open ten links, read the ones that looked credible, run into two sources that contradicted each other.

Either way, you had to decide who to trust, or hold conflicting information in your head until a third source broke the tie.

That decision, repeated enough times across enough subjects, is roughly how you built a working model of something rather than just an answer about it.

AI search skips straight to the tie already broken. That’s the upgrade that cuts down time and effort, because that’s the part that used to be the work.

What gets lost along the way

There’s a term researchers in information science have used since the 1990s for this: information encountering, coined by Professor Sanda Erdelez to describe the accidental discovery of something useful while you’re looking for something else entirely.

It’s the paragraph you weren’t searching for that ends up reorganizing how you think about the paragraph you were, or the source that contradicts the one you just read, forcing you to sit with the disagreement instead of accepting the first answer.

Traditional search was full of friction, and the friction is where a lot of this happened. You had to compare sources because nothing handed you a synthesis. You noticed when two experts disagreed because you were reading both of them instead of a summary that had already reconciled them on your behalf.

Researchers who study how journalists come up with story ideas have found they rarely describe a deliberate method for it. There’s no query that reliably produces a good one. What they describe instead is putting themselves in places and conversations where something interesting tends to turn up, without knowing in advance what it will be. That’s the opposite of asking a targeted question and getting a targeted answer back.

I’m not trying to be nostalgic about how things used to be better. These are simply facts describing how certain kinds of insights are more likely to come about.

AI search resolves the disagreement before you see it. That’s the value proposition and also the cost, since the resolving used to be what turned an answer into an actual mental model of the subject.

Where this appears most

I notice this most in my own work, usually when I’m under time pressure, which is exactly when the shortcut is hardest to resist.

If I ask an AI tool to summarize a market, I get the consensus view: the competitors everyone already tracks, the trends already in every report on the topic. It’s accurate, and it would take me an hour to assemble the same thing myself.

What I don’t get is the layer underneath it, like the odd regulatory filing nobody’s flagged yet or a line in an earnings call transcript that doesn’t quite fit the narrative. Those things used to surface because you had to read fifty documents to write a two-page brief, and got pulled off track by document thirty-one somewhere along the way. At the time that felt like wasted effort, but looking back, it was often where the real value came from.

Hiring is another good example. A colleague told me she now uses AI to shortlist candidates against a job description, which is efficient and mostly fair to everyone in the pile.

What she says she misses though is the resume that didn’t fit the brief at all, the one that used to stop her and make her see the position in a different way.

So the question is not that the AI’s answer is wrong. The issue is that it’s way thinner than the version you’d have by getting there yourself, and the thinness is easy to miss because the answer looks complete.

What do we keep?

I don’t think the answer is to avoid AI search, and I’d be suspicious of anyone telling you to go back to typing ten queries by hand out of principle. I use these tools daily and I’m faster because of them.

What I’ve started doing instead is paying attention to why I’m reaching for AI in a given moment. Sometimes it really is to save time, other times it’s just to skip past the discomfort of not knowing something yet.

While they feel the same in the moment, these are two very different things, and knowing which one you’re doing is most of what separates using AI well from using it on autopilot.

The practical version of this, for me, has been leaving a few pockets of the work deliberately unautomated. I’ll read a report nobody asked me to read, or follow a citation that has nothing to do with what I’m working on, or just let a piece of research take longer than it needs to.

Sure, this might not be efficient, but that’s sort of the point. It gives me room to engage with information differently, to notice things I wasn’t looking for, and every so often one of them turns into something I wouldn’t have gotten to any other way.

Designing for it on purpose

Obviously it’s hard to scale this smoothly into a policy. A company can’t mandate serendipity, and a rule that says “spend twenty percent of research time browsing” would probably just get gamed or ignored.

What companies should pay attention to though is which tasks get treated as pure retrieval and which ones get treated as understanding in progress.

Retrieval is fine to hand over completely. Understanding needs the chance for a detour left in, at least occasionally, or the person doing the work ends up informed about a great many things but truly familiar with very few of them.

That’s true for a company deciding what to automate, and it’s true for me deciding whether to just ask AI or dig in myself on any given afternoon. Same question, smaller scale, but you have to start somewhere.

Stay curious 🙌

gs.

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe's Glimpse: In the last episode, we looked at why human intelligence multiplies rather than adds, and why outcomes matter more than outputs. Missed it? Catch up here! ✨

Buongiorno everyone 👋

Do you check the route Google Maps gives you? I used to. I grew up knowing my city well enough to reroute around traffic on instinct.

At some point I stopped bothering, because the app is usually right and arguing with it felt like wasted effort. I still get where I’m going, I just couldn’t tell you anymore how I’d do it without the app.

I bring this up because I’m seeing the same pattern show up inside companies, at a scale that probably matters more than my commute.

How much is too much

“Human in the loop“ gets repeated in almost every AI conversation I have lately, as if saying it settles anything, when really it doesn’t settle much on its own.

The harder question, the one most organizations skip straight past, is how much in the loop, when, and for which decisions.

A compliance review and a routine marketing email don’t need the same level of human involvement, but companies often apply one policy to both anyway, usually because writing one policy is easier than thinking through twenty edge cases.

Get that calibration wrong, and you pay for it either way, just in different currencies depending on which direction you err in.

Too involved, and everything slows down: teams reviewing every output, asking for a human sign-off on decisions the system was already handling well, until the efficiency gain that justified bringing in the AI in the first place basically disappears.

You end up paying for the technology twice, once to build it and once for the extra layer of people now needed to babysit it.

That’s the too-involved failure mode, and it’s the more visible one, since at least someone usually notices the slowdown and complains about it.

What under-involvement costs

Go too far the other way and the failure looks different. It’s an erosion, and erosions are hard to notice until they matter. Hand a team’s judgment over to a system for long enough, and the skill stops getting exercised without anyone noticing, so by the time someone needs it, it’s gone.

A merchandising team that stops questioning its demand forecasts can look fine for months, right up until it approves an order that turns out badly and nobody can explain why, because nobody kept the habit of reasoning about demand from first principles.

Customer service teams run into something similar when AI drafts every response: the rep who’s fine on the easy 90 percent starts to freeze on the case that falls outside the script, right when a customer most needs someone who can actually think.

Under-involvement carries a couple of other costs, separate from skill erosion entirely.

There’s a security angle, one I hadn’t thought much about until a CISO mentioned it to me: a system that always reacts to the same input in the same way is, by definition, predictable, and predictable systems are easier for someone outside the company to reverse-engineer and exploit.

There’s also a business model risk, since handing your differentiation entirely to an algorithm can flatten what made you distinct to customers in the first place. In categories where consistency matters more than distinctiveness, that’s fine. In others, it turns a premium product into a commodity one.

What companies are trying

Some of the organizations I talk to are experimenting with countermeasures, and I find these attempts telling even when they’re imperfect.

One approach: deliberately introduce small, catchable errors into a system so people stay in the habit of checking rather than rubber-stamping. Another: build in moments where the system simply doesn’t respond, so a person has to make the call instead of defaulting to whatever it would have suggested.

Neither of these is a permanent solution, and I imagine they’re a little awkward to defend in a steering committee, but they’re an honest admission that unsupervised trust in a system has a cost somewhere down the line, even when the system is performing well on every metric you’re tracking.

What we might gain instead

Even with those risks, I don’t want to overcorrect into fear. When something new shows up, my instinct is to get curious about what it changes rather than treat it as the enemy, and I think that instinct is worth trusting here too.

So it’s worth asking what we might actually gain, not just what we stand to lose.

If a routine task gets taken off someone’s plate, you could end up with an idle mind that drifts, or you could end up with a mind that finally has room for something it never had time to notice before.

A planner who no longer builds the weekly schedule by hand might spend that hour on the phone with a client instead, which is usually where the useful information was hiding anyway.

I don’t know that this always plays out well, and I’d be wary of anyone who tells you it does. It depends entirely on what actually fills the space that opens up, and most companies have spent far more energy figuring out what to automate than figuring out what to do with the time that gets freed.

That’s probably the harder question, and it’s getting a fraction of the attention it deserves.

Designing for it on purpose

This isn’t a problem with one fixed solution, and again I’d be suspicious of anyone offering a tidy framework that claims otherwise, since the right amount of involvement depends on the task and on what’s at stake if it goes wrong.

That’s a judgment call every team has to make for itself, probably more than once, as the technology and the context both keep shifting.

What it does require isn’t a formula but a deliberate choice: designing for it on purpose, instead of defaulting to full automation because it’s available, or to full oversight because it feels safer and easier to justify to a board.

In practice, that means people from different functions actually comparing notes on where they’ve drawn the line and why, since no single team sees the whole picture of where it should sit, and what looks like the right calibration in one part of the business can be exactly wrong in another.

If you’re working through this in your own organization, I’d like to hear how you’re drawing that line.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe’s Glimpse: In the last episode, we explored the leadership superpower of thriving in paradox. Missed it? Catch up here! ✨

Buongiorno everyone 👋

We spend a lot of time these days talking about what makes artificial intelligence powerful. How fast it learns, how it scales, how it can see patterns humans miss. How it’s going to transform work, creativity, decision-making, everything.

Maybe we should spend more time asking a different question: what makes human intelligence irreplaceable?

I’ve been working with some simple equations that help me think about this, and it’s helped me reframe the whole AI conversation.

Two different kinds of intelligence

Artificial intelligence looks like this:

DATA + ALGORITHMS = OUTPUT

You feed a system information, run it through a process, get a result. Better data and a better algorithm means a better result. It’s straightforward. And it scales.

Human intelligence works differently:

(KNOWLEDGE + EXPERIENCE) × (JUDGEMENT + CONSCIOUSNESS) = OUTCOME

Notice two things. First, there’s a deliberate distinction between output and outcome. An output is generated. It’s a response. An outcome is what actually happens in the real world as a consequence of that response. They’re not the same thing.

Second, the multiplication sign. Human intelligence isn’t additive. Knowledge and experience amplify each other, but what actually matters is how judgement and consciousness amplify everything else.

The problem with how we measure intelligence

Here’s where organizations start getting confused: they’re trying to measure human intelligence using AI metrics.

→ They compare the radiologist to the diagnostic algorithm on speed and accuracy. The algorithm wins.
→ They compare the hiring manager to the screening algorithm on consistency and bias reduction. The algorithm wins.
→ They compare the strategist to the predictive model on forecast accuracy. The algorithm wins.

But they’re measuring the wrong things entirely. It’s like asking a violinist to compete with a metronome on precision and consistency. The metronome wins easily, but that’s not what a violinist is for.

You can measure what an algorithm does: accuracy, speed, consistency, scale. These are clean metrics. But you can’t measure human judgment using those same metrics because judgment operates on a completely different scale.

A good decision under uncertainty might look messy on paper. It might take longer. It might fail in ways that look inefficient. But the outcome is what matters, and outcomes are shaped by things algorithms can’t touch: responsibility, moral reasoning, the willingness to stand behind a decision when it’s unpopular.

Where the algorithm fails

An AI system can generate an answer that looks flawless on paper: well-reasoned and supported by data, always internally consistent. That’s an output.

But then you implement it in the real world and something goes wrong.

A lot of companies today are working with algorithms to optimize hiring. The output is elegant. It supposedly eliminates bias from resume screening, processes candidates faster, recommends profiles that match historical patterns of success. By every objective measure, it looks better than human selection.

However, the result is: you risk hiring people who look good on paper but struggle when things get ambiguous, who can’t navigate actual complexity, who have no idea why the decisions they make matter beyond executing a task.

The output is clean. The outcome is organizational dysfunction.

That’s because the algorithm optimizes for pattern-matching. It’s looking at what succeeded before and finding more of that. It doesn’t understand what actually makes someone valuable in a human organization, which is something much messier and more contextual than any dataset can capture.

But what is judgement?

When I talk about judgment in that equation, I’m not talking about analytical reasoning. AI does analytical reasoning better than we do. I mean the ability to make a decision when you’re not certain, guided by what you actually care about and what you believe is right.

