Japan, where high expectations can turn into a weapon

I’ve already written about how high the bar is for customer service in Japan - you can read that here. But that same standard has a shadow side.

I’ve worked in customer service for years, with all kinds of clients, and during that time I’ve seen this problem a handful of times, sometimes inside contact centers and other times witnessed it happen in front of me in stores as a bystander.

Japan’s sky-high expectations have a cost. And in Japan, that cost has a name.

Kasuhara.

Not a New Word

Kasuhara is short for kasutamā harasumento. In English, customer harassment.

It covers customers who scream at staff, who demand unreasonable apologies, and who file complaints that, far from being with a constructive aim in mind, are more designed to humiliate the staff on the other side, rather than to fix anything.

The behavior itself is not new here, and it's not new anywhere in the world. Almost every service worker has a story of harassment.

It was during the late 2010s, when labor unions representing retail and service workers began collecting data instead of mere anecdotes so there would be numbers backing up the issue.

The Impacts of Customer Harassment

Customer harassment has many impacts, the most important being the impact on the receiving end of it, the one that the staff needs to deal with. Additionally, companies are also paying for it.

For the person on the receiving end, the cost runs deeper, and Japan has started measuring that too. A 2022 survey by the country’s largest trade union confederation found that among workers who had faced kasuhara in the past three years, 76.4% said it had negatively affected their lives. One in ten quit their job over it.

For companies, the cost is easy to measure. In Japan’s own government survey on workplace harassment, companies that had dealt with kasuhara named one type of harm more than any other: disruption to normal business operations, at 63.4%. This impact you can imagine on how it can translate economically for a company when their employees are not motivated leading to business operations not working correctly.

This is not just a customer service problem. It’s a workplace safety problem.

Finally, with the service sector making up 70% of Japan’s workforce, this isn’t a niche issue affecting a handful of unlucky employees. It touches most of the people who show up to serve you every day.

What Is Japan Doing About It?

In April 2025, Tokyo passed an ordinance against customer harassment that took.

Also, starting October 1, 2026, a national law will make it mandatory for every employer in Japan, from convenience stores to major corporations, to protect staff from customer abuse. In practice, that means three things:

• A clear policy. Staff need to know, in writing, that the company will back them up when a customer crosses the line.

• A safe way to report it. Workers need a channel to flag abuse without fear of being punished for speaking up.

• A plan for serious cases. That can mean a warning, refusing service, banning the customer, or calling the police.

There’s no fine for a company that ignores this. The consequence is guidance and being publicly named as non-compliant. It’s a soft law with a clear message: protecting your staff is no longer optional.

Can a Machine Protect a Human From This?

This is where AI actually gets interesting to me, and not for the reason most people talk about.

There are many ways in which AI could help with customer harassment. For example, AI can listen to a call or read a chat in real time and flag when a conversation is turning abusive, before a manager even notices.

It can also spot patterns across thousands of interactions that a single supervisor never could. It can give a worker a script to de-escalate, or a supervisor an alert to step in. It can also keep a clean record of what happened, so a worker’s word isn’t the only evidence when they report it.

Of course, none of that stops a person from losing their temper. But it can catch it faster, document it properly, and make sure the worker isn’t left to handle it alone.

Most people talk about AI to cut costs. One of its most useful jobs might be protecting the people companies almost forgot to protect.

SoftBank Response to Customer Harassment

On February 2, 2026, SoftBank announced the commercial launch of “SoftVoice,” an AI-powered voice altering solution designed specifically for call centers. The system uses real-time AI voice conversion technology to soften aggressive customer tones during calls while preserving the original content.

Beyond tone moderation, SoftVoice also provides the following features:

• Automated escalation warnings: a three-stage automated message system can issue policy-aligned warnings in cases of abusive language or excessively long calls, helping to de-escalate emotional confrontations.

• AI-powered noise suppression: reduces background noise to improve call clarity and reduce cognitive load.

• Call recording: supports compliance, quality control and risk management.

Why I like this solution is because it is one of the few AI use cases used in contact centers that focus on improving the well-being of the agents.

What AI Can’t Fix

Below is my take on the matter.

Everyone talks about AI as a tool for efficiency. Faster support, lower costs, fewer people needed on the phones. That conversation is important, but it’s not the only one worth having.

Protecting employees is a use case I find far more interesting, and far more human. If you’ve worked in this industry, you already know what this kind of abuse does to a person. Over time, it wears people down. It causes real psychological harm, and this type of harm is not easy to recognize.

Here is where AI can help since it can detect the problem faster and document it better. It can support the worker in the moment by for example, modifying the voice of the customer and making it softer so the receiving end does not need ot suffer it.

However, AI can’t fix why the problem exists in the first place.

While I don´t have all the answers, the one thing I know is that this will only be solved through education. Teaching customers, from a young age, that respect for the person serving you isn’t optional. Teaching companies that protecting staff is as important as protecting the brand. Kasuhara won’t disappear because a system got smarter. It will disappear when people, as a whole, decide it’s no longer acceptable.

AI can catch the problem earlier. Only education can stop it from happening at all.

If this is your first time here, I’m Daniel, and I’ve spent 12 years living and working in Japan, leading Customer Experience and Conversational AI for some of Japan’s biggest companies. I write about what gets lost when technology moves fast, and what Japanese service culture can teach us about bringing it back.

The next article will be coming soon. In the meantime, you might find the following reads interesting:

Recently I had the chance to talk with Kane Simms from VUX World, and we talked about omotenashi, CX, and conversational AI. It was a great conversation on how a centuries-old Japanese philosophy of hospitality can shape the way we design AI-powered customer experiences.

In this article, I’ll walk through the main areas we touched on during our conversation.

From Spain to Japan: my journey into conversational AI

I’m originally from Spain but have called Japan home for 12 years, after calling Edinburgh, Dublin, and Shanghai home. Learning Chinese in Shanghai is actually what led to a job offer in Tokyo.

My path into conversational AI ran through business development, then retail tech, then five years at Rakuten building customer service chatbots across its many verticals, and then going through the shift from pre-LLM to generative AI.

What is Omotenashi?

Omotenashi is a Japanese philosophy of putting the customer first: anticipating their needs, apologizing sincerely when something goes wrong, and solving problems at the right moment. Actually, it’s not different in kind from other good service elsewhere in the world, but I would say the Japanese can bring it to another level.

Small examples stick with me: supermarket staff bowing to the warehouse every time they enter or leave, even when no one’s watching; Shinkansen cleaners bowing to the train car before moving to the next one. It’s become a market differentiator too, Japanese airlines and car companies actively market omotenashi as part of their brand.

Why apologies matter in Japanese customer service

Japanese customer service places huge weight on apologizing sincerely before even attempting to fix the problem. Studies show customers often value a sincere apology more than compensation, the opposite of what I’ve generally seen in the West, where the fix usually comes first.

This extends to how mistakes are handled generally: deflecting blame, even partially, is a real misstep. The expected response is closer to “this is on me, I’m sorry”, putting both parties on the same footing before moving to a resolution. It may connect to how much more individualist Western culture is by comparison; a hedge like “sorry you feel that way” just doesn’t land the same way in Japan.

Why Japan scores low on NPS

Japan consistently scores among the lowest average NPS across industries compared to markets like Brazil, India, or Australia, not because service is worse, but because expectations are higher. The same restaurant often scores lower on a Japanese review platform than on Google Maps.

That gap shows up directly in conversational AI design. Japanese has 10-15 ways to say “sorry,” and choosing the right level of formality for a situation matters; generative AI now makes that calibration possible. Escalation matters just as much: Japanese customers strongly prefer a human when a bot can’t help, and 56% say they won’t return to a brand after one bad interaction.

Ambiguity, anticipation and cultural context in AI design

Anticipation is core to omotenashi, sensing a need before it’s voiced and proactively reaching out, rather than waiting to be asked. That’s as much a design principle as a feature: mapping likely problems by contact type and building in proactive signals wherever possible.

Language adds complexity. Japanese often avoids direct refusal in favor of ambiguity (aimai), used to preserve harmony; words like chotto (”a little”) or demo (”but”) can signal dissatisfaction far more subtly than an outright complaint. Catching that requires prompting AI with specific examples of these soft expressions. Even in person, a phrase like “let me consider it” (kentoshimasu) is, in the right context, a clear no, it just takes time living in the culture to read reliably.

Building conversational AI for the Japanese market

Expectations are simply higher here, so metrics that look weak by other markets’ standards may just reflect a tougher baseline. Beyond apology tone, linguistic etiquette matters: Japanese uses three writing systems, and mixing them inconsistently within a conversation is something native speakers notice immediately. Keeping messages under ~50 characters helps too, since the script is visually denser to read.

Digital personification is a distinctly Japanese opportunity, mascots are everywhere, and giving a chatbot its own named, personified identity tends to land well. Integration with local tools matters too: LINE, Japan’s dominant messaging app, offers deeper business functionality than WhatsApp does in the West, making it close to a must-have channel.

How omotenashi can raise the global CX standard

This isn’t about Japan having answers the West hasn’t found, there’s plenty Japan could learn from a more direct approach too. The interesting question is what each side can take from the other.

What omotenashi offers is a genuine raising of the bar: more accountability, more sincerity, and a habit of pausing to empathize before jumping to the fix. Solving the problem is a given, the harder, more valuable part is pausing first to say “I understand this is frustrating, and now let’s fix it.” That’s the piece worth building into how we design conversational AI everywhere.

Hello and thanks for reading. If this is your first time here, I’m Daniel, and I’ve spent 12 years living and working in Japan, leading Customer Experience and Conversational AI for some of Japan’s biggest companies. I write about what gets lost when technology moves fast, and what Japanese service culture can teach us about bringing it back.

I publish an article weekly, next one coming soon. In the meantime, you might find the following reads interesting:

I was in Singapore last week, working and on the way to the wedding of one of my best friends in Malaysia.

Eleven years in Japan means I can’t land anywhere new without immediately reading the service culture. Singapore made me think harder than most places have in a while, mostly because it didn’t fit the story I usually see and tell.

I almost wrote the easy version of this. Singapore runs on rules, Japan runs on heart. It’s a clean story but it’s also not true, and I’ve lived in Japan long enough to know it isn’t.

1. Pay Peanuts, Get Monkeys

Singapore has a saying: pay peanuts, get monkeys.

Underpay, understaff, undertrain, and the service reflects it. So the country leans the other way. Pay for standards. Audit for them. Make good service the easiest path, not the exceptional one.

There’s a second idea sitting underneath it: kiasu, the fear of losing out.

It’s usually used to describe Singaporeans rushing for a queue or a deal, but it shows up in business too. Nobody wants to be the company that lost the customer because the experience was worse than the one next door. That fear is a kind of fuel. Not affection for the customer. Closer to not wanting to lose.

Here’s the strange part. Every hawker stall, every taxi, every interaction I had in Singapore was genuinely warm. Not just efficient. Warm. If kiasu were the whole story, I’d expect brisk and competent. I got something closer to care. Which tells me the system produces more than what it was designed to guarantee.

2. The ¥50 Nobody Was Chasing

Now the part that doesn’t fit neatly into any system at all. In Japan, you find real attention to detail in moments nobody is being paid extra for and nobody is grading. A shopkeeper re-folding your bag so the logo faces you. Someone running down the street to return ¥50 in change you didn’t notice you’d left behind.

Nobody is auditing that. There’s no fear of losing the customer behind it, because the customer already paid and already left. It happens because the person decided it should.

That’s the moment I usually reach for to make my point about Japan. But it’s not the whole picture, and pretending it is would be dishonest.

3. Where the Incentive Runs Out

Go to a Japanese bank or a government office and you’ll meet a completely different version of the country. The process is rigid. The forms are excessive. The rules exist for the sake of the rules, and the person in front of you often can’t deviate from them even when it’s obviously the right thing to do. I’ve sat through hours of bureaucracy that no customer-first company would ever design on purpose.

This is where Singapore wins outright. Opening a bank account in Singapore is fast, because the bank has a direct interest in you opening it. The incentive is aligned, so the friction gets removed. In Japan, the bank often has no equivalent urgency. The CX is bad not because Japan forgot how to care, but because nobody built a reason to.

So it isn’t “rules vs. heart.”

It’s this: wherever there’s a clear incentive, you tend to get good service, in both countries. Singapore builds that incentive deliberately, almost everywhere, which is why it’s so consistent. Japan sometimes gives you extraordinary care for free, in moments with no incentive attached at all, and sometimes gives you terrible service in the exact places where an incentive should exist but doesn’t.

I don’t think one country does this better. I think they fail differently, and they succeed differently, and the interesting question isn’t which one gets CX “right.”

It’s why care shows up reliably in one place, and randomly in another.

—

I’m in Malaysia next, and the way a restaurant folds a banana leaf told me something about service I wasn’t expecting. More on that soon.

Hello and thanks for reading. If this is your first time here, I’m Daniel, and I’ve spent 12 years living and working in Japan, leading Customer Experience and Conversational AI for some of Japan’s biggest companies. I write about what gets lost when technology moves fast, and what Japanese service culture can teach us about bringing it back.

I publish an article weekly, next one coming soon. In the meantime, you might find the following reads interesting:

Or we can also connect on LinkedIn. I talk about these topics there too.

Part 1: The Country Where the Customer Is God

In Japan, there is a phrase that shapes every service interaction before it even begins: okyakusama wa kamisama desu. The customer is God.

Not “the customer is always right,” which is a business policy. Not “customer satisfaction is our priority,” which is a corporate value statement. God. A word that carries completely different weight, and signals something about how Japanese service culture was built from the ground up, not as a competitive advantage, but as a philosophy.

If you have visited Japan, even as a tourist for a week, you have probably felt this. The department store staff who bows as the elevator doors close, whether or not anyone is watching. The konbini cashier who places your change in your hand with both hands. The restaurant that brings you a warm oshibori towel before you have had a chance to ask for anything. Small moments, each unremarkable on their own, that accumulate into something you struggle to describe when you get home.

I have lived and worked in Tokyo for over twelve years. I have led customer experience teams and spoken internationally about what Japan understands about service that most global companies miss. And I still get surprised. Not because the service is always perfect (anyone who has tried to open a bank account here will confirm that) but because the intention behind it is so consistent that even the imperfections feel considered.

So the question I want to explore in this article is both simple and important for anyone building products or services in Japan: are Japanese customers actually as demanding as the reputation suggests? What does the data tell us that the reputation alone doesn’t? And what does Omotenashi (Japanese Customer Service) have to do with this?

Part 2: Why Japan's NPS Score Looks Like a Disaster (And What It Actually Means)

Here is a data point that stops most CX professionals when they first see it.

Japan’s Net Promoter Score, the widely used metric where customers are asked how likely they are to recommend a company to a friend, sits at -52. That is not a typo. According to SurveyMonkey’s global NPS study, which surveyed consumers across nine countries, Japan sits at the very bottom of the global ranking. Brazil leads with a score of +62. India follows at +51. Japan, the country globally celebrated for its service culture, scores lower than every other nation in the study by a wide margin.

By any conventional benchmark, that score would trigger an emergency board meeting.

So is Japanese customer service actually terrible? Obviously not. The explanation is cultural, and understanding it is one of the most important things any global company entering Japan can do.

Based on the same study by SurveyMonkey, Japan consistently shows the lowest scores when asking customers about their satisfaction on different aspects of the Customer Experience.

HubSpot’s research on international NPS puts it like this:

And other analysis, this time from Qualtrics, also shows Japan as the country with the lowest NPS on average.

Japanese consumers tend to answer surveys the way they navigate most social situations, carefully, modestly, and without extreme positions in either direction.

This is not just a quirk. It is deeply structural. As Wayfinder Japan’s analysis of Japanese survey behavior explains, Japanese respondents tend to anchor their answers toward the middle of any scale, because they compare their response to what they think others would say. The same phenomenon appears in restaurant reviews: on Tabelog, Japan’s most trusted dining platform, 97% of restaurants score 3.5 or below out of 5, and only the top 3% reach between 3.5 and 4. A 3.8 on Tabelog signals exceptional quality. The same score on Google would make you wonder if the kitchen had an off day.

The SurveyMonkey study reinforces this, noting that fewer than one in three Japanese consumers share about their purchases or brand interactions publicly, compared to only 10% of consumers globally who keep entirely silent. Japanese customers process their experiences inwardly. They do not broadcast them. That makes the standard CX measurement toolkit, built around recommendations, reviews, and vocal feedback, a genuinely poor fit for this market.

Part 3: Silent, But Not Forgiving

Here is where the picture becomes more complicated, and more consequential.

Japanese customers are not loud complainers. They do not leave aggressive reviews. They do not call to escalate. As Rikkyo University researcher Shunichi Nozaki has noted, reported by NextShark, Japanese customers would rather move on to a competitor than confront a company about a poor experience. And they move on fast.

The same American Express survey reported by NextShark found that 56% of Japanese consumers said they would take their business elsewhere after just one bad customer service experience, the highest proportion of any country in the study. To put that in context: the equivalent figure for the US was 32%. For the UK it was 37%. Japan was not slightly ahead, it was in a different category entirely.

Think about what that means in practice. You will not get a warning. There will be no angry email, no escalation call, no one-star review explaining exactly what went wrong. The customer will simply not return. And because Japanese culture values discretion, they probably will not tell anyone else why either, at least not publicly.

There is one more layer worth understanding here. The same Wayfinder Japan analysis points out that Japan showed the smallest gap in NPS between companies that are liked and companies that are disliked. Elsewhere in the world, there is a dramatic difference between how people rate brands they love versus brands they tolerate. In Japan, that gap is narrow. Customers hold back both the enthusiasm and the frustration. What this means for businesses is that there are very few visible warning signs before a customer quietly exits for good.

Have you ever left a brand in Japan without saying why?

Part 4: What Japanese Customers Really Want - And Where AI Customer Service Falls Short

So if Japanese customers are difficult to read, quiet when unhappy, and quick to leave, what do they actually respond to positively?

The answer from the data is consistent, and it is not what most technology companies prioritize when they build for efficiency.

A Genesys survey of 5,000 consumers across six countries, including Japan, Germany, the US, the UK, Australia, and New Zealand, found that a full third of Japanese respondents believed customer service had become less personalized in recent years. That was the highest proportion of any country surveyed. While German and British consumers largely felt service had improved with technology, Japanese consumers felt it was moving in the wrong direction.

On empathy specifically, the Genesys study found that only 40% of Japanese consumers said the companies they deal with show genuine empathy when resolving issues. The US figure was 58%. In a country where empathy is not considered a differentiator but a baseline expectation, that gap is not small.

This has direct consequences for AI deployment. An article on Japanese customer support culture from Delight notes that Japan has historically shown slower adoption of automation in customer service precisely because of this emphasis on retaining the human touch. A chatbot that resolves queries efficiently but feels cold or scripted is not seen as a neutral tool in Japan. It is experienced as a step backwards from what the customer expected.

And because Japanese customers won’t complain when a service falls short of their expectations, they will simply leave, a poorly implemented AI solution creates a particularly dangerous situation: invisible, silent churn that is nearly impossible to diagnose after the fact.

Part 5: Omotenashi: The Japanese Philosophy That AI Cannot Replace

All of this data points back to a single concept that has no direct translation in English, and that is precisely why it is so often misunderstood by companies entering Japan.

Omotenashi.

It is usually translated as Japanese hospitality, but that flattens it considerably. It is more accurate to describe it as the practice of anticipating what someone needs before they know they need it, and delivering it without expectation of recognition or reward. No transaction. No performance review. No feedback survey. Just genuine care expressed through sustained attention.

It is why the train conductor bows to an empty carriage. Not because it is in the service manual. Because it reflects who they are.

Japanese service prioritizes building trust and fostering long-term relationships through empathy and thorough problem resolution, in direct contrast to Western approaches, which tend to focus on efficient resolution and meeting defined service level agreements.

What the data in this article reveals, taken together, is something that should inform every CX strategy built for Japan, and honestly, every CX strategy built for anywhere. Japanese customers do not loudly demand more. They quietly expect better. They will not punish you with a complaint. They will reward you with loyalty if you earn it, or remove themselves without explanation if you don’t.

Japan is not perfect at this. Rigid processes, bureaucratic friction, and cautious digital adoption create real pain points that any long-term resident here knows well. But the underlying orientation, toward anticipation, care, and the dignity of the person being served, is something the rest of the world is only beginning to realize it needs urgently.

As AI takes over more of the surface layer of customer interaction, what remains distinctly human becomes not a soft add-on but a competitive necessity. Japan has been practicing that for centuries. The rest of us are still figuring out what we gave away.

Part 6: Where do we go from here?

Now the natural question we can ask ourselves is:

So what?

If I am a company considering providing services to the Japanese market and providing an AI that will directly face customers, what does it mean to know that Japanese customers will be more demanding and have high expectations? How can you prepare for it?

This, exactly, is what we will be exploring in the next articles.

Hello and thanks for reading. If this is your first time here, I’m Daniel, and I’ve spent 12 years living and working in Japan, leading Customer Experience and Conversational AI for some of Japan’s biggest companies. I write about what gets lost when technology moves fast, and what Japanese service culture can teach us about bringing it back.

This is the second article in a series about Omotenashi and how this Japanese customer service philosophy can be the answer to a more human-centric AI. If you want to follow where it goes, subscribe below, and you won’t miss the next one.

In the meantime, these articles might be worth your time:

Or we can also connect on LinkedIn. I talk about these topics there too.

It was my first week working inside a Japanese supermarket when my supervisor stopped me at the warehouse entrance.

“You have to bow,” she said.

I looked around. The warehouse was empty. There were no customers, no colleagues, and no one watching.

“To who?” I asked.

“To the genba [the supermarket floor],” she said. As if that explained everything.

It didn’t. Not yet.

I bowed anyway, not entirely sure what I was doing.

And then I bowed again on the way out. Every time. Empty room or full room, it didn’t matter. I always bowed regardless.

It took me a while to understand what I was actually being taught. It wasn’t a ritual. It wasn’t politeness for its own sake. It was something much more fundamental, that care is not a performance for an audience. It is a standard you hold whether anyone is watching or not.

That single moment in an empty warehouse explains more about what is broken in AI customer experience than any research paper I have ever read.

I have been living in Japan for over twelve years. And every time I look for the answer to what we broke, I find it in the same place. Not in a better model. Not in a smarter prompt.

In a Japanese philosophy that has been thinking about the human side of service for centuries, long before any of us had a chatbot to complain about.

It’s called omotenashi. And I think it’s the missing piece.

Omotenashi: The Philosophy That Was Thinking About This Long Before We Were

Here’s what’s interesting about Japan: this culture has been grappling with a version of the human-service problem for centuries. Not through technology, but through a set of deeply held values about what it means to truly serve another person, values refined through generations of practice in hospitality, retail, and everyday interaction.

To be clear: Japan is not a frictionless service utopia. Anyone who’s tried to cancel a contract here, navigate a bureaucratic form, or deal with a system that cannot deviate from its rules knows that rigidity is real. Japan can make you feel very processed when processes are involved.