It’s the radiologist who doesn’t just look at the image but thinks about the patient’s age, what they can handle psychologically, what their life circumstances are, and then recommends a treatment that makes sense for them as a person, not just as a case.

It’s a leader who sees data pointing one way but notices people are scared and decides to communicate differently than the data alone would suggest.

It’s an architect who knows all the regulations but also understands how a building will feel to someone walking into it.

That’s fundamentally different from what an algorithm does. It’s the gap between knowing an answer and understanding what an answer means.

What actually needs to happen

AI will keep getting better at what it does. It will process more data, generate better outputs, find patterns we can’t see.

The real question is whether we invest in the capacities that make us fundamentally different. Whether we develop the ability to make decisions under uncertainty, guided by responsibility and moral reasoning. Whether we keep asking not just “what can we do?” but “what should we do?”

Because leadership, strategy, creativity, decision-making have never been about having better answers. They’ve always been about what those answers mean, which ones matter, whether you’re willing to stand behind them when things get hard.

An algorithm tells you what happened. You have to decide why it matters.

An algorithm can give you options. You have to choose one and live with it.

The choice ahead

Maybe the most important thing happening right now with AI isn’t the technology itself. It’s the choice we’re making, mostly without thinking about it, about what we want to cultivate in ourselves.

We could treat AI as a way to get through more tasks faster, then just fill the time we save with even more work. Or we could see it differently: as a signal that it’s time to invest in what only we can do. Making judgments. Taking responsibility. Creating meaning from information.

The technology doesn’t choose for us. We do.

Stay curious 🙌

-gs

PS: Thanks to Renato Geremicca for the thinking that sparked this 🙏

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe's Glimpse: In the last episode, we explored the difference between a job and work, and why today that distinction matters more than ever. Missed it? Catch up here! ✨

Buongiorno everyone 👋

The ancient Greeks had a story about wisdom that I’ve always found oddly comforting. Harmonia was the goddess of harmony, and she was the daughter of Aphrodite and Ares. (If Greek mythology isn’t your thing: Aphrodite was the goddess of love, Ares the god of war.)

So harmony emerged from love and war. Beauty and conflict. Compassion and strength.

On the surface, it’s a pretty odd combination. A paradox, even. You’d think the offspring of beauty and conflict would be confused at best. But the Greeks believed something different. They believed harmony emerged precisely because of that tension, not in spite of it. Harmony was the ability to create something coherent from opposing forces.

While watching how organizations handle the contradictions they face today, this story often pops up in my mind.

The contradictions leaders are supposed to hold

A few years ago, in a TEDx talk, I argued that the real superpower of modern leaders isn’t creativity, analytical thinking, or even technological fluency.

Rather, it’s the ability to live comfortably with contradictions, to hold opposing ideas simultaneously without collapsing them into one or the other.

Look at what’s expected of leaders right now:

→ data-driven and human-centric
→ efficient and innovative
→ global in scope but local in execution
→ consistent in brand but adaptable in approach

And now, with AI entering every aspect of business, there’s this additional layer: be both more technological and more human at the same time. More automated and more intentional. More efficient and more purposeful.

Most leaders I talk to experience these as problems to solve, things that need to be resolved once and for all. They feel compelled to choose: are we optimizing for speed or for meaning? Are we building for scale or for craft? Are we letting machines decide or keeping humans in control?

The pressure to pick a side is real and constant.

Why we default to choosing

The way we’ve learned to think about leadership values clarity and decisiveness above almost everything else. We praise leaders who have clear vision and make bold choices.

When you’re ambiguous or sit in the middle of competing demands, people read that as weakness or indecision, so it makes sense that leaders would want to resolve the tension and move forward.

But there’s research on organizational behavior, particularly work by scholars like Wendy Smith and Marianne Lewis, that points to something different entirely. They describe paradoxical leadership as the ability to embrace competing demands simultaneously rather than treating them as mutually exclusive.

Their key insight: we need to recognize that some of the most important tensions in organizations aren’t problems waiting to be solved. They might look like conflicting dynamics, but they’re dynamics you need to learn to manage.

What thriving in paradox looks like

The concept is interesting, but let me show you what this actually looks like in practice.

Imagine you’re the operations leader at a manufacturing company facing this paradox: you need to automate more for efficiency and cost, while also protecting the quality and pride your craftspeople have built over decades.

A typical approach would be to say: we’re investing in automation, retraining people for new roles, moving forward. Problem solved.

But imagine instead you asked a different question: what if automation handled the repetitive parts, but we protected the parts where craftsmanship actually mattered? You wouldn’t try to eliminate the tension. You’d get specific about where each side could contribute most.

So automation takes over the sanding and preparation work, the brutal, repetitive, back-breaking stuff nobody wanted to do anyway. But the finishing, the inspection, the quality decisions stay human. Your craftspeople aren’t displaced, but they’re freed from drudgery to do work that actually requires judgment and care.

Here’s what happens: the paradox doesn’t disappear. It becomes more visible, more woven into the actual process. The tension between machine and human is now explicit and working, and the organization thrives because of it. Efficiency improves. Quality improves. People feel like they’re doing work and not just executing tasks.

That’s what thriving in paradox means: you’re getting specific about where each side contributes most, and you’re making the tension productive instead of trying to eliminate it.

What failing in paradox looks like

I’ve watched leadership failures over the years, and most don’t come from choosing the wrong side of a paradox. They come from leaders who believe there only is one side, who see the tension as a distraction to be eliminated rather than a reality to be worked with.

An organization that tries to be fully data-driven often builds things nobody actually wants.

An organization that prioritizes only human connection over operational efficiency eventually collapses under its own weight.

A company that optimizes purely for AI efficiency creates environments where people feel disposable.

The problem in each case is the same: the leader decided to solve the paradox by eliminating one side.

The paradoxical leader holds both tensions consciously, and yes, it’s uncomfortable. But that discomfort is intentional: it’s what keeps you honest about trade-offs instead of pretending they don’t exist.

How to actually start thinking this way

The practical shift is simpler than it sounds, though it requires a different kind of thinking. Stop asking which side you should pick. Start asking where each one matters most.

Take the data versus human tension. The questions to begin with are: what does the data reveal about what people actually care about that we might miss through intuition alone? What patterns in the numbers tell us something about human behavior we wouldn’t see otherwise? How do we use quantitative insight to actually deepen our understanding of the people we serve?

Or consistency versus adaptability. Again, it’s not about which one wins. It’s about what is core to who we are, and where that core requires us to adapt. A luxury brand using the same marketing template in Tokyo and São Paulo isn’t consistent, it’s tone-deaf. Real consistency means showing up differently in each place while remaining true to what actually defines the brand. The consistency is in the principle, not the execution.

The shift is in moving from either/or questions to how-do-they-work-together questions. Instead of picking one side and following it blindly, you’re asking “what does each enable that the other can’t?” You’re creating conditions where both can operate at their strongest, revealing different truths about your situation.

That’s the superpower. And it’s learnable.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe’s Glimpse: In the last episode, we explored 5 and a half ways to constructively dissent. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I was on a call last week with a client who kept using the word “optimize“ to describe how they wanted to use AI. Optimize the workflow. Optimize the hiring process. Optimize the decision-making. And I kept thinking: optimize for what? Speed? Cost? Efficiency?

The question underneath that is: what are you actually optimizing?

Because if you’re trying to optimize something that’s just a transactional task, you optimize for speed and cost. But if you’re trying to optimize something that someone experiences as work, something bound up in how they see themselves, then optimizing purely for efficiency might be completely wrong. You’d be solving for the wrong thing entirely.

As Seth Godin has recently suggested, this all comes down to something we rarely talk about: the difference between a job and work.

A job is transactional. You do X, you get paid Y. It has a description, deliverables, measurable output. You can hand it off and make it more efficient.

Work is different. It’s what you do when the doing itself matters to you, when it’s connected to how you see yourself or what you care about.

A surgeon can perform a job: run through a procedure, execute it flawlessly. But the moment they genuinely care about the person on the table, when they’re making real judgments about trade-offs that no protocol covers, when they’re carrying actual responsibility, that’s work.

It’s the same with a teacher who notices something’s wrong with a student before anyone else does, or a craftsperson who cares about the quality of something no one will ever inspect.

Where it gets tricky

The complicated part is that these things aren’t always separate.

Sometimes a job becomes work through repetition and care. Sometimes something that starts as pure drudgery, like running reports, processing data, gradually accumulates meaning as you get better at it and understand what it actually means.

And of course this also works the other direction. Work can gradually become just a job. The thing that mattered starts feeling like routine, the passion fades and you’re left executing the task without the engagement.

And this brings us to the problem we face today. Right now, as organizations are figuring out what to automate, they’re not really distinguishing between these.

They’re asking: what can AI do faster and cheaper? And the answer to that question doesn’t really take into account whether what you’re automating is a job or work.

Why this is important

When you automate something, you’re removing more or less from people’s lives depending on what that something is.

Some things are just jobs. They’re repetitive, exhausting, something you endure. For those people, automation is a relief. It would be patronizing to pretend that everything that feels like drudgery is secretly noble.

Then there’s a middle ground. Those activities that start as routine and gradually accumulate meaning through mastery and engagement. You run reports and eventually understand what they mean. You answer customer questions and develop genuine care.

Again, the reverse happens too. Something you used to care about becomes mechanical. The engagement fades and it’s just the task. These in-between things are easy to overlook when you’re automating, but they matter because the meaning isn’t fixed, it depends on how you’re doing the work.

Finally, there are the things where doing it is part of who you are. A radiologist who spent years developing diagnostic intuition doesn’t just lose a job if an algorithm surpasses their technical skill. They lose something about how they understand themselves.

The problem is that the tendency today is to automate without asking which category something falls into. Lots of organizations just optimize for efficiency across all three without asking what they’re actually removing.

The blind spots in this argument

I need to be honest about the limitations of this thinking, though.

First, there’s a privilege embedded in it. The ability to distinguish between job and work, to care about meaning in your labor, is largely a luxury of people whose work already contains it.

For someone working multiple jobs to survive, the philosophical concern about losing meaning can sound like a complaint from people who’ve never had to choose between meaning and rent.

Second, meaning is fluid and deeply personal. Humans adapt quickly. We build profound identities around things that didn’t exist twenty years ago: software engineering, social media management, digital marketing.

What matters to one person might not matter to another. The work of tomorrow might look completely unrecognizable to us today. Which means you can’t just set universal rules about what’s meaningful and what’s not.

The answer changes depending on who’s doing the work and when. We might lose less and gain more than we think, or the opposite. The point is we can’t know in advance.

So what’s actually at stake?

The real risk isn’t that AI takes over human activity. We’ve been automating things for centuries.

The real risk is that we optimize for efficiency without ever asking what we’re optimizing for. Organizations that treat AI as purely a cost-reduction machine will end up with environments where people feel like machine supervisors rather than people doing something that matters.

Organizations that actually ask how AI can amplify human judgment and care rather than just replace it build something different.

At an individual level, it means asking yourself honestly: as more of what I do gets delegated or automated, what’s left that I would actually call my work? And am I investing in it, or just hoping it takes care of itself?

The thing about work that actually matters to people is it’s never just about the task. It’s about doing something you’re genuinely invested in, something you care about. That’s what’s at stake when we automate things away without thinking about which things those are.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe’s Glimpse: In the last episode, we explored the idea of anti-CEO and why the leader the future needs is the opposite of what we’ve been training for. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I recently found myself in a tense situation: misaligned with a business partner on something that mattered.

I prepared carefully: collected evidence, framed my points, wrote down my non-negotiables, and rehearsed the conversation in my head multiple times.

I’ve learned over the years that in these situations the risk isn’t really in pushing back, but in HOW you push back.

Get it wrong and you damage the relationship or lose credibility. Get it right and you can shift direction without destroying trust.