But underneath the rigidity, there is an orientation, particularly in human service interactions, toward the other person’s well-being that goes beyond efficiency. That’s the part worth paying attention to. And it’s exactly what AI keeps leaving out.

What Omotenashi Actually Looks Like in Practice

Below I explore 3 of the main principles that drive Omotenashi that I always think back to when I see that AI went wrong:

Kikubari (気配り) - anticipating needs before they’re expressed

Kikubari is the practice of noticing what someone needs before they ask. In a traditional Japanese inn, your tea appears just as you’re thinking about wanting it. In customer experience terms, it means reaching a customer before they have to file a complaint. Today’s AI is almost entirely reactive, it waits for input. Kikubari is proactive. It asks “what will this person need next?” not “how do I respond to what they just said?”

Shitsurai (室礼)- the experience lives in the details

In Japan, the care given to a space before a guest arrives matters as much as the interaction itself. The temperature of the room, the placement of an object, what’s visible and what isn’t. In digital terms, this is the difference between an AI that technically answers a question and one that makes the entire interaction feel effortless. It shows up in edge cases handled gracefully, in context remembered across channels, in a system that doesn’t make you repeat yourself.

Kuuki wo Yomu (空気を読む) - reading what isn’t being said

Literally “reading the air.” Japan has a deep cultural practice of sensing the emotional state beneath an interaction, the hesitation, the frustration hiding inside a polite phrase, the thing someone needs but hasn’t found the words for yet. Most AI treats an angry customer and a calm one with exactly the same cadence. That’s not a technical limitation anymore. It’s a design choice. And it’s the wrong one.

Where This Goes Next

In the next three articles, I’m going to take each of these values, go deep, and provide practical frameworks for anyone building or leading customer experience in the age of AI.

What does it actually look like to design for kikubari? How do you build shitsurai into a conversational system? And what would it mean for an AI to genuinely read the air?

Technology nowadays is impressive. What it’s missing is an orientation toward the person, not the ticket.

We have the speed. What we need now is the soul.

End.

This is the first article in a series about Omotenashi and how this Japanese customer service philosophy can be the answer to a more human-centric AI. If you want to follow where it goes, subscribe below, and you won’t miss the next one.

In the meantime, these articles might be worth your time:

Or we can also connect on LinkedIn. I talk about these topics there too.

I have been working at Uber for 6 months, but Japan has been delivering food for more than 400 years.

That sentence says a lot about what I want to explore in this article.

The Bikes That Never Spill

I lived in Asakusa for five years. It’s one of Tokyo’s oldest neighborhoods, narrow streets, wooden shopfronts, the kind of place that makes you feel like you accidentally walked into the wrong century and then decided to stay.

Every morning, before the city fully woke up, I’d see them. Riders from local soba and ramen restaurants navigating the streets on bicycles and three-wheeled motorbikes, balancing trays of hot food without spilling a drop. No app, no GPS, no branded backpack. Just a person who knew every corner of the neighborhood, delivering food from a restaurant that knew its customers by name.

I watched this for five years and never stopped finding it remarkable.

But what stopped me the first time wasn’t someone balancing noodles on a bicycle. It was the bike itself.

Those delivery bikes in Japan are specially equipped with shock absorbers and stabilizers, specifically designed so the soup doesn’t spill, so the presentation arrives intact, so the customer receives not just food but the full experience the restaurant intended.

Think about that for a second. Someone, somewhere, looked at a delivery bike and thought: the suspension isn’t good enough for what we’re trying to do here. And then fixed it.

That is omotenashi made physical. A philosophy built into the engineering of a vehicle, not a value on a poster, not a training program. A shock absorber on a bicycle.

What Demae Actually Was

But Demae was never just logistics.

Food was delivered on the restaurant’s own dishes. Real plates, real bowls, not disposable containers. The rider would return later to collect them. The whole system ran on trust, between the restaurant and the neighborhood it served, between the rider and the customer who opened the door.

That trust wasn’t built through a rating system or a satisfaction survey. It was built over years of showing up, getting it right, and caring enough about the details that the customer never had to think about it.

Japan Grows Up With High Standards

To understand why Demae worked the way it did, you have to understand what Japanese customers expect from food, and where those expectations come from.

It starts young. Japanese children grow up with bento boxes prepared with extraordinary care. Not just nutritious but beautiful, carefully arranged, color balanced, thoughtfully presented. The message absorbed early is that food is not just fuel. It is an expression of care for the person receiving it.

Walk past any restaurant in Japan and you’ll see glass display cases outside the entrance showing exact replicas of every dish on the menu, perfectly crafted and precisely presented. Before you even walk through the door, the restaurant has already made you a promise.

By the time food delivery existed in Japan, the customer already knew what good looked like. The standard was set long before the service began. That’s the part most people miss.

What Delivery Looks Like When Hospitality Comes First

Most industries think about delivery as the last step of a transaction. You order, they bring it, done.

Japan never saw it that way.

In the Demae tradition, delivery was an extension of the restaurant’s hospitality. The rider represented the kitchen. The presentation of the food at the door was as important as the cooking itself. The relationship between the restaurant and its customer didn’t end when the order was placed, it continued through every step until the meal was finished and the dishes were collected.

The details are the message.

What the World Can Learn

I’m not saying the old way is always better and I am not saying Japan got it 100% right. The modern food delivery industry has brought extraordinary convenience to millions of people and created entirely new ways for restaurants to reach customers they never could before. That is genuinely valuable.

But there is something in the Demae tradition worth carrying forward. The question Japan has always asked, implicitly, culturally, without needing to write it in a strategy document, is not “how fast can we deliver?” It is “how does the customer feel when the food arrives?”

That is a different question. And it leads to different decisions. It leads to shock absorbers on bicycles, to real dishes instead of disposable containers, to riders who know the neighborhood well enough to navigate it without GPS.

Japan didn’t invent food delivery. It invented food delivery done with care.

400 years later, that’s still the hardest part to get right.

End.

About this newsletter

CX Japan explores what Japanese service philosophy can teach the world about building experiences that technology can enhance but never replace.

Written by Daniel Orenes Ferrandez, CX leader at Uber Japan with 12 years inside Japan’s most demanding companies.

If you’re not subscribed yet, join below, one article a week.

About the author

Daniel Orenes Ferrandez

I lead Customer Experience at Uber Japan. Before that I spent four years building conversational AI at Rakuten. I also speak at international conferences, including The Economist Corporate Network and the European Chatbot Conference. I have been living in Tokyo for 12 years.

→ Find me on LinkedIn
→ More at danielorenes.com

I have spent almost 12 years working in the Japanese market, 10 of them inside Japanese corporations. That time has shaped how I think about leadership more than anything else in my career. One topic I particularly like to research is executive presence and how differently it shows up depending on where in the world you are.

Most writing on this topic treat it as a universal thing, like there is one playbook everyone should follow. After working across cultures for over a decade, I realized this is not exactly how it works and in this article will analyze it based on my own experiences.

Subscribe now

What is executive presence?

EP is the ability to make people respect you, to portray a competent image so that your opinion actually gets taken seriously.

A senior position might get you in the room, but it does not guarantee any of that. I have sat across from people where it was clear their age or tenure was the only reason they were there. They had the title but somehow you could feel they did not really have the room. Nobody was leaning in or particularly engaged. The gap between having a position and actually commanding respect is what EP is really about.

Generally, EP can be divided in 3 areas:

Gravitas

Whether people take you seriously. This comes from how you make decisions, how consistent you are, and whether people sense you actually mean what you say.

Communication

Not just public speaking. How you run a meeting, what you do when you are not talking, and whether people feel heard after a conversation with you.

Appearance

How you show up, how polished you look, and what your digital presence says about you before anyone has even met you in person.

These three matter. But after a decade inside Japanese companies, watching leaders succeed and fail across very different cultural environments, I would add a fourth, which might me more important than the other three.

What most writing on this topic misses

EP is not a fixed standard. It is defined by the people around you and the context you are operating in.

In most Western corporate environments I have seen presence built through visibility. Speaking up, taking clear positions, projecting confidence. The room rewards people who fill space and speak with certain air of authority.

However, by living and working in Japan, I realized story is a bit different here.

Below area values that would be crucial to develop EP in Japan, but not necessarily might be important in other cultures:

Ma (間) — Silence

In most Western meeting rooms, silence is uncomfortable. People rush to fill it. In Japan it is the opposite. Ma is the deliberate use of pause and empty space, and it signals that you are thinking seriously before you speak.

A well placed silence in a Japanese meeting room carries more authority than a confident answer delivered too quickly. I have seen Western professionals talk themselves out of credibility simply by being unable to sit with a moment of quiet. Learning to be comfortable with Ma is one of the first and hardest adjustments for anyone coming from a Western professional environment.

Kenson (謙遜) — Humility

In Western cultures humility is often seen as a soft trait, nice to have but not particularly linked to authority. In Japan it is structural.

Kenson, the practice of humility and self-deprecation, is how you signal that you understand your place within a relationship and within a hierarchy. Presenting yourself as the smartest person in the room, speaking about your own achievements openly, or positioning yourself above others, even subtly, will quietly destroy your credibility.

The leaders I saw command the most respect in Japan were often the ones who deflected credit, acknowledged others first, and never made their own status the point of the conversation.

Kuuki wo Yomu (空気を読む) — Reading the Atmosphere

Literally translated it means reading the air. In practice it means being able to sense what is happening in a room without anyone saying it out loud. What the group is feeling, what has not been said but needs to be acknowledged, whether the moment is right to push forward or to pull back.

During my time in Japan I came to realize how important this skill can be. More than one time I have heard colleagues saying this like for example: “Ah, Tanaka-san, yes he is good, but he still needs to improve his air reading skills when meeting with clients”.

Japanese professionals spend years developing this skill and it is treated as a serious indicator of emotional and social intelligence. In a Western context we sometimes call this reading the room, but in Japan it goes much deeper. Missing these signals, pushing through when the room has already decided, or failing to notice discomfort that nobody is voicing, are some of the most common mistakes I watched Western executives make.

Nemawashi (根回し) — Consensus

In Western environments consensus can feel slow. Why spend weeks aligning everyone when a decision can be made in a meeting? In Japan consensus is not bureaucracy. It is the foundation that makes execution possible.

Nemawashi refers to the process of quietly laying the groundwork before a formal decision, talking to stakeholders individually, understanding concerns in advance, building alignment before anyone sits down in the same room.

By the time a decision gets made formally, everyone already knows what is coming and has had the chance to shape it. Leaders who skip this process and try to push decisions through in the room are seen as careless at best, and arrogant at worst. The most effective leaders I worked with in Japan had already done the real work before the meeting even started.

In my time working in Japan, I have watched confident, capable Western executives walk into Japanese boardrooms and lose the room completely, not because they lacked skill, but because they were showing the wrong version of presence. Loud where quiet was expected. Decisive where patience was the signal or direct where indirectness was the language.

I have also watched much quieter leaders hold a room of senior Japanese executives simply by knowing when to stop talking.

Neither style is wrong. But only one works in a given room, and I learned that from experience.

How to actually develop it

You cannot self-assess this one. EP lives in the perception of others, and what you think you project and what people actually receive can be very different things.

Ask the people around you, but ask the right question. Not a generic “how am I doing” because that gets you nothing. Ask how they feel after a meeting with you. Whether they leave with clarity or still unsure. Whether they felt listened to or like you were just waiting for your turn to speak.

One honest answer to that is worth more than any framework.

And if you are working across cultures, how you gather that feedback is itself cultural. In Japan, nobody will tell you directly that you misread the room. I had to build relationships where honesty could eventually come through, and learn to pay attention to what was not being said as much as what was.