That experience gave me an opportunity to formalize five and a half principles worth going back to in these situations (spoiler: the half tip is that a pig with a tie is still a pig.) Hopefully they’ll help you too.

(1) Shift from “I disagree” to “I want this to succeed”

Leadership trust erodes when pushback looks like resistance, but it holds (or even strengthens) when it looks like commitment to the same goal.

The framing matters enormously. “I disagree with this approach” triggers a defensive response. “I’m fully aligned on where we’re going, and I think we should pay attention to something that could derail us” opens a conversation.

Lead with alignment, then introduce friction. You’re trying to protect progress, not blocking it.

Careful: I’m not talking about manipulation. It’s simply a matter of acknowledging a truth: in most cases, you probably do share the same ultimate goal. The disagreement is about the path to get there. Make that clear upfront.

(2) Bring data and opinions

Opinions alone can create hierarchy conflicts. The HIPPO effect is real: the Highest Paid Person’s Opinion tends to win, so when it’s just your view against someone else’s, rank becomes the tiebreaker.

Data alone, on the other side, can create methodology conflicts. Numbers can be arranged to say almost anything, and people argue about sample sizes, timeframes, and definitions. Ultimately, the substance can get lost in debates about measurement.

Data and opinion together create conversations.

If you can quantify the risk and then share your read on it, you move the discussion from personal to professional. “I think this is risky” is easy to dismiss. “Here’s what happened the last three times we tried something similar, and here’s why I think this time could follow the same pattern” is harder to wave away.

Authority is harder to invoke when facts are already on the table.

(3) Choose your moment deliberately

Timing is a strategic asset, and most people waste it.

Pushing back in a public setting forces leaders to defend their position to save face, and once that happens, a constructive resolution becomes virtually impossible. They’ve stated their view in front of others, so changing it now looks like weakness.

A private conversation, instead, gives everyone room to reconsider without losing credibility.

Same message. Completely different outcome.

This doesn’t mean you need to wait weeks for the “perfect moment.” It means recognizing that the middle of a board meeting is usually the worst possible time to surface a fundamental concern. Find 20 minutes before or after. The investment pays off.

(4) Name the tension before it names you

One of the most disarming things you can say is: “I want to flag something that might sound like resistance, but isn’t.”

It pre-empts the defensive reflex and signals intellectual honesty before the other person has a chance to misread you. It acknowledges that what you’re about to say could be interpreted negatively, and you’re choosing to say it anyway because it matters.

This works because it demonstrates self-awareness. You’re not oblivious to how your pushback might land. You’ve thought about it, and you’re doing it deliberately for good reasons.

Most conflict escalates because people mistake tone for intent. Naming the tension removes that ambiguity.

(5) Offer an alternative, not just a critique

Pushback without a constructive path forward reads as noise, but if it comes with an alternative it reads as leadership.

Before any difficult conversation, answer one question honestly: “What am I proposing instead?“

If you can’t answer that, you might not be ready for the conversation yet. Critique is easy, and almost anyone can point out problems. What separates valuable dissent from complaint is offering a better option.

This doesn’t mean your alternative has to be perfect or fully formed. It means you’ve thought beyond “this won’t work” to “here’s what might work better.” Even a rough alternative demonstrates you’re trying to solve the problem rather than win the argument.

The half principle: Know when to leave

Possibly the most important one.

If, after applying all five principles, the response is still “shut up, I’m the boss”... leave.

I mean it. As recognition that you’re in the wrong place, nothing more dramatic.

A leader who can’t welcome honest, well-intentioned dissent is a dictator with a good LinkedIn profile. And as an old saying in my home region Calabria goes: a pig with a tie is still a pig. 🐷

This isn’t about being precious or unwilling to take direction. It’s about recognizing a fundamental incompatibility. If you’ve done the work and the response is still authoritarian dismissal, the problem doesn’t lie in your approach but in the relationship.

Some people genuinely believe leadership means never being questioned. Unfortunately you can’t fix that. You can only decide whether you’re willing to work under those conditions.

Why this matters

Constructive dissent is one of the most valuable things you can offer an organization. It’s also one of the riskiest.

The organizations that succeed are the ones that figure out how to make dissent safe and productive. The individuals who succeed are the ones who figure out how to deliver it in ways that get heard rather than dismissed.

These five and a half principles won’t guarantee you get your way, but they will significantly increase the odds that your concerns get genuinely considered rather than shut down immediately.

And sometimes, that’s all you can ask for.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe’s Glimpse: In the last episode, we explored why AI is forcing professional services to rethink how they charge. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I wrote about something recently on Forbes México that made a few people uncomfortable: CEOs are, on average, more incompetent than ever before.

Before anyone accuses me of easy cynicism, let me explain. I don’t say this with malice. I say it because I think it’s true, and because understanding why matters more than the provocation itself.

I was thinking about this after reading a newsletter by my friend Seth Godin on the topic, which got me reflecting on how much has changed.

A century ago, Henry Ford could drive his own car. He understood, with reasonable precision, what it took to build one. The gap between what a business leader needed to understand and what they actually understood was manageable. The boss knew more than their subordinates about the things that mattered most, and that’s where their authority came from. 🚗

Now think about what a CEO is expected to have an informed opinion on in any ordinary work week: generative AI adoption, global supply chain fragility, cybersecurity posture, digital marketing ROI, talent retention in hybrid markets, geopolitical risk, ESG reporting, data privacy regulation, asset tokenization, DEI agenda.

That list applies equally to a twenty-person company and a Fortune 500.

Nobody can master all of that. It’s a structural impossibility.

Why the model is broken

Here’s what’s happening: the more complex business environments become, the wider the gap between what’s expected of the top leader and what any human can reasonably know.

I’m not saying individual CEOs are worse people or less prepared professionals. I’m saying the leadership model we’ve inherited (and still teach in most business schools) is obsolete by design.

The problem is the concept of CEO itself.

There’s real pressure that comes with the role. The pressure to appear like you know, to speak with more conviction than you have, to let instinct substitute for information there’s no time to gather.

I’ve done it, and I suspect most leaders have.

Academic research on this is consistent. The Dunning-Kruger effect has been widely documented. And there’s also a more dangerous variant in executive environments: institutional pressure toward false certainty. 💭

In a boardroom, well-founded doubt looks like weakness. Saying “I need more information” or “I’m not certain” gets read as indecisiveness. Leaders learn quickly that appearing confident matters more than being right.

The result is organizations that project certainty while flying blind. They make big investments based on instinct dressed up as strategy, and looking decisive becomes more important than being right.

The anti-CEO: a proposal

The term “anti-CEO” has been around for a while. What I want to focus on is what it means in practice, because I think we need more specificity about these competencies.

The anti-CEO isn’t a heroic figure. It’s a set of competencies that are often exactly opposite to what we’ve valued and rewarded for decades. It’s about knowing differently, leading differently, and extracting value from places where the traditional CEO doesn’t even look.

Let me walk through five inversions that define this approach.

(1) From certainty to comfort with ambiguity

The classic CEO was trained to give answers. Their days are structured around decisions, approvals, directions. The language is declarative: “We do this,” “We go there,” “We close this.” Ambiguity gets fought and eradicated.

What I’m seeing work differently: leaders who’ve developed what psychologist Elliott Jaques called tolerance for abstract uncertainty. The ability to hold multiple hypotheses open at once, to make reversible decisions while accumulating information, to update positions without seeing that change as failure.

When variables change faster than planning cycles, premature certainty creates more problems than well-calibrated doubt. 🎯

Satya Nadella is probably the clearest example. When he arrived at Microsoft in 2014, he found a culture that rewarded knowing over learning. Well, he replaced “know-it-all” with “learn-it-all”, and Microsoft’s market cap went from $300 billion to over $3 trillion in a decade.

(2) From vertical expertise to connector intelligence

The traditional executive built their career deepening in one domain. The knowledge pyramid was vertical: you climbed step by step on the same axis.

The anti-CEO is a domain connector. Someone who understands enough about disparate systems (technology, human behavior, regulation, culture) to detect the intersections where value gets created and where invisible risks lurk.

People who connect groups of knowledge that normally don’t communicate generate way more value for their organizations than the deepest experts in those same groups. The value is in the bridges, not the towers. 🌉

(3) From talking to listening with strategic intent

We associate executive authority with voice: the grand speech, the articulated vision, the board presentation.

The anti-CEO understands that the most valuable information for making good decisions rarely arrives from top to bottom. It arrives from the edges: the salesperson who hears customer objections, the engineer who knows what’s technically possible before the market does, or the customer service team that knows when a product is losing traction.

What works: designing deliberate systems to capture those weak signals before they become urgent alarms. Listening as strategic practice, and actually acting on what you hear, closing the trust loop that invites more people to speak honestly. 👂

Ray Dalio institutionalized this at Bridgewater with his “idea meritocracy“ system: any employee, regardless of rank, can (and must) openly question any decision.

(4) From perfect execution to imperfect experimentation

The 20th century rewarded efficiency. The CEO was the guardian: eliminating variance, standardizing processes, building control systems that guaranteed reproducibility.

Now competitive advantage comes from discovering the unknown before others do. And discovering requires experimenting. And experimenting involves failing.

The key is building organizational capacity to learn faster from inevitable errors than competitors. The anti-CEO designs systems that convert failure into actionable information before it gets expensive. 🔬

Amazon calls it “working backwards.” Apple calls it “1,000 ideas to get to one.” Same approach: exploring what’s possible rather than perfecting what’s known.

(5) From positional authority to earned authority

The classic CEO’s authority was largely positional: it derived from the role, the title, the org chart. People followed the CEO because the CEO was the CEO.

In high-complexity environments, that logic flips. The most talented professionals (especially younger ones) follow judgment. Rather than titles, they respect coherence between what gets said and what gets done. ⚡

Authority becomes a flow. It renews every time you make an honest decision on uncertain terrain, and it erodes every time you pretend to know what you don’t.

The gap we are facing today

There’s an elephant in most boardrooms: the gap between the skills the leadership market incentivizes and the skills the current context demands.

CEO selection processes still mostly prioritize track records of growth in benign environments, ability to articulate simple narratives about complex realities, and relationship networks that guarantee immediate legitimacy.

McKinsey projects that between 40% and 60% of the most valued executive competencies today will be irrelevant or insufficient by 2030, as a consequence of automation, AI, and the reconfiguration of work models. 📊

We’re talking about the core of the executive role. Something to think about.

Where does this leave leadership

The anti-CEO makes decisions with as much determination as the classic CEO, sometimes more. They just make them from a different place: from clarity about what they actually know, what they don’t, and what they need to learn.

Henry Ford knew how to build a car. No 21st century CEO can know everything needed to compete in current markets. That’s an impossible standard.

The real goal is building organizations that learn faster than their competitors and convert uncertainty into advantage.

That’s what it looks like when you stop pretending you can be what you structurally can’t be.

And maybe there’s more authority in that honesty than in the performance we’ve been rewarding.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe’s Glimpse: In the last episode, we looked at what Luckin Coffee and Starbucks teach us about strategic trade-offs. Missed it? Catch up here! ✨

Buongiorno everyone 👋

Did you know that the billable hour has a birthday? Most people don’t.

In 1940, a Boston lawyer named Reginald Heber Smith published a pamphlet through the American Bar Association urging his colleagues to keep daily time sheets and measure their work in “hours and tenths.”

His goal was admirable: bring some discipline and transparency to a profession that had long operated on vague flat fees, handshakes and mutual confusion about what exactly had been delivered and why it cost what it cost. ⚖️

Smith’s innovation spread beyond law. By the 1960s, hourly billing had become the standard across the entire professional services world: consulting firms, advertising agencies, PR houses, accounting practices, architecture studios.

The logic was compelling and truly reasonable at the time: knowledge work is hard to scope in advance, every client situation is different, and the most honest proxy for value is the time expert humans spend on the problem.

The meter had been installed. And for eight decades, it ran without serious challenge.

What the meter was actually measuring

The output model (bill for time, deliver tangible deliverables) solved a real problem: it made an intangible service legible.

A client may not understand the strategic insight behind a marketing campaign, but they understand “200 hours at $350/hr, resulting in 15 creative executions and a brand guidelines document.” It’s auditable and it fits neatly into procurement processes. 📊

That legibility came with a structural cost that nobody talked about much: the model systematically misaligns incentives.

The slower and less efficiently a firm works, the more it earns. Volume of output gets rewarded over quality of outcome. And the client, buried in deliverables, often struggles to connect what they received to what actually changed in their business.

For most of the 20th century, this tension was manageable. The inefficiency tax was real but tolerable, because there was no credible alternative. Measuring outcomes is hard. Attribution is messy.

And so the meter kept running, and everyone agreed not to ask too many questions.

Why past disruptions failed

This time is different.

I know, I know. We’ve heard that before. Every major technology wave of the past century was supposed to disrupt the billable hour, and none of them did.

Calculators didn’t kill accountants. Word processors didn’t empty law offices. Excel didn’t replace CFOs.

The reason is consistent: every previous technology automated the mechanical layer while leaving the judgment layer intact. The calculator compressed arithmetic, but left financial interpretation to humans. Spreadsheets eliminated manual calculation, but left analysis to analysts. 🧠

In each case, the value being sold (expert human judgment) remained scarce, and therefore remained billable.

This is also why something economists call the Jevons paradox held: when technology made professionals more efficient, it often increased total demand for their services. More efficiency created more work. Instead of slowing down, the meter spun faster.

Why this time is different

Generative and agentic AI breaks this pattern in a structurally different way.

While previous technologies helped professionals work faster, AI produces the work itself. A strategic memo gets generated. A legal brief gets written. Market data gets synthesized and interpreted.

And agentic systems don’t stop at producing outputs: they pursue goals end-to-end, removing the need for human orchestration across multiple layers. 💪

The critical difference is economic: the marginal cost of cognitive output is approaching zero.

That has never happened before. No previous technology made the professional deliverable itself nearly free to produce.

When marginal cost of production trends toward zero, a pricing model built on production costs collapses.

Plus, there’s a second variable: clients can increasingly access the same AI tools directly. The access advantage (”we have the tools, data, and expertise to use them”) that historically justified the intermediary is disappearing.

The Jevons paradox may still operate, but only for professionals who successfully shift their value to what AI cannot replicate: contextual judgment, accountability, trust, relationship, applied ethics. Outcome rather than output.

The numbers are already moving

Research from 2025 found that 60% of senior marketing leaders in the US have already reduced agency spending as a direct result of AI adoption.

More strikingly, 83% say they would cut agency spend further if they could fully automate content creation, and that moment is arriving faster than most agencies built their strategic plans around. 📉

Agentic AI systems can already autonomously configure and monitor advertising campaigns, generate creative variations at scale, synthesize research, draft legal documents, and produce first-cut financial analyses. Tasks that have historically justified entire staffing pyramids of billable junior hours.

The consulting world is ahead of the curve on acknowledging this. McKinsey recently disclosed that a full quarter of its fee pool is now tied to measurable client outcomes rather than time spent.

The framing has flipped: instead of clients asking “what will this scope cost?”, they’re coming with “here is the outcome we need, what will you charge us to deliver it?“

Gartner projects that by 2025, over 30% of enterprise software contracts will incorporate some form of outcome-based pricing, up from roughly 15% just a few years ago.

The risks of the outcome model

Before we declare the billable hour dead, it’s worth being clear-eyed about what the alternative involves. ⚠️

Outcome-based pricing only works cleanly when outcomes are measurable, attributable, and within the service provider’s control. In practice, that’s a narrow set of conditions.

A consulting firm can reasonably tie fees to cost savings it directly implemented. It cannot be held responsible for whether those savings survived a geopolitical shock six months later.

Similarly, an ad agency can accept performance-linked fees on a direct-response campaign with clean conversion tracking, but cannot easily price a brand-building effort whose effects materialize over three years.

There’s also a selection risk: when fees are contingent on results, service providers rationally gravitate toward lower-risk mandates. The really hard, transformational work becomes harder to price and easier to avoid.

And there’s a transparency problem. Hourly rates are comparable. Outcome-based proposals are not. When two firms bid with different metrics, success definitions, and risk-sharing structures, procurement can’t evaluate them well.

That creates space to obscure pricing through complexity, which is arguably no better than obscuring it through inefficiency.

The models already emerging

Three alternative structures are gaining traction.

Success fees (base retainer plus kicker) work when the outcome is binary and unambiguous. They get complicated when the outcome is continuous or partially achieved, and create a perverse incentive: advisors may rush to close rather than optimize terms. 💰

Revenue sharing aligns interests powerfully but creates accounting complexity. Who owns the measurement methodology? What happens when macroeconomic factors intervene? These questions kill deals in legal review.

Consulting for equity is the highest-conviction form of outcome alignment. It’s also what most directly challenges the business model: it converts predictable fee income into illiquid asset income. Partners get excited. Finance teams get nervous. And it raises a question no one wants to answer: if the firm is willing to bet its fee on the outcome, what does that say about how confident it was when billing by the hour?

None of these models is complete. What they share is recognition that the old equation (time in, deliverable out, invoice sent) no longer captures where the value is.

What this actually requires

The transition from output to outcome is more than a pricing decision. It’s a strategic repositioning that touches how firms hire, measure performance, select clients, and build capabilities.

Firms need to get clear about what outcome they’re actually responsible for and where that responsibility ends. That takes more specificity than most are used to, more creativity in contracts, more willingness to work with structures that don’t fit the standard templates. 📋

It also puts pressure on clients in a way they’re not used to. They have to define what success looks like before the work starts. Turns out that’s pretty difficult. Maybe that’s the real friction here: clients have to commit to clarity upfront, and many preferred the ambiguity the hourly model allowed.

Reginald Heber Smith’s time sheet solved a measurement problem in 1940. That problem hasn’t disappeared, though. It just has a different shape now.

Those who figure out that new shape first will redesign what it means to get paid for thinking.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe's Glimpse: In the last episode, we explored the age of curation and why choosing well matters more than making more. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I’ve been thinking about what makes something easy to choose.

Not what makes it appealing, or impressive, or aspirational. Just: easy to choose. Clear enough that when someone is standing there deciding between options, yours is the obvious answer for a specific reason.

This came up recently when I was looking at what Luckin Coffee has pulled off in China. It’s a useful example because it shows something counterintuitive about competitive strategy:

You don’t need to be loved to win. You need to be easy to choose.

For years, Starbucks built what is probably the most studied brand system in modern retail. Not just coffee, but a ritual, a reason to stay, a place where paying $6 for a latte feels entirely reasonable because you’re also buying something harder to name. It works. It still works. ☕

Then Luckin came along and essentially said: what if none of that matters to a large chunk of the market?

No atmosphere. No barista names on cups. No invitation to linger. Just: open the app, order ahead, pick up, leave.

It’s now the largest coffee chain in China by store count.

Two different models about what customers are buying

The tempting read here is “convenience beats brand.” I don’t think that’s quite right, and I think it’s worth being more precise about why. 🤔

These aren’t two versions of the same strategy. They’re built on completely different assumptions about what customers are actually buying:

STARBUCKS:

• Optimizes for emotional connection

• Builds preference before the purchase moment

• Invests in physical space and ritual

• Accepts higher cost, slower throughput

• Asks for your time

LUCKIN:

• Optimizes for friction removal

• Wins at the purchase moment itself

• Invests in app, density and speed

• Accepts lower margin per transaction

• Gives your time back

Both work. Both serve real needs. They just require completely different operational realities, different cost structures, different ways of measuring success. 📊

Why splitting the difference doesn’t work

What I find interesting, having watched a lot of organizations try to navigate this kind of choice, is how often the instinct is to split the difference.

A nicer store, and also faster. A strong brand, and also constant discounts. An experience, and also maximum efficiency.

Sounds pragmatic. Usually just means you’ve diluted both propositions without committing to either.

Both Starbucks and Luckin made sacrifices. That’s precisely why both work. Clarity came from the trade-off, not despite it. 💡

Think about a retailer who wants to be “premium and accessible.” Beautiful stores (premium signal) with aggressive pricing (accessible signal). Personalized service (high-touch) with self-checkout (low-friction).

This is how I see this going:

→ Expensive stores that confuse price-sensitive customers, with service that feels inconsistent because staff doesn’t know whether to engage or step back.

The execution might be fine. The problem is lack of clarity about what they’re willing to sacrifice.

Trade-offs across industries

This pattern shows up everywhere. 🗺️

Hospitality: Ritz-Carlton optimizes for anticipating needs. Citizen M optimizes for self-service efficiency. Both work. A hotel that tries to be both ends up with staff hovering uncertainly while guests struggle with under-designed self-service kiosks.

Airlines: Singapore Airlines accepts lower seat density for comfort. Ryanair packs planes tight for price. A carrier that tries both ends up with mediocre comfort at prices that don’t justify the compromise.

Software: Salesforce accepts complexity for customization depth. Notion accepts limited enterprise features for ease of use. Products that try to be both end up overwhelming casual users while frustrating power users with missing capabilities.

Automotive: Volvo optimizes for safety and reliability. Porsche optimizes for performance and driving experience. A car brand that tries to be both ends up with confused positioning and customers who aren’t sure what they’re paying for. 🚗

The successful players aren’t trying to be everything. They’ve made a clear choice about what they’re optimizing for, and they’ve accepted what that choice excludes.

Why organizations avoid the trade-off

The question that tends to make leadership teams uncomfortable, in my experience, is not “what do we want to be?” It’s the more concrete version:

• Are we willing to lose price-sensitive customers to protect the experience?

• Are we willing to give up the experience to compete on speed and scale?

• Are we willing to sacrifice margin for market coverage?

• Are we willing to accept slower growth to maintain quality?

• Are we willing to serve fewer segments really well instead of all segments adequately?

Most strategies I’ve seen avoid answering those questions directly. They stay in the abstract long enough that nobody has to commit to the sacrifice. 💭

There are real reasons for this. Trade-offs feel risky. Excluding customers feels like leaving money on the table. Saying “we’re not for everyone” feels limiting.

And in some organizations, making the trade-off explicit can create internal conflict. Different functions want different things. Sales wants accessibility. Brand wants premium positioning. Operations wants efficiency. Product wants features.

So the compromise becomes: let’s try to do both. Let’s find a way to have it all.

The market doesn’t work that way though. Customers don’t evaluate you against your intentions, they evaluate you against alternatives. When alternatives have made clear trade-offs, they’re easier to choose. 🫵

What clarity actually looks like

Luckin is a useful reminder that you can build something very large without being loved. But you have to be genuinely, operationally, relentlessly easy to choose.

Half-measures don’t get you there.

Clarity doesn’t mean perfection. It means making it obvious what you’re optimizing for, and then delivering on that consistently.

Starbucks customers know what they’re getting: a place to work, meet, or take a break. The coffee is good enough, but that’s not really the point.

Luckin customers know what they’re getting: coffee that’s fast, cheap, and convenient. The experience is fine, but that’s not really the point either.

Both are easy to choose because the value proposition is clear. The trade-off is visible. What they’re willing to sacrifice is obvious.

I think this matters more now than it used to. Markets are more crowded. Customers have more options. Attention is fragmented. When someone is trying to decide between alternatives, clarity helps.

So what’s the trade-off your organization is still sitting on? Where are you trying to split the difference between two fundamentally different value propositions?

There’s no universal answer. It depends on your market, your capabilities, your competition. But I think avoiding the question is more dangerous than making the wrong choice. When you don’t make the trade-off explicit, you make it by default. And default trade-offs usually produce mediocre positioning.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe's Glimpse: In the last episode, we explored how AI is becoming the new UI and what that means for customer experience. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I’ve been noticing something about how I consume content lately.

I used to seek out new things. New articles, new voices, new perspectives. Now I find myself doing something different: filtering. Deciding what not to read, what not to engage with.