A conversation that stayed with me

Last month, I participated in the European Chatbot Conference. There were plenty of discussions around GenAI, automation, and the future of customer interactions, but one comment, almost casual, stayed with me.

Oyikansola, a PhD researcher in AI and Data Science, with whom I had the pleasure of sharing a panel, mentioned that in some parts of Africa, people don’t speak just one language and instead, they mix them. English, local dialects, sometimes multiple languages within the same sentence, switching naturally without even thinking about it.

This made me think about something I have not thought too much about before. LLMs can understand multiple languages, even if we mix them. However, can they also answer in the same way, mixing multiple languages and, most importantly, in a way that feels right?

This is how people actually communicate

Across the world, millions of people communicate by blending languages in ways that feel completely natural to them. In Africa, some countries mix multiple languages or dialects in one sentence. In Mexico, conversations often move fluidly between Spanish and English, especially in professional or urban environments. In the Philippines, Taglish, and in Singapore, Singlish are part of everyday communication.

Even in more structured environments like Japan, particularly in business or tech contexts, it’s common to hear sentences that incorporate English terms seamlessly into Japanese conversations.

The mismatch with conversational AI

Now compare that to how most conversational AI systems are designed. They tend to assume a much simpler model of communication, one user interacting in one language at a time.

Users are often expected to choose a language up front, and from that moment on, the system assumes consistency. English, Spanish, and Japanese are each treated as separate and self-contained.

But real conversations do not follow those boundaries and sometimes languages are mixed.

For a long time, conversational systems struggled with this reality because they were built on rigid structures such as rules, intent classification, and predefined flows. Now with LLMs mixing languages is not a foreign thing anymore. However, is not as easy at seems.

LLMs changed something, but not everything

This behavior is not new. In fact, it has a name: code-switching.

Code-switching describes how people naturally move between languages within a conversation, often within the same sentence. Not as an exception, but as a normal way of communicating.

For Conversational AI, it is not the same to understand code-switching as it is to respond to it, and even less to design an experience around it.

Models like ChatGPT or Gemini can already understand mixed-language inputs quite well. But when it comes to responding, if languages are mixed, something still feels slightly off.

Accuracy vs authenticity

This leads to a more fundamental question: should conversational AI aim for accuracy, or authenticity?

If the priority is accuracy, the path is relatively clear. Responses are clean, grammatically correct, and consistent, typically within a single language. This approach makes systems easier to evaluate, scale, and control, particularly across multiple markets.

But accuracy alone can create interactions that feel distant. The system works, but it does not fully resonate with the user.

On the other hand, aiming for authenticity means trying to mirror how users actually communicate. This includes allowing mixed languages, adapting to tone, and reflecting conversational patterns more closely.

The challenge is that current systems are not yet fully capable of doing this in a reliable way. While they can imitate mixed-language communication, the result is not always convincing. When authenticity is attempted but not achieved, it can feel forced or artificial, which can be just as problematic as being overly rigid.

The trade-off is real. Choosing a single language is often the safest and most consistent option, but safety comes at a cost. The moment users are forced into a single, “correct” way of speaking, the system stops adapting to them, and instead asks them to adapt to the system.

Why this matters for CX

This is where the discussion moves beyond language and into customer experience.

From a CX perspective, especially when operating across multiple markets, supporting a language is not the same as understanding how that language is used in practice. The differences are often subtle, but they have a meaningful impact on how users express intent, emotion, urgency, and even politeness.

For example, Mexico and Spain may share the same language on paper, but the way it is used in conversation can differ significantly. Similarly, communication styles in Philippines differ from standard English in ways that go beyond vocabulary.

If conversational AI systems ignore these nuances, they risk delivering experiences that are technically correct but fail to feel natural or inclusive to the user.

Designing for real conversations

This does not mean that every chatbot should immediately start mixing languages. Instead, it highlights the need for more intentional design decisions.

Companies should begin by designing for how users actually speak, rather than how systems expect them to speak. This includes testing with real user inputs, which often involve mixed-language scenarios, rather than relying on idealized or standardized language.

It also requires making conscious decisions about when clarity should take priority over naturalness, and when a more conversational approach adds value to the experience.

Perhaps most importantly, this should be treated as a deliberate design choice. In many cases today, it is not a choice at all, but rather a side effect of how systems have traditionally been built.

Inclusivity is more than language coverage

We often frame inclusivity in conversational AI in terms of how many languages a system supports. While this is important, it only addresses part of the challenge.

Inclusivity is not just about language coverage. It is about communication, and more specifically, about aligning with how people naturally express themselves.

The reality is that there is no perfect answer. Whether you decide to prioritize accuracy, authenticity, or a balance between the two, there will always be trade-offs. In some cases, consistency and clarity will matter more. In others, sounding natural and familiar will have a bigger impact on the experience.

What matters most is not the specific choice you make, but the mindset behind it.

Designing conversational AI for multiple markets requires acknowledging that users are not uniform, even when they share the same language. Each market has its own nuances, habits, and ways of communicating, and those differences should be considered from the beginning.

When those scenarios are explored upfront, decisions around language, tone, and behavior become intentional. Because instead of taking the safest or most convenient path by default, you are making conscious trade-offs based on the experience you want to create.

In the end, inclusivity is not about trying to perfectly replicate how everyone speaks. It is about recognizing those differences, designing with them in mind, and making deliberate choices rather than one-sided assumptions.

Anything else is not really inclusive. It is just standardized.

NRF is the biggest retail event in the world, where the latest trends shaping the industry are announced and validated. I used to attend in person in New York at the Javits Center, and every time it was a great experience.

It was the perfect place to discover emerging trends, meet potential partners and clients, and understand where retail is heading next.

NRF happens every year, and I’ve made it a tradition to summarize the key highlights.

Here are the 5 highlights that stood out the most at NRF 2026 and a final remark on what I found to be one of the most remarkable part of this year’s NRF: Gemini Enterprise for Customer Experience and Customer Experience Agent Studio.

Subscribe now

1. AI is moving from back-end optimization to the front-end customer journey

AI has been in retail for years, mostly in forecasting and supply chain optimization.

However, from this year things are changing and AI is now becoming the consumer-facing interface. Shoppers are slowly transitioning away from traditional search bars and toward AI assistants to discover products, generate recipes, and build dynamic shopping lists.

One example of this is Kroger, an early adopter of Google’s Gemini-powered shopping solutions. Kroger is deploying a personal shopping assistant using AI to bridge the gap between “I want to cook this tonight” and “Here is what’s actually on the shelf right now.”

More Information: Kroger

2. Agentic commerce is real, and checkout is becoming invisible

This year, one of the biggest trend was agentic commerce. In other words, AI that can take action and complete purchases.

The strongest example was the Walmart + Google partnership, which will allow users to discover Walmart and Sam’s Club products directly inside Gemini, including real-time availability and pricing. Loyalty accounts can even be linked for personalized recommendations.

More information: Walmart

3. Omnichannel 3.0: Moving to “Unified Commerce”

mThe industry has officially moved past “Online vs. Offline.” The trend is now Unified Commerce, where inventory, payments, and customer data exist in a single, fluid pool.

• Endless Aisle: Retailers like Target and Wayfair are using “endless aisle” technology to let customers scan a floor sample in-store and have it sent to their home immediately, ensuring they never lose a sale due to floor space constraints.

• Smarter Pricing: The value strategy is shifting. Companies like Revionics demonstrated how retailers are using analytics to decide exactly which products need competitive pricing and which promotions are actually driving profitable growth.

4. The most valuable retail tech is empowering frontline teams

While high-tech consumer AI gets the headlines, many retailers at NRF 2026 focused on “operationalizing” the physical store. The goal is to make the in-store experience as data-driven as a website.

• Bite-Sized Training: Sprouts Farmers Market showcased how they’ve gamified employee training. By replacing long manuals with 3-minute “daily reinforcement” games on mobile devices, they’ve seen a 94% engagement rate. Better-prepared staff leads to smoother execution on the floor.

• Computer Vision (CV): Retailers are increasingly using ceiling-mounted CV to detect out-of-stocks in real-time, ensuring that the digital inventory matches the physical shelf perfectly.

5. Value strategy is shifting from discounting to smarter pricing

With consumers still price-sensitive, value was a major topic. But NRF 2026 showed retailers don’t want a race to the bottom. Instead, they’re becoming more surgical with pricing, promotions, and private label strategy.

An example discussed widely was Revionics, highlighting how retailers can use analytics to decide which products must be competitively priced, and which promotions actually drive profitable growth.

The Game Changer: Google’s Big Move Into Retail Customer Experience

One of the biggest announcement at NRF 2026 came from Google.

Google launched Gemini Enterprise for Customer Experience, designed to unify shopping and customer service into one AI-driven experience. It also introduced Customer Experience Agent Studio, allowing companies to build and deploy AI agents that can take autonomous action.

But the true breakthrough was the launch of the Universal Commerce Protocol (UCP), developed with major partners like Shopify, Walmart, Target, Etsy, and Wayfair.

UCP is meant to standardize agentic commerce, enabling AI assistants to guide consumers from discovery all the way to checkout, without forcing them to jump between apps. It also ensures retailers can keep control of the customer relationship while still benefiting from Google’s AI ecosystem.

In other words, Google is trying to build the “infrastructure layer” for AI-powered commerce.

Career progression is a topic I am truly passionate about. At work, I find deep joy in sitting down with my team members and talking about their career plans, ambitions, doubts, and goals. And it’s the same on LinkedIn and with friends.

This is a topic that isn’t foreign to anyone. To some degree, we all care about it, in one way or another.

In the past few months, I’ve noticed a clear trend in my conversations with colleagues, friends, and connections: the impact of AI on people’s careers. Questions like, with AI growing at this pace, will my job be automated? Should I start learning new skills? Should I look for a job that won’t be impacted by AI? Or should I look for a job that uses AI heavily and go all in on the AI world? These are all valid questions. And unfortunately (or fortunately, depending on how you see it), there is no straight answer.

In this article, I’d like to explore this topic and share my opinion and conclusions.

Subscribe now

The Myth of the Linear Career Path

For a long time, many of us grew up believing that careers were meant to be linear. You study something, you get a job in that field, you climb step by step, and eventually you reach some kind of “destination.”

And maybe at some point in the past that was true, or at least it was a convenient illusion we could all agree on. But today, it’s pretty obvious that linear career paths are no longer the norm.

Careers now look more like messy maps than ladders, shaped by unexpected opportunities, shifting industries, and the constant reinvention of what “valuable skills” even mean.

AI Is Reshaping Careers Faster Than Ever

And if there’s one force accelerating this shift, it’s AI. We’re living through a moment where the job market is changing at a speed most people have never experienced before.

New roles are being created almost overnight, while roles that were once stable are becoming less relevant or rapidly transforming. Some might argue that this has always happened, that every technological wave has temporarily disrupted industries, from the rise of computers to the internet to mobile. That’s true. But what makes AI feel different is that it’s not only transforming one sector, it’s challenging almost every industry at the same time, from finance to healthcare to education to customer service.

This isn’t a temporary wave. It feels like a deeper shift that is rewriting how work itself is defined.

Don’t Let Fear Be Your North Star

Because of that, I’ve been hearing the same theme repeatedly in conversations lately: fear. I’ve spoken with many people who are trying to change jobs, pivot careers, or jump into “AI-safe” roles mainly because they’re worried about what will be automated next. And honestly, I get it. It’s natural.

Wanting to protect your future is not irrational, it’s responsible. But fear should not be your north star. If your career decisions are driven primarily by avoiding what AI might replace, you may temporarily reduce risk in the short term, but what happens when that new “safe” path gets automated too? AI is evolving too fast for career planning to be based purely on escape routes.