Not because the quality is bad. Most of what I see is competent, well-executed, polished. But it all feels like it’s sitting at roughly the same altitude.

We’re now producing more than ever: ideas, visuals, opinions, content. Creation has been radically simplified, and what used to demand technical skill, teams and time can now be generated almost instantly.

Yet despite this explosion of output, I feel strangely underwhelmed. 😒

The abundance problem

For decades, creativity worked as a differentiator. If you could design better, write better, present better, or tell a more compelling story, you stood out. Skill itself signaled quality.

AI hasn’t eliminated creativity, but it seems to have flattened how much skill matters.

The gap between “good” and “very good” has narrowed, and competent output isn’t rare anymore. Walk into any meeting and it’s quite likely that the slides will look professional, the copy will read smoothly, the visuals will be polished. 💅

This creates an odd situation: when everything looks well-executed, being well-executed stops mattering as much.

So we’re producing exponentially more content, all of it competent, and none of it standing out based on execution alone. Which brings us to the second problem: human attention wasn’t built for this.

When everything is available, always, instantly, people don’t feel empowered. They feel tired. I notice this in myself and in the companies I work with.

Faced with too many options of roughly equal quality, the brain looks for shortcuts and trusted signals, desperately seeking guidance. 🗺️

This is where something different starts to have value: helping people understand what matters. Making sense of the abundance instead of just adding to it.

Curation, basically.

What curation means and why it matters now

Curation gets misunderstood as aesthetic preference or taste. “I like this, I don’t like that.”

Real curation is strategic decision-making. Curating means deciding what deserves attention, establishing relationships between ideas, creating contrast and narrative, signaling what matters.

Algorithms struggle with this. They optimize for patterns, for what’s already worked. Curators can intentionally break those patterns. ✂️

While AI generates content really well, humans are still better at interpreting what it means and why it matters.

This distinction becomes critical when you look at what’s actually being produced. We’re in a phase where visuals are polished, language is fluent, concepts are well-packaged.

The issue is that much of it feels interchangeable. AI tools push outputs toward what’s statistically familiar. They train on what already exists, thus gravitating toward what’s already been done.

So even if everything is well-executed, execution loses some of its power to differentiate. 📍

What becomes rare is having a recognizable point of view. A perspective distinct enough that people can orient around it. And this is exactly what curation provides: the human judgment that says “this matters, that doesn’t” in a sea of equally competent options.

How brands and people should position themselves

The most interesting brands I’m seeing aren’t trying to be everywhere or endlessly conversational anymore. They’re acting like editors. Choosing carefully, speaking with intent, referencing culture without chasing it.

Instead of trying to be “relatable to everyone,” they try to be meaningful to someone specific. ✨

This applies beyond brands. Personal leadership, institutions, how individuals show up professionally. Authority now comes from knowing what to prioritize.

I think this is partly why we’re seeing people shift how they build their presence. Constant posting gives way to deliberate choices about what to amplify or ignore.

Much of modern communication has leaned heavily on proximity. Sounding casual, familiar, ordinary. That helped humanize brands and leaders in a digital world.

But we got to a point where we see that proximity alone doesn’t seem to create aspiration. People don’t follow others just because they feel similar. They follow those who embody some version of progress or growth they find compelling. Forward movement matters. 🚀

Desire has direction, and direction requires perspective.

And guess what can provide that? Curation. It tells you: of everything available, this matters. This connects to that. This deserves your attention.

Making sense of abundance

This isn’t about gatekeeping or nostalgia.

It’s about helping people make sense of things when there’s too much to process. When you’re drowning in competent content, someone who helps you figure out what actually matters does something genuinely useful.

This is what matters now: selecting with clear intent, connecting ideas across domains, framing complexity without dumbing it down, and offering coherence when everything feels fragmented. 💡

When AI accelerates output, human judgment becomes the layer that adds value.

I’m watching companies struggle with this. They have content teams producing constantly. AI tools making it even easier to generate more. But they’re not getting more impact (sometimes less.)

The bottleneck has shifted. Creation is abundant and easy. What’s scarce is the ability to make that creation mean something, and that meaning comes from deliberate choices about what to elevate and what to ignore.

What this means practically

If you’re building a brand, leading a team, or trying to establish your own voice, maybe the questions are shifting.

“How do I create more?” becomes less urgent. “What do I choose to elevate?“ becomes more pressing.

“How do I stay visible?” matters less. “What perspective do I actually offer?“ matters more.

“How do I keep up with everything?” stops being the goal. “What do I deliberately let pass by?“ becomes the skill to develop.

This needs different capabilities. Judgment. Synthesis. Having a point of view. Maintaining constant presence becomes less important.

It also takes confidence to let things pass by. To not comment on everything. To resist the pressure to always be adding something to the conversation.

The way I see it, the age of AI could also become the age of curation. AI handles creation well. What people actually need help with is figuring out which of all those created things deserves their attention.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

I have a confession to make.

I wasn’t the most talented student growing up.
I didn’t graduate from a top business school.
I was never in the top 10% of my class.

And yet, I built a career that allowed me to become CEO of global organizations, earn a PhD, co-author international bestsellers, and teach at world-class universities.

How does that math work?

One word: EXPOSURE.

I placed myself in rooms I had no guaranteed right to be in.
Different countries. Different industries. Different conversations.
I spent years working side by side with thinkers like Philip Kotler and engaging with CEOs and CMOs of some of the most iconic brands in the world.

Those rooms changed how I think.
They changed how I decide.
And eventually, they changed how others see me.

The people who accelerate fastest aren’t always the most talented. They are the most exposed: to better questions, sharper minds and bigger standards.

That’s why I’m launching The Next Leap.

A small, global community built around one principle: curated exposure.

🎥 A library of strategic breakthroughs designed to instantly elevate your point of view
🎙️ Monthly fireside conversations with top CEOs (Apple, Uber, Levi’s, Hugo Boss, Spotify and more)
🤝 Peer exchange with ambitious professionals from around the world
🔥 Live sessions with me where we challenge assumptions and think at a higher level

Not a course.
Not a newsletter.
A room.

The kind of room that compresses decades into months. The room I wish someone had opened for me twenty years ago.

If this resonates with you, simply reply “EXPOSURE” to this email, and share it with friends or colleagues who might be interested.

The opportunity is limited to 100 spots eligible for early-bird rates, so make sure to secure yours before they fill up.

Stay curious! 🙌

-gs

Previously on Giuseppe's Glimpse: In the last episode, we explored the B2Cization trap and why B2B borrowed the wrong lessons from B2C. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I’ve been noticing something about how I interact with services lately.

I used to navigate menus, click through pages, search for options, compare things side by side. Now I’m increasingly just asking.

Saying what I need → typing a question → getting an answer

When I look at what’s happening with the companies I advise, I see this everywhere. The interface is shifting from something you navigate to something you have a conversation with.

How we got here

For years, we obsessed over interfaces. Menus, buttons, apps, websites.

We spent enormous energy designing customer journeys across channels. Mobile experience, desktop experience, in-store, call center. Each one carefully crafted and separately optimized.

The goal was omnichannel: be everywhere, stay consistent, let customers choose their path.

What’s emerging now feels a lot different. Thanks to generative AI and natural language processing, the interface is becoming less about navigation and more about interpretation. 🤔

It all starts with one entry point: “How can I help you?“

Behind that sits an intelligence layer that pulls from corporate knowledge, helps call center agents, supports in-store staff, powers chatbots and generates business intelligence.

Same brain serving multiple functions (and no separate channel management needed.)

From many doors to one front door

Think about it this way:

Omnichannel = many doors. You can enter through mobile, web, store, phone, chat. Each is a separate experience that hopefully connects somehow.

Optichannel = fewer, better doors. You optimize for the channels that actually matter and make them work really well together.

Solochannel = one “front door” for intent, many invisible back doors for execution. You express what you need once and the system figures out how to handle it. 📱

Recent Gartner research predicts 30% of Fortune 500 companies will offer service through only a single AI-enabled channel by 2028. They also say 70% of customers will start service journeys via conversational AI by then.

Those exact numbers might not hit, but the direction feels real.

This could mean:

→ Customers don’t need to figure out which channel to use or navigate complex menus. They just say what they need.

→ Companies maintain one intelligence layer instead of separate systems, knowledge bases and training programs for each channel.

→ Answers stay consistent. The AI doesn’t contradict itself across different touchpoints. ✅

→ Service shifts from reactive to proactive. When the system understands intent, it can anticipate needs.

Where I see this breaking down

This won’t work everywhere. A few areas where I think solochannel struggles:

Regulated industries can’t afford the risk. In healthcare, finance or legal services, AI hallucinations create liability nobody can manage. You can’t have an AI giving incorrect medical advice or wrong financial guidance. 🏥

Luxury and high-touch sectors risk losing their differentiation. Part of what justifies premium pricing is the human expertise, the relationship, the brand personality. An AI layer optimized for efficiency might make everything feel the same, which is exactly what luxury brands can’t afford.

Technical fragility becomes a real problem. One intelligence layer means one point of failure. If it goes down or gets trained on bad data, everything breaks. With multiple channels, at least one failure leaves others running.

Exploratory journeys don’t translate well to search. Many customer journeys aren’t about getting a quick answer. They’re about browsing, getting inspired, comparing options, building trust through exploration. You can’t search your way to “What’s best for me?” or “Show me something I didn’t know I wanted.”

For these, we’ll probably see hybrids. Conversation helps navigate, but you still see visual options to explore. 🔍

The disintermediation risk

There’s another risk that I think companies aren’t paying enough attention to.

The solochannel may not belong to the brand.

If people start using third-party AI layers (think ChatGPT, Perplexity, future versions of search) to find answers, those platforms become the interface. The brand becomes just a data source the AI pulls from.

You lose the relationship. You lose the ability to shape the experience. You become a commodity provider feeding someone else’s intelligence layer. 💭

Brands could build AI layers that outperform third parties by offering better answers and owning the entry point.

But that takes serious investment in AI capabilities, data quality and integration. Many companies won’t or can’t do it. They’ll end up answering questions through platforms they don’t control.

The post-UI era

For many sectors, this transition feels hard to avoid.

UX stops being primarily about front-end design and becomes about systems, data and governance.

How good is your knowledge base? How clean is your data? How well do your systems integrate? How do you govern what the AI can and cannot say?

These become the competitive differentiators, not the color of your buttons or the flow of your menus. ⚙️

Channels won’t multiply anymore. They’ll collapse into one intelligence layer that mediates everything.

I’m watching this happen with clients across industries. Airlines using AI layers to handle everything from booking changes to baggage claims. Banks using them to serve both customers and internal staff. Retailers using them to unify online and in-store experiences.

The ones doing it well aren’t just deploying chatbots. They’re rethinking their entire information architecture: how knowledge flows, how systems connect, how humans and AI divide the work.

My mixed feelings

I’m not entirely sure how I feel about this shift.

On one hand, it makes things genuinely easier. Less cognitive load, faster answers, more consistency. As someone who’s spent years studying customer experience, these are meaningful improvements.

On the other hand, I wonder what we lose. Browsing can be enjoyable. Discovery has value. Sometimes the friction in a journey is where trust gets built, where you learn something unexpected.

But whether we like it or not, this is probably where things are headed for many industries. The real questions are: how do we build it responsibly, where do we apply it, where do we deliberately keep the human or exploratory parts that actually matter.

What do you think? Are we ready for this? Or is technology again moving faster than our wisdom about using it?

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe's Glimpse: In the last episode, we explored what makes innovation actually work through the story of Uniqlo. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I’ve been spending a lot of time recently advising B2B clients who want to learn from my experience in the B2C sector.

The conversation usually starts the same way: “Our platform feels terrible compared to consumer apps. We need to be more like them.”

They’re right that something needs to change. But I think they’re often chasing the wrong thing.

What people mean when they say B2Cization

For years, B2B companies operated on an assumption that complexity in the product justified complexity in the experience. If what you’re selling is complicated, the interface can be complicated too.

That’s not really working anymore.

I don’t reset my expectations when I switch from Spotify to Salesforce, or from Amazon to some enterprise platform. The bar for what feels “good” gets set by the best thing I’ve used recently, and that’s usually consumer software. 🌐

People call this B2Cization. Everyone’s talking about it, but I keep seeing companies interpret it as “make everything simpler and prettier.”

That’s not quite it. B2Cization is more about making complexity navigable when it needs to be there. Making things human-scale even when they’re technically complex.

B2B used to hide behind clunky interfaces, dense documentation, hard-to-navigate systems. The thinking was: this signals we’re serious and robust.

Now clunky just reads as inefficient. If your B2B experience wouldn’t fly in B2C, you’re losing ground and might not notice until it’s too late.

The metric problem

Here’s where I see things go sideways most often: companies assume that because the experience should feel like B2C, the metrics should look like B2C too.

But metrics encode how decisions actually work.

B2C decisions happen fast, individually, and you can usually reverse them if you’re wrong.

B2B decisions are rare, involve multiple people, have high stakes and can make or break careers.

When teams start tracking CTR, CPC and conversion speed the way they would for consumer products, they optimize for the wrong outcomes. 📊

I was reviewing performance data recently for a SaaS company. Their ad click-through rates looked great. Marketing was celebrating.

Then we looked at who was clicking: junior people at small companies. Curious, yes. Able to buy? No. Lots of engagement, zero path to revenue.

CTR and CPC still tell you something useful: whether your message makes sense, whether targeting is roughly correct. But they don’t tell you if you’re reaching buyers or just browsers. If you’re building real confidence or just generating curiosity.

High click-through from the wrong people looks healthy on a dashboard but quietly kills your ROI. Getting attention from people who can’t buy is actually the most expensive kind of attention. 💭

What metrics should actually measure

If B2Cization is happening (and it is), it should push B2B toward better measurement, not just faster measurement.

Less focus on volume, more on relevance. Less on speed, more on legitimacy.

That means elevating signals like:

• Are we showing up in the right accounts? (ICP penetration)

• Are multiple roles from the same organization engaging? (buying-group coverage)

• Are we tracking accounts over time, not just individual sessions?

• Are we measuring engagement that leads to sales conversations, not just marketing activity?

• Are we tracking time to meaningful dialogue, not time to click?

These don’t make exciting dashboard slides, but they connect to what actually happens in deals. 📈

B2C metrics track persuasion. B2B needs to track reassurance.

While B2C buyers might be trying to maximize delight, B2B buyers are trying to minimize risk, build internal consensus and not get fired for a bad choice. Nobody loses their job for picking the vendor that got vetted by seven people over three months. They definitely lose their job for the quick decision that looked exciting but failed in practice.

B2Cization does and should change expectations about usability. What it doesn’t change is the fact that B2B buying requires trust, credibility and careful evaluation. 🤝

What this looks like in practice

When B2Cization works:

→ Your interface doesn’t require training to navigate. The design helps instead of fighting the user. Complex functions are accessible without being overwhelming.

→ Your explanations are clear but not dumbed down. You don’t pretend the complexity doesn’t exist, you just don’t bury people in it unnecessarily.

→ Your customer journey matches how B2B actually works: multiple stakeholders, long timelines, the need to build consensus internally.

When it fails:

→ You’re celebrating clicks from individuals when deals require organizational buy-in. You’re measuring speed when the process requires depth. You’re tracking engagement from people who will never have budget authority.

→ Expectations about experience have converged between B2B and B2C. How decisions actually get made has not. Treating them the same because interfaces should feel similar is how you end up with metrics that look good but mean nothing. 📉

I see this play out constantly. Marketing wants metrics that show momentum fast. Sales needs metrics that predict revenue months later. When B2Cization gets interpreted as “be more like consumer,” it makes this worse by importing urgency into a process that runs on deliberation.

You can make B2B feel like B2C and get attention, but you win contracts by understanding how B2B actually works. The trick is doing both without confusing which metrics matter for which outcome.

What I’m taking from this

B2Cization is happening whether companies want it or not. User expectations are set by the best experiences they have anywhere, which are usually consumer experiences.

So yes, make your B2B platform feel as good to use as consumer software, remove unnecessary friction, design like you’re designing for actual humans.

But designing the experience and designing the measurement are separate problems. Confusing them costs money.

Companies that sort this out will have smooth experiences and metrics that connect to real outcomes.

Companies that don’t will have dashboards full of healthy-looking numbers generated by people who will never buy.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe's Glimpse: In the last episode, we explored what power actually means and why the strongest organizations aren't the ones with the tightest grip. Missed it? Catch up here! ✨

Buongiorno everyone 👋

A few months ago, I was in South America for the launch of the Spanish edition of my book. After the tour wrapped up, I managed to carve out a month for one of the most incredible trips of my life: Patagonia, both Chilean and Argentine.

Weeks among frozen fjords, thousand-year-old glaciers, landscapes that felt like another planet. 🏔️

Before leaving, I needed thermal shirts for the hikes. Something that would keep me warm in brutal temperatures but still let me move freely. After checking a few places, I went to Uniqlo because I knew their story from my work in retail innovation and was curious.

I bought their Heattech Ultra Warm Crew Neck T-shirt. It was really good: kept me warm without the bulk, moved naturally, worked exactly as promised.

Standing in that Patagonian cold, wearing a shirt engineered in a Japanese lab, I thought: this is what real innovation looks like. It solves an actual problem. 💡

A clothing company that calls itself a tech company

Uniqlo is the world’s third-largest clothing retailer. But its founder, Tadashi Yanai, refuses to call it a fashion company.

“Uniqlo is a technology company,” he said.

People thought that was strange at first, but then he built a multi-billion-dollar business doing things differently than everyone else. 👔

Yanai grew up in his family’s small menswear shop in 1950s Japan. While studying economics at Waseda University, he developed what most people would consider an unusual obsession: treating clothing like an engineering problem.

Not self-expression. Not trend-following. An engineering problem you solve with data and testing.

One win, nine losses

Yanai’s management philosophy is straightforward: “One win, nine losses.”

Most fashion brands design what they think will sell, produce large runs and hope the market agrees.

Yanai, instead, ran controlled experiments. Staff tested new products, layouts and ideas as small trials. Ninety percent didn’t work. 📊

The difference was what happened after failure. They dissected it, collected data, documented what they learned, adjusted the system, then tested again.

Fashion treated like a science experiment.

This approach led them to partner with Toray Industries in the early 2000s.

Labs instead of design studios

Toray doesn’t make fashion fabrics. They engineer materials for planes, cameras, electronics. Stuff that has to perform under extreme conditions.

Together, Uniqlo and Toray built labs with microscopes and climate chambers. 🔬

Heattech came out of this in 2003. It looked like a basic undershirt, but the fabric blend kept people warm without adding bulk. Thin, flexible, functional.

They sold tens of millions within six years.

More followed. Ultra Light Down jackets in 2009 that packed tiny but stayed warm. Airism fabric in 2012 that managed sweat. Each took years of lab work. Again: material science, not trend forecasting.

Controlling the whole chain

What I find worth noting is how Uniqlo controls the entire process. Design, production, distribution, retail.

They call this SPA: Specialty store retailer of private label apparel. Most competitors work with dozens of suppliers. Uniqlo runs one connected system. ⚙️

This gives them speed. They start with small production runs. RFID chips track sales in real time and, if something sells fast, warehouses ship more within days.

Months of work for competitors happens in weeks for them.

I think this shows where data actually matters. Not predicting what people want before you test, but responding fast once you know.

Data from Japan doesn’t work in London

Nonetheless, going global almost killed them.

2001: Opened stores in London → picked wrong locations → failed

2005: Tried the US → chose suburban malls → failed again

“We opened 21 outlets and closed 16,” Yanai admitted. 🌍

Japanese sizing didn’t work for Western bodies. Tokyo store layouts felt wrong in New York. Cultural differences they hadn’t thought about made products flop.

These expensive lessons changed everything.

They stopped replicating Japan everywhere, and each market became a place to learn. They connected online data to store experiences and made constant adjustments based on local preferences. Test, fail, adjust, test.

Eventually it worked globally, but only after learning that data from one market tells you nothing about another.

This is what I find most interesting about their story. Data was essential, but only when combined with human judgment. When you’re working with humans, copying data-driven strategies from elsewhere rarely works. You need to pay attention, adapt and make calls. 💭

The “one win, nine losses” philosophy accepts this reality. Most experiments fail, even with excellent data.

But if you use that data strategically while remembering you’re dealing with human beings who have specific social, economic, historical and cultural contexts, then you’re more likely to get to that tenth experiment. And when you do, you win big.

What this means for everyone else

The vertical integration that takes Uniqlo from lab to store in weeks is probably what most businesses need now for custom products.

Control your chain, cut delays, manage quality. Use data to see what’s working and move fast. But don’t expect data to predict what humans want before you try.

I see this often in my work. Organizations pile up massive amounts of data, build dashboards, hire analysts. Then they expect the data to tell them what to do.

But data without a business case is like fuel without a destination. It gives you power but no direction.

Uniqlo started with a clear problem: make better clothing through material science. Data helped them test and iterate, but the vision behind it was fully human.

Three things worth remembering

Standing in Patagonia wearing that Heattech shirt, I kept thinking about what makes innovation actually work.

I think the Uniqlo’s story teaches us three things:

1. Innovation needs to solve a real problem. You can have unlimited technology and data. If it’s not fixing something real for real people, it’s just expensive R&D going nowhere.

2. Data needs judgment. Data shows what happened, judgment decides what to try next. Uniqlo’s “one win, nine losses” works because they have opinions about what to test. Data just confirms which opinions were wrong.

3. Good technology is present but invisible. You shouldn’t notice RFID chips or supply chains. You just notice the product works and it’s there when you want it. Tech that needs explaining probably isn’t working.

Yanai built a tech company that happens to sell clothing. Most companies do the opposite: add tech to existing operations and hope for value.

Starting with the problem rather than the solution seems to make the difference.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe's Glimpse: In the last episode, we explored how AI agents are sitting between brands and customers, changing what it means to speak to your audience. Missed it? Catch up here! ✨

Buongiorno everyone 👋

Whether we like it or not, we’re all part of power dynamics at work.

Sometimes you have power over decisions, resources or direction. Sometimes you’re on the receiving end, waiting for approval or working within constraints set by someone else. These dynamics shape how things get done, how fast they move and how people feel about their work.

Over the years, I’ve watched organizations handle power in very different ways. Some use it to unlock potential. Others use it to maintain control. The difference in outcomes is dramatic.

Why the mismatch? I think it starts with a misunderstanding about what power actually means.

What is power?

To understand what power is, we need to look at its etymology. The word comes from the Latin potere, which combines potis (able) and esse (to be). Literally, “to be able.”

Not to control. Not to command. To be able.

This is quite different from how we use the word now, right? When someone says “she has power” in an organization, we usually mean she can make decisions that others have to follow. We think of power as something you hold over people. 💪

But the original meaning suggests something else entirely. Power as the ability to make things happen. To create conditions where action becomes possible. It’s closer to capability than control, closer to enabling than commanding.

When I look at it like this, it reframes a lot of what I’ve seen in organizations. Somewhere between ancient Rome and modern corporations, we changed what power means, and I think that shift matters more than we realize.

How organizations changed what power does

Most companies today are still built on models from the 19th and 20th centuries. That was an era focused on predictability, reliability and repetition. You got efficiency through standardization, scale through specialization, control through hierarchy.

In that world, power had one job: drive people toward predefined actions.

Work got broken into tasks. Tasks got assigned to roles. Roles got governed by rules, procedures and supervision. People weren’t there to decide what to do. They were there to execute what they were told. 🫡

From an operational perspective, this made sense. These systems worked incredibly well in stable environments. They reduced variance, increased output and made mass production possible at scale never seen before.

You could argue (and many do) that industrial progress required exactly this kind of power structure.