The Most Underrated Skill: Knowing Yourself

But to me, the bigger issue isn’t even the automation risk. The real mistake I see is that many people are doing massive external effort, but not enough internal work. We spend endless hours tailoring CVs, optimizing LinkedIn, researching companies, networking, and preparing for high-pressure interviews and case studies.

Yet we rarely invest the same level of intensity into understanding ourselves. Do we really know our values? What kind of work energizes us? What environments bring out our best performance? What we’re naturally good at, and what we’re forcing ourselves to be good at? In an era where skills are constantly being redefined, self-awareness becomes one of the few stable competitive advantages.

This is something Japan taught me in a way no career book ever has. Japanese culture has a concept called ikigai, roughly translated as "reason for being." It sits at the intersection of what you love, what you're good at, what the world needs, and what you can be paid for. It's not a career framework. It's a life framework. But after 12 years watching Japanese professionals navigate their careers, I've noticed something: the ones who stay relevant through disruption aren't the ones who chase the safest role. They're the ones who know exactly what they're for. That kind of clarity is harder to automate than any technical skill.

AI Should Be a Variable, Not the Whole Equation

Of course, it’s still smart to evaluate how AI might impact your career. Ignoring the trend completely would be naive. But AI should be one variable in your decision-making, not the entire equation.

The majority of your effort should go into understanding your strengths, your motivations, and the kind of impact you want to create. Because if you build your career around who you are, rather than what you’re afraid of, you’ll be far more adaptable no matter what AI disrupts next.

Build a Career That Evolves With You and.. Have Fun With it

In the end, the future belongs to people who can evolve without losing themselves. Careers will keep shifting, roles will keep changing, and uncertainty will remain part of the landscape.

But if you anchor your choices in your values, your curiosity, and your unique human strengths, you won’t just survive the AI era, you’ll shape it. Fear is useful as a warning signal. But it’s a terrible compass.

It’s fair to say we all want to get better at what we do. We want to grow, improve, and feel we’re moving forward. When we look for inspiration, we often look at CEOs of big companies and assume they’ve mastered it. While there’s much to learn from them, I believe the real key to growth lies in the effort we each put into learning and improving a little every day.

That’s why I decided to dive into CEO Excellence and the more I read, the more I recognized patterns I had already lived, working inside Japanese corporations for over a decade, where many of these leadership principles aren't frameworks. They're just how things are done. The book analyzes six competencies that define great CEOs. I will not say this is a light read, it is dense, but it’s full of valuable lessons. In these two articles, I share what I learned and how anyone, regardless of title, can apply these mindsets to grow both professionally and personally.

In Part 1, I explored the first three: setting the direction, aligning the organization, and mobilizing leaders.

Link to Part 1: https://substack.com/home/post/p-159112191

In this second part, I’ll go through the last three: engaging the board, connecting with stakeholders, and managing personal effectiveness, adding some of my own experiences along the way.

Engaging the Board

Personal Experiences

In the first Japanese company I worked for, I was taught that making a habit of sharing progress regularly was essential. At first, I didn’t quite understand why it was needed. I thought results should speak for themselves. But later, when I started to hold management positions, I realized the importance of engaging regularly with stakeholders.

Sometimes I would assign a project or task to a team member and only check in near the end. It worked at times, but other times I’d find out about challenges too late. I learned that this wasn’t about control, it was about alignment. Regular check-ins ensure that everyone stays updated on changing priorities or variables. They also require managers to be concise and clear about goals and expectations so team members understand exactly what success looks like.

From a manager’s perspective, I also learned how much communication can get lost between leadership and operational teams. Acting as a bridge between both sides became one of the most valuable (and challenging) things I have experienced. It helped leadership understand the challenges my members were facing with clients, and helped my team understand leadership’s direction and reasoning. Finding that middle ground where everyone moves forward together is what true engagement means. Also, I would say what makes middle-management work particularly rewarding.

For ICs

For individual contributors, engaging your own “board” means proactively communicating with your manager or stakeholders, even if they haven’t asked. Don’t wait for them to check in. Share progress, flag risks early (very important!), and invite feedback before decisions are final. It shows accountability and builds trust.

For Managers

Create a rhythm of open communication. Encourage updates not as a formality, but as a way to strengthen collaboration. And when your team reports progress, listen closely and guide them with clarity.

Connecting with Stakeholders

Personal Experiences

In large organizations, you rarely operate alone. At Rakuten, especially while working on AI-related projects, I learned that aligning with stakeholders is crucial to make things work. I often tell my team that even the best idea won’t move forward without alignment.

For example, when we developed GenAI solutions for different business units, progress didn’t come from the technology alone. It came from how well we involved everyone, business teams, technical groups, and compliance. Taking the time to align everyone on the same purpose created momentum. Without that alignment, even a brilliant idea can easily stall.

For ICs

For individual contributors, connecting with stakeholders means understanding how your work affects others. Instead of focusing only on your tasks, take a moment to see how your outputs impact another team’s work or the overall customer experience. It’s not just about doing your part, but about contributing to a shared success.

For Managers

Help your teams see beyond their own boundaries. Bring customer stories or stakeholder feedback into team discussions. When people understand how their work fits into a broader ecosystem, motivation and accountability naturally increase.

Managing Personal Effectiveness

Personal Experiences

To me, this is the most important mindset, and that’s why I wanted to close with it. Managing yourself is the foundation of leadership. Over time, I’ve learned that if you don’t manage your time and energy, you can’t lead others effectively.

Every day, I dedicate a block of time early in the morning to what I call Focus Time. At first, honestly, I used it more as a backup slot than a real focus time, something I could replace if an urgent meeting appeared. But recently, I started treating it less as a flexible time slot and more as non-negotiable, as important as any other meeting. During that time, I reflect, prioritize, and think from a higher perspective. It helps me separate what’s truly important from what’s merely urgent.

For ICs

For individual contributors, managing personal effectiveness means protecting time for deep thinking and learning. Don’t let your calendar control your day. Spend a few minutes every morning reviewing what’s most impactful and what can wait.

For Managers

For managers, it’s about setting the example. If you’re always running from one meeting to another without reflection time, your team will follow that pattern. Protecting your focus time shows others that discipline and self-management are part of high performance.

Final Thoughts

The more I grow in my career, the more I realize leadership isn’t granted by title, it’s earned through consistency. And that consistency comes from staying curious, being intentional, and never stopping the process of learning and reflection.

Really, you don’t need permission to lead. You just need the habit of doing it every day. Remember, you’re the captain of your own boat, you decide the direction, the pace, and the harbors you want to explore along the way.

I’ve spent years watching Japan figure out problems that the rest of the world is only now starting to ask. Online grocery is one of them. Five years ago I wrote about the early days, when adoption was low, cash was still king, and ordering groceries on your phone felt slightly exotic in Tokyo. A lot has changed. But the most interesting part isn’t the numbers. It’s what Japan’s approach to service is revealing about where this industry is actually heading.

Article Here.

Back then, adoption was limited, convenience stores ruled daily life, and online grocery shopping felt like a side bet.

Today, the sector has grown into a major retail force. But the real story isn’t just logistics or technology, it’s how customer experience (CX) is being redefined.

1. Market Scale & Trajectory

Japan’s online grocery market has moved from niche to mainstream. In 2024, the market was valued at USD 40.4 billion and is projected to soar to USD 271.9 billion by 2033, growing at a 23% CAGR. This represents a huge leap from 2020, when penetration stood at just 16.6%, far behind Singapore’s 41% and Hong Kong’s 36%. The pandemic acted as a catalyst, exposing millions of first-time users to app-based grocery shopping and embedding new consumer habits that are still present today.

2. New Winners & Competition Dynamics

The competitive landscape looks very different now than it did five years ago:

• Uber Eats has become one of the most recognizable delivery platforms in Japan. Originally focused on restaurant meals, it has expanded into groceries with pilots such as My Basket in Tokyo. Uber’s logistics expertise, brand trust, and ability to connect rides, food, and groceries under one ecosystem give it a unique edge.

• Rakuten Seiyu Netsuper has strengthened its omnichannel presence, reinforced by KKR’s 2025 acquisition of Seiyu and its integration into the Rakuten Points loyalty ecosystem.

• Amazon Fresh continues to leverage Prime membership and advanced logistics to secure share.

• Aeon × Ocado are betting on automation, with robotic fulfillment centers planned by 2027 to scale same-day delivery.

• Wolt and Foodpanda experimented by expanding into regional cities rather than Tokyo first, though Foodpanda has since scaled back globally.

• Temu, Shein, and soon TikTok Shop are reshaping consumer expectations with a focus on aggressive pricing and convenience.

What’s clear is that competition is no longer about speed alone. The winners will be those who deliver the most trusted, integrated, and habit-forming experiences.

3. How Consumers are Shopping Differently

Consumer behavior has also shifted dramatically. Japanese shoppers are now comfortable with cashless payments, driven by the boom of PayPay and mobile wallets like Suica and Pasmo. Urban consumers are moving away from large weekly stock-ups to smaller, more frequent “micro-basket” deliveries. And perhaps most importantly, trust, reliability, and precision remain central to how customers evaluate services.

• Cashless adoption: Digital payments are now mainstream, reducing friction at checkout.

• Shopping patterns: Smaller, frequent deliveries are increasingly common, especially in cities.

• Service expectations: Freshness, packaging quality, and punctuality are non-negotiable.

A personal story illustrates this shift: I read a LinkedIn post from a customer who ordered pasta, only to find the sauce missing. Instead of just issuing a refund, the platform refunded the cost and delivered the missing item. That small act transformed frustration into loyalty. In Japan, service recovery can matter as much as the service itself.

4. What CX Leaders Should Focus On

The next phase of growth in online grocery won’t be defined by logistics alone but by the quality of customer experience. Precision in delivery matters more than ever, with customers expecting real-time updates and narrow delivery windows. Personalization, powered by AI, will allow platforms to predict reorders and even offer natural-language interfaces for shopping. Ecosystem integration, whether through Rakuten Points or PayPay, is becoming a decisive factor in building habits. And finally, success will depend on how well platforms support their partners, retailers, couriers, and delivery staff, because their experience ultimately defines the customer’s.

Service recovery also plays an important role. Mistakes will happen, but the way a company responds to them can either erode trust or deepen it. In a market like Japan, where reliability is expected, even small gestures can turn errors into loyalty.

Rethinking How We Measure CX

For years, metrics like NPS, CSAT, or FCR have been considered the gold standard for measuring customer experience. They still matter, but relying on them too heavily risks giving a false sense of progress.

Automation is a good example. Many companies highlight increases in auto-resolution rates as evidence of improvement. But if total support volume remains unchanged, then the root causes of customer issues are still unresolved. In practice, the company has become better at deflecting questions but not at eliminating them.

True CX leadership means going deeper. It requires analyzing the sources of customer contacts, understanding pain points at their root, and working across the business to eliminate them. The future of measurement is not about asking, “How much can we automate?” but rather, “How well do we understand our customers so we can solve problems before they arise?”

5. Technology on the Streets: Robots Enter the Scene

Technology isn’t just shaping the back end of online grocery, it’s now visible on the streets of Tokyo. In 2025, both Uber and Rakuten rolled out delivery robots in central locations: Uber in Nihonbashi, and Rakuten in Toyosu. These pilots are more than a showcase of innovation; they represent a test of how automation can blend into the daily lives of Japanese consumers.