Where this starts breaking down

What worked in predictable environments struggles in volatile ones. 🥴

As markets became more dynamic, customer expectations more fluid and technology more pervasive, the same mechanisms that once created efficiency began to produce friction.

I’ve seen what happens when power gets used mainly to direct:

• Decisions slow down

• People stop taking initiative and wait for direction from above

• Responsibility gets fuzzy

• Everyone optimizes for following rules rather than solving problems

The critique from people who study organizations is that power designed to control action often ends up suppressing judgment. You stay safe, but you also stay slow. 🐌

I don’t think this is anyone’s fault. It’s just a mismatch between how we designed organizations and the world they now operate in. We’re using 20th-century logic for 21st-century problems.

What happens in complex environments

In complex environments, leaders can’t know enough to prescribe the right action at every level. You can’t issue instructions for everything because you don’t have all the information.

The challenge shifts from “get everyone aligned on what to do” to “create shared understanding of what matters.”

Power needs a different role here. Instead of telling people what to do, it needs to make them able to figure it out themselves. 💡

This connects back to that original meaning: power as capability rather than control.

From what I’ve seen, power used this way tends to:

• Create clarity about direction without dictating every step

• Build people’s capabilities so they don’t need constant supervision

• Spread decision-making to where the information actually lives

In this version, you measure leadership by how much initiative and ownership flows outward, not by how many decisions flow upward for approval. 🎯

What this requires from leaders

This doesn’t mean abandoning structure or accountability. People who warn against “empowerment” have a point when they say it can become chaos dressed up as freedom.

The shift is more subtle.

Leaders need to move from being the source of answers to being the people who create conditions for good decisions:

• Clear intent about what you’re trying to achieve

• Shared principles for how to make choices

• Decision rights that sit close to where the work happens

• Trust that’s backed up by competence

When this works, organizations get more resilient. Not because they’re tightly controlled, but because they’re genuinely capable of responding to what’s in front of them. 🕺

People stop waiting for permission. They see what needs doing and do it.

Where is power?

Have you ever thought of how we recognize power in organizations?

If we pay attention to it, we’ll notice that we often look for it in the wrong places. Who makes the final call. Who controls the budget. Who needs to approve things. We equate power with authority.

But if the Latin root is right, and power really means “to be able,” then maybe we should look for it elsewhere. In how many people can make good decisions without escalation. In how quickly teams respond to problems without waiting for direction. In how much gets done when leaders aren’t in the room.

The most powerful organizations might actually be the ones where capability is widely distributed rather than tightly held. Where “to be able” applies to many people, not just a few at the top.

That’s harder to build than traditional hierarchy. It requires different instincts from leaders. But I suspect it’s the only approach that works when the world moves faster than any single person can process.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe's Glimpse: In the last episode, we explored why being busy is often a choice and how delegation actually works. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I’ve been thinking about what happens to information when AI agents start making decisions for us.

Over the years, I’ve worked with dozens of brands on their customer journey strategies. The approach was always similar: optimize the product page, get the messaging right, appear at the ideal moment in search results. The assumption was that if you put good information in front of people, they’d find it, understand it and decide.

Now, I’m starting to question whether that model works the way it used to.

How people receive information is changing

In customer journeys now, people often don’t receive brand information as companies designed it. They receive a version that’s been filtered and interpreted by an AI agent.

Consider how these questions get answered today:

“What’s the best product for me?” 🤔

“Which brand lasts longer?”

“Which model offers the best value?”

“Which service is most reliable?”

Information doesn’t arrive as something to search for anymore. It arrives as an answer.

Product information used to mean spec sheets, reviews, blog posts, comparison videos. You’d search, read, interpret and decide. Now an agent does most of that work for you. It extracts data, compares options and hands you a processed conclusion: “this is the best one for you, and here’s why.” 🫵

What this means for brands

A few things I’ve noticed:

1. Brands are losing narrative control. An agent can contradict brand claims. If your marketing says “best battery life” but customer data suggests otherwise, the agent synthesizes from all available sources. It won’t simply repeat your message. 📢

2. Competition is changing nature. Companies don’t compete based on advertising quality anymore. They compete based on how they appear in an agent’s synthesis.

3. The point of influence is shifting. It’s moving from content you control (websites, catalogs, ads) to synthesis engines where decisions form. You’re not trying to be visible. You’re trying to be selected by an algorithm reading on behalf of someone else.

4. Product information is becoming closer to pre-selected recommendations with supporting data. The agent presents a conclusion rather than options to evaluate. 🎯

Which parts of the journey get delegated?

From what I can see, people delegate three main areas:

1. Pre-decision research

   Filtering options, reading reviews, comparing features, summarizing trade-offs. This is the time-consuming part. AI handles it faster and without the cognitive shortcuts we rely on.

2. Standardized purchases

   Flight comparisons, hotel bookings, commodity products like electronics or beauty items, repetitive services. When differences between options are relatively small, agents decide. 💡

3. Post-purchase tasks

   Warranty management, customer service interactions, reorders, subscription handling. Agents are simply more efficient here.

A practical example

Ten years ago, buying a phone meant:

• Searching “best smartphones 2016”

• Reading several articles

• Watching review videos

• Perhaps visiting stores

• Making a decision

Marketing spoke directly to consumers. 🗣️

Now the process looks different. You ask an agent: “I have €600, need a good camera and battery life. Which phone should I buy?”

The agent processes reviews, specifications, availability, pricing, brand reliability and post-sales data. Then it returns 1-2 recommendations.

You might still factor in how the brand feels, its design, what it represents. But these elements act as final filters rather than primary criteria.

What happens to branding?

This is where it gets interesting for me.

Brand identity isn’t disappearing, but its function is certainly evolving.

For humans, it remains important. Visual identity, tone, aesthetics and storytelling… these continue to create emotional connection and a sense of belonging. ❤️

But for AI agents, these elements are invisible. Agents don’t “see” the brand. They process performance data, review patterns, reliability metrics and factual consistency.

So maybe branding doesn’t vanish. It bifurcates: emotional for human touchpoints, data-driven and rational for agents. 🤷

This means brands need clarity about which customer journey elements serve humans and which serve agents. Different objectives require different approaches.

The search engine question

I’ve been considering Google’s position in this shift.

If AI agents are answering questions directly, why would anyone open Google? Queries might start with your agent rather than a search engine.

But I think Google has assets that standalone language models lack:

• Historical query data spanning 25 years. They understand search intent, not just keywords.

• Behavioral information. Click patterns, engagement time, navigation paths. This can’t be reconstructed from text alone. 📈

• Integrated services. Maps, Gmail, YouTube, Android, Chrome. A much larger operational context than standalone LLMs have.

• First-party data at scale. Actual user behavior rather than just content.

Whether they leverage these advantages effectively is another question.

Industry-specific impacts

Travel illustrates this shift clearly (though it applies broadly).

Competition in this field won’t be between destinations or booking platforms. It will be between how agents synthesize information about these options.

Searches like “best hotels in Madrid” will give way to requests like: “Organize 5 days in Madrid, €1,500 budget.“ The agent proposes options, you approve, it handles bookings for flights, hotels and experiences. ✈️

This raises questions for platforms like Booking, Tripadvisor and Expedia. If agents provide optimal answers, what role do these platforms play?

Perhaps they become data sources for agents rather than consumer destinations. Perhaps they need to integrate into the synthesis process rather than the booking endpoint. The path forward isn’t obvious yet.

The uncomfortable part

What makes this shift tricky is that most brands are still optimizing for human attention. SEO, content marketing, brand campaigns... all designed assuming people will see them, read them and decide.

But if the decision increasingly happens inside an agent’s synthesis process, and humans only see the output, then showing up in traditional ways might matter less than being structurally legible to algorithms. 🔢

This requires different capabilities. Less about persuasion, more about accurate data representation.

Some brands will adapt to serve both audiences: emotional branding for human moments, optimized data for algorithmic processes. Others will continue investing in approaches designed for a customer journey that’s evolving past them, investing in beautiful campaigns that agents never “see.” 👀

The timeline and completeness of this shift remain uncertain. But if your strategy assumes customers directly consume your content and form their own opinions about your brand, it seems worth considering what changes when an algorithm does that consumption and opinion-forming first.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

Previously on Giuseppe’s Glimpse: In the last episode, we explored how AI agents are rewriting the rules of customer loyalty and why companies now have two audiences. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I need to tell you something about myself that’s not particularly flattering.

I’m the stereotypical representation of the Busy Being Busy mindset. I’ve been CEO three times, I’m a divorced father living in a different country than my kids, I’ve called five different cities home in the last 20 years, I’ve advised 279 companies (just signed that last contract), got a PhD, co-authored three books with Philip Kotler, and taught more than 10,000 students over the last decade.

I’m telling you this not to brag, but to admit something: for a big chunk of my career, I was terrible at delegation. 💼

And I’m pretty sure I’m not alone. When was the last time you heard a business leader say “I feel like I can take on more work because I’ve delegated effectively”? Those people probably exist, but they’re like an endangered species.

Since most of us struggle with this, I want to share a simple framework I use now when I’m overwhelmed. But first, let me tell you why delegation is so hard in the first place.

The mistakes I made (so you don’t have to)

My first time as CEO, I was convinced I had to control everything to prove I deserved the job. I thought delegation meant giving up control, and giving up control meant failing.

What actually happened is I became the bottleneck. I was working constantly and convinced this was what leadership looked like. 🤦

The second CEO role was different. I walked into a situation where the team wasn’t strong enough for what needed to happen. So I went back to doing everything myself, but this time I had a reason: “I can’t trust anyone to do this right.”

Looking back, that’s such obvious circular logic. The team wasn’t ready because I never gave them a chance to get ready. I never delegated anything meaningful enough for them to grow into. And if you don’t have the right people, guess who’s responsible for creating the conditions for them to develop? For attracting and hiring great ones? 👈

The third time, I thought I’d learned. I delegated more. But I made a different mistake: I tried to be “fair” by delegating equally to everyone. Same types of projects, same level of responsibility, rotating opportunities so nobody felt left out.

What I didn’t realize is that delegating based on equality rather than competence doesn’t actually help anyone. The strongest people don’t get stretched, and the ones who need more development don’t get the targeted support they actually need.

What I wish I’d understood earlier

The word “delegate” comes from Latin: de (away, out) + legare (to send, appoint, entrust). Which means you’re not just offloading work. You’re granting someone the authority to act on your behalf.

That’s a completely different thing. 😲

When I finally got this, I started asking myself: what actually requires me? Not what I’m good at or what I enjoy. What truly requires my specific authority, relationships or expertise that nobody else in the organization has?

Turns out, that list was much shorter than I thought. Maybe 20% of what I was doing.

The rest? I was doing it because it made me feel important, or because I’d always done it, or because I didn’t trust anyone else, or because I thought speed mattered more than developing people.

Once I got more honest about this, things started to shift. My teams got stronger, and I had actual time to think about bigger questions instead of just reacting to whatever landed on my desk that day. 📚

Uncomfortable truths about delegation

A few things I learned that I wish someone had told me:

(1) Delegation feels slow at first.

Teaching someone to do something takes longer than doing it yourself. This is genuinely frustrating. But if you never pay that cost, you stay trapped forever doing the same tasks.

(2) Most leaders (including me) are terrible at asking for help.

We spent our whole careers being the person with answers. Admitting you can’t handle everything feels like weakness. But that’s where delegation actually starts, with saying “I need you to own this.” 🙋

(3) I used to think delegation was about trust.

Like, I need to trust you first, then I’ll delegate to you. But I had it backwards. Clarity comes first. If I’m crystal clear about what success looks like, what the boundaries are, and what decisions are yours versus mine, then trust builds naturally from there.

(4) Tasks versus outcomes matter more than I realized.

If I delegate a task (”write this report”), I get someone who follows instructions. If I delegate an outcome (”figure out why our customer retention dropped 15% and propose solutions”), I get someone who thinks like an owner. The second one takes more initial effort to set up, but it’s the only way people actually develop. 💡

(5) The biggest barrier to delegation was my own ego.