For customers, the potential value is clear: reliable, contactless delivery that fits urban lifestyles. For companies, robots offer the promise of cost efficiency and scalability at a time when courier availability and labor costs are ongoing challenges. But the true measure of success will again come back to CX. Robots must not only deliver on time, but also meet Japan’s high standards for safety, courtesy, and convenience. A robot that arrives with the right groceries, at the promised time, and in good condition can reinforce trust. A misstep, a wrong item or a confusing interaction, could just as quickly damage it.

In this sense, technology is not the answer in itself. It’s a tool. Its role is to enhance the human experience, reduce friction, and allow companies to focus on fixing the root causes of customer pain points.

Uber Delivery Robot

Rakuten Delivery Robot

6. Looking Ahead

Japan’s online grocery sector reflects the country’s unique DNA: loyalty-driven, precision-focused, and convenience-oriented. Technology and new entrants will continue to accelerate growth, but the winners will be those who design end-to-end experiences that blend trust, empathy, and operational excellence.

Five years ago, the big question was whether online grocery shopping would take off. Today, the question is: Who will set the standard for CX in this space?

If you know me, you know I’m a retail geek. I find a lot of joy when I walk into a new supermarket or convenience store. I like to explore the tech they use, new goods I have not seen before, and their distribution. Also reminds me of my early days in Japan working at a retail-tech Japanese company.

On June 23, Lawson and KDDI opened their “Real×Tech LAWSON” flagship at Takanawa Gateway City, showcasing how real omotenashi and cutting-edge technology live together inside a convenience store.

In this article, I will explore Japan’s retail digital transformation, dive into what’s happening at this Lawson store, and think about what might come next.

The Key to Retail’s Future

Retailers around the world are pouring ever more into technology. Gartner data shows global IT spending hitting a record US $5.43 trillion in 2025, up nearly 8 percent from 2024, much of it driven by AI and automation initiatives . Narrowing to retail specifically, experts forecast that global retail tech investment will climb past US $260 billion by 2027.

By 2050, Japan’s population is expected to drop to around 108.3 million, a decline of roughly 14 percent from 2020 levels, as fewer births and an ageing workforce reshape the retail landscape.

Combine that with rising labour costs and changing consumer habits, and you get a perfect storm pushing retailers to test and implement new technologies at their stores.

For example, a month back, I walked into a supermarket in Tokyo where carts were powered with tablets where you could self-scan the products either with your phone or a scanner. You can also find cashier‑free stores tucked in Tokyo alleys, supermarkets trialling robotic shelf‑stocking, and chains rolling out smart shopping carts with built‑in scanners. All of these point to one thing: digital transformation isn’t a nice‑to‑have; it’s a must.

Lawson’s Takanawa Store: A Living Lab

Lawson has been on this journey for some time, partnering with KDDI, rolling out mobile app integrations, and piloting small‑scale tech stores since 2019. After KDDI took a 50 percent stake last year, they’ve doubled down. At Takanawa Gateway City, they’ve packed in experiments across four key areas:

• AI‑powered signage: touch a shelf label and the overhead display shows product stories, sustainability info or limited‑time promos, helping you shop with more confidence.

• Smart‑city integration: live feeds on weather, train delays and local events stream to in‑store screens, turning the store into a community hub. This digital signage can also use information from Takanawa Gateway City (Urban OS) to display real-time information such as weather, train delays, and congestion levels.

• Robotics and AI ops: Automated bots handle beverage restocking, cleaning, and even frying “Karaage‑kun,” while AI agents crunch tag‑and‑camera data to forecast staffing needs, aiming for a 30 percent cut in store operation time by 2030.

An automatic cleaning robot that cleans the store in about an hour. It does everything from suction to wiping with water.

• Remote service and AI Ponta: video booths connect you on demand to specialists in telecoms, finance or healthcare, while a friendly Ponta robot offers chat, fortune‑telling and product tips through natural conversation.

Where Retail Tech Goes Next?

This flagship is part showroom, part lab. Over the coming months, we’ll see which tech sticks aand how quickly will Lawson scale up these features across its 14,000 stores.

A couple of years back, when I was in NY on a business trip, we visited an Amazon Go store and honestly, I feel that Amazon was already doing quite good at the time. If the goal is to make the store unmanned, this approach of Lawson can work. However, if the goal is to gather relevant consumer data, get to know each of your customers, and personalize customer journeys, Amazon's approach was closer to this.

Why? Because apart from being unmanned, Amazon Go would recognize each client inside the store, your walking patterns, where you looked, which products you bought, which ones you took from the shelf but did not buy, and left again. All of this relevant data is used to personalize future interactions with the user, not only in the store but also online.

Digital transformation in Japanese retail is racing ahead, and Lawson’s Takanawa store gives us a glimpse of “future convenience” today. I’ll be keeping an eye and providing new updates.

Not long ago, Klarna made headlines with a bold move. The company claimed it had successfully replaced its entire customer service team with AI. No more waiting on hold. No more human agents. Just instant, automated responses.

But recently, Klarna admitted they were wrong.

That kind of honesty is rare, especially in tech. It’s also valuable. Because what happened at Klarna is more than just a failed experiment. It’s a mirror for any business thinking about how to use AI in customer experience.

This article isn’t about criticizing Klarna. It’s about learning from their journey.

Why Customer Service Needs a Different Playbook

Customer service sits at the intersection of emotion and information. It’s not just about giving the right answer. It’s about how the answer is delivered. Tone matters. Timing matters. And so does trust.

One wrong answer from a chatbot in marketing? Not ideal, but manageable.

One wrong answer in customer service? That could mean losing a customer. Or worse, losing trust. And as we’ve seen recently with some banks, that can cost millions in reputation damage.

What makes this space hard to automate isn’t just the volume of queries. It’s the unpredictability. The long tail of questions. The frustrated tone behind a refund request. The panic in a user who is locked out of their account.

AI doesn’t do well with emotion. At least not yet. And customers feel the difference.

That’s why companies need a different playbook when it comes to automating support.

Start at the Front Line

In Japan, there’s a concept called Genba, the real place where the work happens. For customer service, that’s your frontline team. The agents who speak to users every day.

They know where the friction is. They know which requests are simple and which ones spiral into complex issues. They know which tools work and which ones just slow them down.

So before you bring in a chatbot or build a workflow, go to your Genba. Listen. Learn. And then design your automation strategy based on real-world needs.

Here’s a simple framework to start with:

• Automate the obvious. Password resets, delivery tracking, and FAQs. These are perfect for AI.

• Personalize the tricky. If AI handles it, make sure it pulls the right data so the answer feels tailored.

• Keep the humans. For emotionally charged, urgent, or complex requests, real people still do it best.

3 Key Learnings from Klarna’s AI Shift

1. Don’t Automate in a Vacuum
AI should be built with your support teams, not instead of them. Your agents understand the nuances that bots miss. Bring them into the design process early.

2. AI Isn’t a One-Size-Fits-All Tool
Just because AI works well in one part of the business doesn’t mean it will in another. Customer service demands accuracy and empathy. That’s a unique mix. Don’t assume your existing AI setup can just be repurposed for support.

3. Human Experience Is Still a Differentiator
People remember how you made them feel. A fast response is good, but a human response, at the right moment, is unforgettable. That’s how loyalty is built.

The Future Is Hybrid

AI is here to stay. It’s not a question of whether companies should use it in customer service. It’s a question of how.

The best setups won’t be AI-only or human-only. They’ll be smart hybrids. AI for scale and speed. Humans for context and care.

And most importantly, a strategy that starts with listening, both to your customers and your team.

Klarna’s story is a reminder. Just because something can be automated doesn’t mean it should be.

The companies that get this right won’t just save money. They’ll build trust. And in the long run, that’s worth more than any short-term efficiency win.

But what if I tell you that not much has changed?

Throughout history, humans have looked for trusted sources to answer life's big questions. In early societies, people turned to shamans or elders for guidance on health, survival, and community decisions. Religious books have also been used as moral and value guidance. During the Middle Ages and the Renaissance, scholars like Aristotle or Confucius were sought after for their teachings.

One of the most interesting forms of knowledge seeking happened during ancient Greece, in which the oracle of Delphi was recognized as one of the ultimate sources of knowledge, and many important people, including politicians and even a friend of Socrates, would visit the oracle asking for guidance.

Looking at the present, I don´t think I am exaggerating if I say that no one can live without the internet, and when someone wants to know something, ChatGPT is the default way to go. With the rise of LLMs, apps like ChatGPT or Gemini, powered by are becoming the new oracles.

In this article, I explore what the Oracle of Delphi and LLMs have in common and end with some thoughts on the best ways I believe we should utilize these powerful tools.

Meet the original chatbot: The Oracle of Delphi

The Oracle of Delphi, known as the Pythia, was a high priestess at the Temple of Apollo in ancient Greece. She entered trance-like states that were believed to be divinely inspired, delivering cryptic messages to those who came from all corners of the Greek world.

Her prophecies shaped major political decisions, military campaigns, and philosophical debates. Even famous figures like Alexander the Great and the philosopher Socrates are said to have sought her guidance. Though her words were often vague, their power lay in interpretation.

Today, in an age dominated by data and machine learning, we find ourselves engaging with a new kind of oracle: large language models. And just like the Pythia, these systems reflect back not only information, but our hopes, fears, and the way we choose to ask.

Similarities of the oracle with LLMs (Large Language Models)

1. Ambiguity is a feature, not a bug

• The Oracle of Delphi gave cryptic, poetic answers open to interpretation. LLMs, while seemingly clearer, often respond in ways that depend heavily on how the user phrases the prompt.

• In both cases, the responsibility to find meaning rests with the human. Misinterpretation can lead to poor decisions.

• Socrates’ story is a perfect example; he didn’t take the Oracle’s words as fact but questioned their meaning for years. LLMs invite the same kind of interrogation, if we’re wise enough to do it.

2. Garbage in, garbage out

• At Delphi, what you asked and how you asked it mattered. Vague or leading questions could get you unexpected answers.

• LLMs function the same way: well-crafted prompts lead to better, more useful output. Poorly framed ones can yield confusing or biased replies.

• Prompt engineering today works the same way. The more precise and elaborate your question to an LLM, the better the response you will get.

3. We still worship at the temple of authority

• The Pythia was seen as a mouthpiece of the gods, her words held immense authority. People made life-and-death decisions based on her responses.

• LLMs, while not divine, often carry similar perceived weight. Users trust them for advice, facts, and even emotional support.

• In both cases, there’s a danger in over-trusting the output without questioning its source, limits, or intent.

4. Reflection of Human Bias

• The Oracle’s responses were shaped by the priests who interpreted her words and the cultural lens of the time.

• LLMs are trained on human-created data, absorbing and amplifying our collective knowledge and our biases.

• Neither is neutral. Both reflect the values, fears, and worldviews of the people who surround them.

Socrates vs LLMs

In my point of view, one of the most interesting stories about the Oracle of Delphi is about the philosopher Socrates. According to Plato’s Apology, a friend of Socrates asked the Oracle if anyone was wiser than him.

The Pythia replied that no one was wiser than Socrates.

Socrates was surprised by this answer, as he didn’t see himself as wise at all.

He didn’t take the Oracle at face value.

When told “no one is wiser”, he went out and questioned everyone. Politicians. Poets. Craftsmen. He found out that while they all believed they had knowledge, they couldn’t explain it clearly.

His insight? He wasn’t wise because he had the answers. He was wise because he knew he didn’t.

That’s how you use ChatGPT. Not as the truth but as a sparring partner. Similarly, large language models today don’t give us final answers. They reflect back our assumptions, provoke more questions, and, when used well, push us to think more deeply about what we truly know.

What can we learn from ancient knowledge?

We’ve always turned to oracles in times of uncertainty, and now we carry one in our pocket. But just like the ancients knew, blind trust is risky. Large language models are powerful, but they’re tools, not truths.