I’m still working on this one. There’s something seductive about being the person everyone needs, the one who has to approve everything, who’s in every meeting, who knows every detail.

It feels like importance. But it’s actually the opposite. If the company can’t function without you for a week, you haven’t built anything resilient. You’ve just made yourself the single point of failure. 💣

I think about this now: being busy all the time is often a choice I’m making, usually without admitting it. Being overwhelmed protects me from harder questions, like whether I’m actually adding value or just adding activity.

What actually changed for me

A few years ago, I heard about this concept from the tech world called the “bus factor.” It’s the question: “How many people need to get hit by a bus before the company collapses?”

If the answer is one or two, you’re fragile. 🚌

I started noticing this pattern in myself. I’d be the person who had to approve things, who knew all the details, who was in every important conversation. At the time I thought that was what good leadership looked like.

But really, I’d just made myself the bottleneck for everything.

After so many years working with Philip Kotler, my brain now organizes things in lists of four words starting with the same letter. So when I’m overwhelmed, I use the 4Ds framework:

→ Do it (only if it requires your unique talent)
→ Delegate (outcomes, not tasks)
→ Defer (schedule for the right time if not urgent)
→ Delete (remove unnecessary tasks)

Sounds simple. The tricky part is actually being honest about what belongs in the “Do it” category. I always want to put more there than probably should be.

I still catch myself thinking “it’s faster if I just do it myself.” And sometimes that’s true. But it also means choosing to stay busy instead of building something that works without me.

Stay curious 🙌

-gs

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.

English speakers, you can skip ahead! 🇮🇹💙

Amici! Visto che parliamo di customer experience e loyalty, volevo ricordarvi che esiste il mio corso “Mastering CX“, dove affronto questi temi in modo approfondito. Se vi interessa: https://products.giuseppestigliano.com/ 🚀

Buongiorno everyone 👋

I’ve spent a good chunk of my career thinking about loyalty programs. The business kind. Airlines, coffee shops, credit cards, all of it.

For years, the logic was simple: get people to come back and you win. So we built elaborate systems. Points, tiers, memberships, birthday coupons. We tracked every purchase and rewarded repetition.

The industry convinced itself that loyalty was something you could engineer, an emotional bond between customer and brand that belonged to both of you. 💖

I’m starting to think that entire framework is breaking down.

Your loyalty program wasn’t built for this

Here’s what’s changing: more and more decisions aren’t being made by humans anymore. They’re being made by AI agents working on our behalf.

These agents search, compare, negotiate, renew subscriptions, evaluate options. They’re the ones deciding which brand to choose, which vendor to reorder from, which service to switch to. And they don’t make decisions the way people do. 🤖

Traditional loyalty programs assumed customers would remember your brand, compare prices themselves, seek out deals, and repeat purchases out of habit or convenience. Those assumptions only work when humans are doing the deciding.

But if an AI agent is handling your grocery reorders, it doesn’t “remember” your favorite brand. It retrieves current data. It doesn’t get attached to the coffee you’ve always bought. It optimizes based on price, reviews, delivery speed, whatever parameters it’s been given. 🛒

I was talking to someone who works in tech recently and they said something that stuck with me: “Loyalty lives where decisions are made. And decisions are moving out of human heads.“

I think they’re right, even if I’m not totally sure what that means yet.

What agents actually care about

Loyalty has always been part rational, part emotional, part structural (the hassle of switching). AI agents strip away most of the emotional part.

They don’t care about your brand story. They don’t feel nostalgia. They don’t identify with your values unless those values translate into measurable outcomes: price, performance, reliability, sustainability metrics.

This makes me wonder what happens to brands that built their whole identity on emotional connection. Are they suddenly at a disadvantage? Maybe. Or maybe emotional loyalty just matters in different contexts now. 🤷

At the same time, I can see how this might actually help some brands. The ones that always focused on doing the basics really well (fast delivery, accurate orders, good service) might finally get rewarded for that instead of losing to competitors with better marketing.

Points programs won’t save you

Most loyalty programs are essentially bribes. Buy ten coffees, get one free. Fly enough miles, get a free upgrade. It’s transactional. ✈️

AI agents don’t respond to that. They’re not going to stick with your airline because they’re chasing status. They’ll just book whoever has the best combination of price, schedule, and reliability for that specific trip.

What matters instead is consistency. Does your product do what it promises? Are your delivery times accurate? Do you have fewer returns than competitors? Lower total cost of ownership? Better sustainability data?

If the answer is yes, consistently, the agent chooses you. If not, it doesn’t. There’s no forgiveness for an off day. No loyalty that builds up over time. Just: did you perform this time? ⚡

In a weird way, this is more fair. You get chosen because you’re actually better. But it also means smaller companies with less data might become invisible to these systems. I don’t know how that plays out…

Agents don’t procrastinate

Something I’ve noticed over the years is that a huge amount of “loyalty” is really just inertia.

People stay with mediocre banks because switching is annoying. They keep subscriptions they barely use because canceling requires effort. They stick with insurance companies that are probably overcharging them because comparing alternatives feels like homework. 💤

AI agents don’t experience friction. They never think “I’ll deal with that later.” If something better exists, they find it and switch. Instantly.

This is terrifying for industries built on customer laziness. Telecom. Banking. Insurance. Utilities. A lot of their “loyalty” was really just people not bothering to leave.

When an agent can compare every option and execute a switch in seconds, that comfortable cushion disappears. And honestly, maybe it should. But it does mean a lot of business models need to be rethought. 🤔

Two different games

So what do companies actually do with this?

I think you have to play two games now. One for the agent, one for the human.

For the agent: you need clean data. Transparent pricing. Reliable metrics. Machine-readable everything. Performance that’s consistent and measurable. 📈

For the human: you still need emotional connection. Brand meaning. The reasons someone would care enough to set their agent’s parameters to prefer you in the first place.

The companies that figure out how to do both will probably be fine. The ones that only know how to do one... I’m less sure about.

Is this the end of loyalty?

Not really. It’s just changing shape.

It’s becoming more data-driven, more performance-based, more immediate. The emotional part doesn’t disappear entirely. It just matters at different moments in the relationship. ⚖️

There are people who think this shift will make markets more efficient and reward genuinely good companies. There are others who worry it’ll turn everything into a commodity and concentrate power in whoever controls the agent platforms.

Both opinions seem reasonable to me.

What I’m fairly certain about is this: if your customers start delegating purchase decisions to AI agents, and your entire loyalty strategy is built around points and perks and emotional storytelling, you’re going to have a problem.

Because loyalty isn’t just going to be about how customers feel about you anymore. It’s also going to be about how you show up in an algorithm’s evaluation. And those are two very different challenges.

What companies need to understand: you now have two audiences, the human who sets preferences and the agent who executes them.

Stay curious 🙌

-gs

Previously on Giuseppe’s Glimpse: In the last episode, we explored the Cobra Effect and why misaligned incentives turn solutions into bigger problems. Missed it? Catch up here! ✨

Buongiorno everyone 👋

I’ve been thinking a lot lately about why some messages stick and others don’t.

Last week, I sat through a presentation where someone explained their product for twenty minutes. Features, benefits, integrations, use cases. All true. All relevant. And yet by the end, I couldn’t have told you what the thing actually did.

This made me think of something that happened years ago.

Steve Jobs stood on stage in 2001, holding something small enough to fit in his palm. The market was already crowded with MP3 players. Competitors were shouting about megabytes and technical specs.

Jobs said five words: “A thousand songs in your pocket.“

That’s it. The focus was not on storage capacity, file formats or technical superiority claims. Just an instant mental picture of your entire music library, anywhere you go. 🎵

The iPod wasn’t necessarily better technology, but it was better communication. And I think Jobs understood something that most of us forget: people don’t really want to think if they don’t have to.

Our brains are lazy (in a good way)

I don’t mean that in a cynical way. Our brains are remarkable, but they’re also pragmatic. We process 11 million bits of information per second. According to research, we make roughly 35,000 conscious decisions every single day. If we deeply analyzed each one, we’d never leave the house.

So our brains evolved shortcuts, what psychology calls heuristics, that let us navigate complexity without burning out. Research suggests that 95% of human behavior is driven by feelings and emotions, not rational analysis. 🧠

Daniel Kahneman, the Nobel Prize-winning psychologist, put it perfectly: “Thinking is to humans like swimming is to cats. They can do it; they just prefer not to.”

And if you’re leading change, launching something new, or trying to move people toward a vision, this matters more than you might think.

Why visualization works

Here’s what makes “a thousand songs in your pocket” so effective: you can see it immediately.

Your brain doesn’t need to translate megabytes into meaning. It doesn’t need to compare specifications. The image forms instantly, and with it, the desire.

Compare that to how most organizations communicate:

• “We deliver innovative solutions”

• “Our platform enables digital transformation”

• “We provide best-in-class customer experiences”

What does any of that actually look like? 🤷

M&M’s figured this out decades ago: “Melts in your mouth, not in your hand.” Two sensory images. You can feel both. It’s concrete, specific and memorable.

When I work with teams on transformation initiatives, this is where things often break down. The strategy is solid. The technology works. But the communication is abstract. People nod politely in meetings and then... nothing changes.

And I don’t think that’s because they don’t care. Rather, their brains couldn’t grab onto anything concrete enough to remember or act on.

The trap of saying everything

There’s another pattern I see constantly, and it comes from a good place.

Brand owners, founders, leaders: they believe in what they’re building for many reasons. So when it’s time to communicate, they list everything: “We’re great at A, B, C, D, E, F and G.”

This is called goal dilution. And it’s deadly. 💀

A partially engaged human mind (which, let’s be honest, is most minds most of the time) can’t hold seven things. It can barely hold two. When you say you’re excellent at everything, people remember nothing.

IKEA understood this. Their promise is singular: affordable, well-designed furniture you assemble yourself. Love it or hate it, you know exactly what IKEA means.

I’ve watched transformation programs stall because the “why” had twelve bullet points. Brilliant products that failed because their value proposition needed a flowchart. The teams usually had good reasons for every feature they mentioned. But when you try to say everything, nothing lands.

What this means for how we lead

If most decisions are emotional and our brains prefer shortcuts to analysis, then maybe clarity is the whole game.

When you’re driving change, whether that’s AI adoption, a new way of working, or a shift in culture, it might help to ask:

Can people see it? Not understand it intellectually, but visualize what success looks like? If your team can’t picture it, I’m not sure how they’re supposed to move toward it. 🎯

Are you focusing on one thing? You might have ten reasons your initiative matters. But in my experience, you’re better off picking the most visceral one and letting that be the anchor. Everything else can come later.

Does it connect to something human? Data explains what we do. Emotions explain why we do it. If your communication doesn’t touch something real - pride, relief, possibility, even frustration - it probably won’t land.

This isn’t about manipulation or oversimplification. It’s about recognizing that if you want your message to get through, you need to work with how people’s brains operate.

The best communicators aren’t the ones with the most to say. They’re the ones who know what’s worth saying. 💫

A thousand songs

Jobs didn’t invent the MP3 player. He just knew that people don’t buy gigabytes. They buy the feeling of having their entire music collection with them on a morning run.

That clarity, that commitment to a single, visual, emotional truth, changed an industry.

I think about this every time I’m tempted to add “just one more point” to a presentation. Or when I see strategy documents that require a decoder ring.

Somewhere in your organization right now, someone is trying to explain why change matters. They’ve got a slide deck with seven key pillars and fifteen bullet points.

What if they just found their “thousand songs in your pocket” instead?

Stay curious 🙌

-gs

P.S. This episode was inspired by insights from Michael Aaron Flicker and Richard Shotton’s Hacking the Human Mind, a book packed with smart behavioral science that every communicator should read. Check it out here. 📚

Oh, wow! You made it to the end. Click here to 👉 SHARE this issue with a friend if you found it valuable.