If we don’t sharpen our questions, recognize our biases, and stay curious, we risk repeating the same mistakes, only faster, and at scale. The future doesn’t belong to the oracle. It belongs to those who learn how to listen, interpret, and choose wisely.

Earlier this month, I visited Retail Tech Japan 2025 at Tokyo Big Sight, one of the biggest retail events in Asia. It’s one of those events I try not to miss, partly because I used to attend every year with the Japanese retail-tech company I worked for when I first arrived in Japan, and I have very nice memories of it.

It's also a great way to know the latest technology trends in the retail industry. This year, it was packed with the latest innovations in retail, and it’s always a great place to see how the industry is adapting to some very real challenges in Japan, from labor shortages to shifting customer expectations.

This year, three main themes stood out to me:

• The growing role of GenAI and avatars in retail

• The rapid shift toward in-store automation

• The push for more sustainable operations

In this article, I will provide an overview of what are the main challenges Japan will be facing from now to 2050 and will go deep into each of the tech trends I noticed during my visit to Retail Tech Japan 2025.

Japan Macro Trends Impacting the Retail Industry

Japan’s retail industry stands at a pivotal moment shaped by deep national trends that extend far beyond commerce. Four major forces are redefining the landscape and accelerating the need for innovation. Interestingly enough, my home country Spain is also facing similar challenges such as the aging population.

1. Aging Population and Shrinking Workforce
   With over 29% of the population aged 65 or older and declining birthrates, retailers face both a shrinking customer base and a growing labor shortage. This dual pressure is pushing the industry to rethink everything, from store design and product lines to how operations are staffed and supported.
   

2. Labor Shortages Across the Board
   As Japan’s workforce continues to shrink, frontline retail jobs are harder to fill. This has led to a surge in automation, from self-checkout kiosks to AI-driven customer service tools. Retailers are increasingly looking to technology not just for efficiency but for business continuity.
   

3. Low Productivity in Traditional Retail
   Despite Japan’s global reputation for innovation, many retailers still rely on manual processes and outdated systems. I could see this first-hand during my first years working in Japan when I had the chance to work at a supermarket and experienced first-hand how many of the inventory checks were done manually. This highlights the need for digital transformation across everything from inventory management to personalized marketing.
   

4. Climate Pressure and Sustainability Expectations
   If you have lived or visited Japan, you might have realized how much dependent the retail industry is on using plastic for everything. Also, Japan is one of the countries most vulnerable to climate-related disasters. At the same time, both consumers and regulators are demanding more sustainable practices. From eco-friendly packaging to energy-efficient store formats, retailers are expected to lead in environmental responsibility, often with tech playing a central role.

Retail-Tech Trends Observed at Retail Tech Japan 2025

1. GenAI + Avatars: Looks Are One Thing, Use Case Is Everything

AI-powered avatars were everywhere. NEC, Toshiba, and several other companies showed off virtual agents that looked polished and human-like. In particular, NEC’s avatar was surprisingly natural in appearance and tone.

But here’s what matters more than just how “real” they look: what these avatars do. The tech that impressed me most wasn’t about flashy graphics, it was about how these agents connect with backend systems to answer real customer questions, guide them through store processes, or even make personalized product recommendations. This is where AI orchestration comes in: making sure the avatar can tap into customer history, inventory systems, and the context in real time.

One of the most practical use cases we can see in the industry now is GenAI-powered customer service. Think of a virtual agent answering questions like “Do you have this in stock?” or “Can I return this item without a receipt?” It’s the kind of use case that doesn’t require a moonshot to implement, but can immediately reduce support costs and free up staff, something especially valuable in Japan where consistent, high-quality service is the norm

.

2. Automation in Stores: My Everyday Shopping Is Changing Fast

During the event, all the big IT vendors such as Toshiba TEC, Fujitsu, NEX, etc. were showcasing their self-checkout solutions. But what struck me most about automation wasn’t just what I saw at the event, but what I’ve seen outside it.

Over the past year, my local supermarket, Life, has undergone a quiet transformation. It used to be a fully staffed checkout experience. Now, almost every checkout lane is self-service. One staff member watches over four or five stations, stepping in only if something goes wrong. It's fast, efficient, and reflects how automation is no longer just for tech demos, it’s here, and it’s scaling.

I also recently visited another Tokyo supermarket testing out self-serve carts using tablets with barcode readers so users can scan the products by themselves and save checkout time when paying at the end. The concepts are similar to what the company Caper is doing in the US but less advanced.

These solutions aren’t just about convenience, they’re solving a growing problem: labor shortages. Retailers in Japan don’t have the luxury of large staff pools, and automation is increasingly becoming a necessity, not just a nice-to-have.

3. Sustainability

Sustainability didn’t dominate the floor, but it was there, woven into a lot of the solutions on display.

I saw systems designed to reduce food waste by tracking expiration dates and optimizing shelf restocking. Others focused on improving energy efficiency, smart refrigeration units, for instance, that adjust based on store traffic or ambient conditions.

Additionally, at the 2025 Osaka Expo, 7-Eleven introduced hydrogen-powered smoothie machines and energy-generating floors in their futuristic stores. These innovations not only offer novel customer experiences but also contribute to sustainable energy practices by utilizing renewable energy sources and harnessing kinetic energy from customer foot traffic. ​

Final Thoughts

Retail Tech Japan 2025 left me with the feeling that the future of retail isn’t just about cutting-edge technology, it’s about solving real-world problems, problems that Japan needs to start facing now. Whether it's AI avatars taking over repetitive support questions, automation helping stores run with smaller teams, or sustainability efforts reducing waste, these innovations are moving fast and finding their place in daily operations.

Terms like ownership, accountability, and responsibility are often heard in the corporate world. When these terms are well comprehended by employees, it can improve how a corporation works. Otherwise, it can work against it.

Hence, it is important for all employees to understand these concepts clearly. Among the three concepts, ownership is the most interesting one. The reason is that while responsibility and accountability are both decided by external agents, ownership is decided by oneself and is completely voluntary.

Ownership is often key to our progress in both work and life. I often reflect on its meaning, effects, and the ways to encourage it in others. In this article, I define responsibility, accountability, and ownership, and focus on the importance of ownership. I also provide some recommendations on how to cultivate an ownership mindset in an organization and how to develop it as an individual.

Responsibility, Accountability, and Ownership

While sometimes these terms can be mistaken, each is very different from the other.

• Responsibility: This one is probably the easiest to understand of the three. It is part of a member’s role in a team or project. When you are responsible, your outputs are of high quality, and members increase their trust in you. Responsibility can be seen when results are given. It is more of a rational thing to do, once a member understands their goal.

• Accountability: Usually, either a team or an individual is held accountable for the results of a project. For example, someone can be responsible and accountable for an activity of a project, but not accountable for the total result of the project. This accountability might fall under the project manager or a higher position. This is not chosen and is usually assigned externally.

• Ownership: This is the most interesting of the three. Ownership is created, adopted, and accepted by oneself and not requested externally. This is more about proactiveness, initiative, and developing a feeling of owning the processes and the results, without waiting to be told.

The importance of ownership

Contrary to responsibility and accountability, ownership is usually not written in a job description and is not something that can be measured by numbers. However, it is something that people can feel when it is there and in the same way, very easy to see when it is not there.

Taking ownership of one’s work can be beneficial for oneself and for the organization. At the individual level, displaying ownership will make you a better professional, accelerate your growth, and make other colleagues and superiors notice, increasing the trust they have in you.

At the organizational level, creating an environment that fosters ownership among employees is what all corporations should be chasing after. However, this is far from easy, and below we are exploring in which ways this can be achieved.

For corporations, how can you create an environment that fosters ownership in members?

So, at this point, you might be asking yourself, how can a corporation achieve employees increasing their ownership of their work? In fact, should a corporation even expect all its employees to display ownership when they are already meeting their responsibilities?

Many leaders’ goals are to move employees from a responsibility mindset to an ownership mindset. However, this is not an easy task. To do that, a corporation should aim to provide a work environment in which responsibilities and accountabilities are clear enough that encourage employees to increase their ownership. It also requires leadership to move from a top-down approach to one in which employees are encouraged to find their own ways to meet the goals and have enough space to be creative and explore.

Only by doing this will employees start having the tools and inspiration to display more ownership. Even with an environment that encourages ownership, ownership is an intrinsic trait, something that comes from oneself. Hence, there is a possibility that employees will still not develop it. In these cases, usually, other variables come into play, such as salary, the personal situation of employees, etc

For individuals, how do you develop ownership?

Some practical advice I could give to start building ownership is the following:

• Be proactive: Don´t wait to be told what to do. Actively anticipate and suggest improvements.

• Actively seek feedback to improve yourself: This means you are prioritizing your development, even though you know you hear things about yourself that you might not like.

• Go the extra mile: If possible, try always to go one step further than what was expected of you.

• Even though you are not accountable for a specific task or project, act as if you were. This is a good situation because it means that, ultimately, if the result is not favorable, you will not be the one accountable. However, you will have shown you care about it and acted as if it were yours, which will be noticed by your team and superiors.

Conclusion

Understanding the differences between responsibility, accountability, and ownership is key in organizations. Particularly, taking ownership by being responsible for own’s actions, decisions and results can be of great significance for one’s career and organization. It shows you care about your work, that you are committed, and can be trusted.

An organization will always benefit from having employees with ownership, but creating an environment in which it is encouraged is not easy and requires organizations to take conscious steps toward it.

A Japanese supermarket recently gave its staff smile scores.

Not a manager’s opinion. An AI.

450 factors. Facial expression, voice tone, greeting delivery. Returning a single number.

Your smile is 42%. Not good enough.

This is the same country where taxi drivers wear white gloves. Where staff bow as you leave. Where the word for customer literally means “honored guest of the highest standing.”

I worked at a Japanese supermarket, and I can say that the pressure to make customers feel appreciated is real.

But Japan didn’t need AI to care about service.

So why did it go here, and what does that tell us about where the rest of the world is heading?

It is no secret that Japan is one of the countries with the highest, if not the highest, dedication to the client. The customer in Japanese is called O-kyaku-sama (お客様).

If we break it down, the “O” is an honorific, “Kyaku” means the client, and “Sama” literally means god, and is also usually used when the subject of the sentence is someone of very high standing.

This means that not only do companies strive to provide good customer service but also, customer expectations are quite high and demanding. When talking about CX, proof of this can be Japan scoring the lowest in the NPS scores on average.

This means that on average, customers in Japan tend to be more strict when expressing how satisfied they are with a brand.

In this situation, it is not surprising that companies go the extra mile to make sure that their clients are happy and their expectations are met.

Last year, one of Japan's biggest retailers, created a lot of noise in the industry by announcing the implementation of an AI solution to monitor their employee’s smiles, which has created a lot of controversy in terms of ethical issues.

This use of AI opens different debates, for example, is this an ethical use of AI? Is this impacting EX? And if so, are there other ways to improve CX to the detriment of EX?

In this article, we will talk about Japan's strict approach to CX, how this Japanese company is exactly using AI to monitor employees’ smiles, and the ethical dilemma that is created depending on the use we give to AI.

Japan's high CX expectations

Omotenashi culture - a cult to the client

Japan places high importance on client satisfaction. There is a word in Japanese that actually refers to the art of putting the client always first which is “Omotenashi” (御持て成し).

This concept is deeply ingrained in Japanese society and culture, and it can be seen everywhere: train stations, restaurants, hospitals, and everywhere else.

Some examples of it are store staff welcoming clients (rather loudly sometimes), taxi drivers automatically opening taxi doors, or staff going out of their shop to say goodbye to a client with what looks like an eternal bow in the middle of the street.

NPS, a challenge for companies in Japan

Companies use NPS to measure loyalty by gauging how likely they are to refer a brand to their friends and family. When measuring the NPS, companies need to take into account the country and the industry they are in among others.

For example, by looking at the picture below, we can see that while on average, Japan tends to to provide a low NPS, customers in countries like Brazil or India tend to be more on the promoter side.

The tale of an AI smile solution

Having laid the ground for how important is for Japanese companies to keep the client happy at all moments, now we are in a better position to understand why a company can go as far as deploying an AI solution to monitor their employee´s smiles.

Last year, one of the largest retail companies in Japan deployed an AI solution called ‘Smile-kun’, becoming the first company to utilize AI to assess the tone and demeanor of its customer service employees.

The system is designed to evaluate and standardize the smiles of staff members across 240 locations and is capable of analyzing more than 450 factors, including facial expressions, voice volume, and the way employees deliver greetings. Based on these insights, the system assigns an overall attitude rating to customer-facing staff.

Additionally, it incorporates a gamified element, motivating employees to enhance their scores by improving their demeanor.

After a three-month testing period, the company reported a 1.6 times improvement in the attitude of their staff towards customers. How they measured it is not clear, but it certainly raises ethical questions about the morality of putting your workers under these kinds of expectations when expressing themselves.

With this in mind, it is no surprise that the implementation of this solution has been criticized and categorized as another form of potential harassment to the staff and has been compared to another initiative from a big fast food retailer in Japan called ‘Smile Zero Yen’ in which the company pressured staff to provide smiles arguing that smiling is a thing free to do.

About the ethical use of AI

AI is here to stay, and when used right, it can create a real impact. In healthcare, AI-powered diagnostics like Google’s DeepMind detect diseases earlier than doctors. Japan uses AI for earthquake predictions, giving people life-saving extra minutes. In environmental protection, AI helps track illegal fishing and deforestation in real time.

But AI can also go wrong. AI-driven smile monitoring is a good example since it can add stress to our employees instead of improving customer experience. Even in hiring, AI can reinforce bias, in the news we have seen some recruiting tools that discriminate against women. Without ethical oversight, AI risks doing more harm than good.

So, how do we measure this?

To measure how ethical AI is, there are different fairness metrics like demographic parity to detect bias, and explainability tools like SHAP to ensure transparency. Also, regulatory frameworks like the EU AI Act set guidelines for companies to use AI responsibly.

My takeaways

Going back to the example of AI being used to measure smiles, if the ultimate goal is to improve customer experience, there are probably more effective ways to achieve this without (potentially) negatively impacting employee experience and mental well-being.

A More Strategic Approach to CX

Rather than starting with AI as the solution, companies should first take a structured approach to improving CX. An example of steps that could be followed are:

1️⃣ Analyze the customer journey: Map out key touchpoints where customers interact with your brand. Identify moments of friction that impact satisfaction.

2️⃣ Identify key pain points: Use customer feedback, support ticket trends, and VoC data to identify the biggest challenges affecting CX.

3️⃣ Connect pain points to business impact: Assess how these challenges influence revenue generation, customer retention, and operational costs. Prioritization should be based on impact, not just ease of implementation. However, sometimes, it could also make sense to prioritize based on ease of implementation, what in the industry we call the ‘low-hanging fruits’, prove it works, and then scale it.

4️⃣ Prioritize and act: Once pain points are clear, define which ones should be addressed first and determine the best mix of solutions, whether process improvements, employee training, or AI-driven enhancements.

AI as an Enabler, Not the Starting Point

Once pain points and goals are clear, AI can help automate, assist, and optimize, but applying AI arbitrarily without understanding the real customer challenges will lead to suboptimal results.

Particularly, in the customer service industry, we are seeing significant advancements in AI that enhance both CX and EX. Some impactful approaches include:

📌 Automating repetitive inquiries: AI-powered chatbots and IVRs can handle routine requests, freeing up employees to focus on more complex and value-driven interactions. Studies show that automation can reduce handling time by up to 30-40% (and depending on the use case even more).

📌 AI-assisted employee support: AI Assistants can provide real-time suggestions, knowledge retrieval, and response optimization to the agents working at the contact centers, helping them respond to user inquiries faster and more accurately.

📌 Reducing employee stress with AI: Some companies, like SoftBank, are experimenting with AI-powered tools that modulate the tone of irate customers, helping to create a healthier work environment.

If you liked this article, these articles might be worth your time:

​

By 2050, countries like Japan and Spain are projected to experience significant demographic shifts. Japan's population is expected to decrease to approximately 105 million, with 38% of its citizens aged 65 and older. Similarly, Spain anticipates that 30% of its population will be 65 or older by 2050. Given these trends, it's crucial to emphasize the importance of designing user interfaces (UI) and user experiences (UX) that are accessible and intuitive for the elderly.​

A Conversational AI Christmas Tale

A couple of years back, we bought an Alexa for my grandma, who just turned 90 years old and has been living alone since my grandpa passed away. Initially, I thought she might not use it. However, contrary to that, Alexa has become an important part of the house.

She uses Alexa to do many things, for example, asking for the weather, setting up alarms, playing old Spanish folk songs, asking history questions, and even practicing some math.

One time, I heard from my uncle that my grandma called him, worried because she had asked Alexa to help her practice the multiplication table of 6. In the meantime, she left the room and let Alexa continue counting. When she came back, Alexa had continued with the multiplications up to very high numbers, and she was worried about not taking better care of her since she must have been tired. Of course, she knew that the machine could not get tired, but nevertheless, she did feel a bit guilty.

It wasn’t until one night when I stayed at her house that I realized what was going on. After we went to sleep, in the silence of the night, I could hear my grandma talking with Alexa before sleeping. She was asking Alexa to set an alarm for her, checking the weather for the next day, and wondering whether she should wear a coat to church or not. To my surprise, she also said good night to Alexa.

That produced in me a strange feeling hard to describe. In some way, I felt sad about the fact that she was talking to an AI as if it were a person, and it reminded me of how great it would be if she could tell my grandpa instead. On the other hand, it made me happy that even though Alexa is not a real person, it was right there, filling a real gap in life.

This made me reflect on the value of AI, one I would not have thought too much about before but has now become such an obvious and important part of the AI puzzle: the elderly and the value that AI can provide in countries with continuously decreasing birth rates.

Can AI replace humans?

Here, I am not talking from a business point of view in terms of revenue or cost efficiency. I am talking about the emotional side. Can AI really fill the emotional gaps that we humans are desperate to fill? Love, sadness, anger, frustration… Can AI help us improve our emotional well-being?

The obvious answer would be, well, no. AI can help us improve efficiency at our jobs and even make the world more sustainable, but AI does not understand feelings. Or does it?

When talking about this, many of us probably think of the movie Her, in which its main character falls in love with an AI-powered operating system.

The reality is that while AI does not have feelings, this does not mean it can’t produce feelings in us. In the same way that a video game or a movie can make us happy or even bring us to tears, AI can also help fill gaps of loneliness or love.

In Japan, we have seen examples that, as crazy as they may sound, show just how deeply AI can integrate into human lives. A year ago, we saw the first wedding between an AI avatar and a human. They divorced last year, and I still wonder why.

CX for the elderly

We also bought a smartphone for my grandma this Christmas, and after some research, we decided to purchase one that was said to be easily customizable for the elderly. I created a manual for her to call and send messages, cleaned the desktop and phone of all useless apps, increased icon sizes, and even created direct call icons so she wouldn’t get lost looking for contacts to call. Despite all this, it was still difficult for her to grasp the fundamentals of how to use it.

Knowing how to use a smartphone, while second nature to us, the “young,” is not so intuitive for older people. After researching, I have yet to find a product that specifically targets the elderly in this sense.

However, voice is a different game, and Conversational AI solutions are much more intuitive to use than smartphones, since you just need to, well, talk.

Now, every time we discuss implementing a chatbot for our clients, I can’t help but also think about the elderly and how easy and intuitive the solution will be for them.

Particularly, Conversational AI solutions such as Alexa or Google Home need to place special emphasis on this demographic. In countries like Japan or Spain, this will become a necessity. While they might not be the clients, they will certainly be the users.

Closing notes

This experience with my grandmother made me realize that AI is not just about automation, efficiency, or business impact, it can also serve an emotional and social role. While AI may never replace human relationships, it can help bridge gaps, ease loneliness, and provide a sense of companionship in ways we might not have considered before. As societies continue to age, we must think beyond the young and tech-savvy, ensuring that AI is built for everyone, including those who need it most.

You don’t need permission to lead

In recent years I had the chance to grow into more leadership-oriented roles where I could experience first-hand what it takes to motivate, align, and enable a team of professionals to achieve specific goals while aligning with company objectives.

As I reflect on my own leadership learnings, I discovered that one of them is empowerment. I believe everyone has a hidden gem inside them, and under the right circumstances, it can shine. It's my responsibility to discover what truly motivates each team member.

To improve our leadership skills, I believe it is natural that we first look at CEOs, since they are the most explicit proof of leadership, we see them at our internal kickoffs, at big conferences, and in the news. Hence, this raises the question of, what skills they possess that made them reach where they are. And, can we apply those same skills to our careers, regardless of where we are to improve and advance?

I understand that the factors that make CEOs reach where they are are not only the skills they develop and apply but also factors such as background, education, luck, and network come into play. However, regardless of that, without the right skills, they probably would have never made it.

From my own experiences and from the research I have done, I reached some conclusions I would like to share. Let´s dive into it.

What makes a great CEO?

Let’s start by identifying which ones are the skills that make great CEOs. To do this I am going to leverage the main learnings from a book I recently read called CEO Excellence, written by Carolyn Dewar, Scott Keller, and Vikram Malhotra.

This book explores what are the main traits and characteristics that differentiate great CEOs from not-so-greats based on the performance of their companies and other factors. The authors interviewed some of the most successful CEOs and identified 6 mindsets that distinguish the best leaders from the rest. They are:

• Setting the Direction – Defining a bold vision and strategy.

• Aligning the Organization – Ensuring the right people, culture, and structures are in place.

• Mobilizing Leaders – Building a strong executive team and empowering them.

• Engaging the Board – Managing relationships with stakeholders and governance structures.

• Connecting with Stakeholders – Balancing shareholder interests with customers, employees, and society.

• Managing Personal Effectiveness – Staying resilient, focused, and continuously learning.

In the section below, we will explore how anyone can apply these 6 mindsets to their jobs to be better professionals.

How to apply each CEO mindset to our roles?

Now, let´s go one by one analyzing how ICs and people managers can leverage them to improve their work. In this article, we will focus on the first 3.

Setting the direction

The best CEOs are masters in setting the right direction, to do that, they define a clear and bold vision, make strategic choices, not just plans, and drive relentless execution.

For ICs

Don´t focus just on tasks, focus on impact, meaning you can always have a mindset of prioritization, giving more priority to high-impact tasks rather than for example, getting lost in multiple low-impact activities. Also, try to be more proactive than reactive.

When talking about Customer Success, I can say first-hand that this is one of the most challenging parts for Customer Success Managers. It is very easy for CSMs to get lost in multiple low-impact operational tasks and forget about the high-impact ones, such as bringing real value to our clients.

For people managers

In the same way that a CEO sets the vision for a whole organization, a people manager, regardless of the team size, should define the vision for the team is leading. In other words, give your team members a Why? that will move them to push forward even under difficult times. In the same way, apart from setting the vision, a people manager needs to enable members to work towards it, meaning, unblocking them from roadblocks in the way.

Aligning the organization

Great CEOs ensure everyone is working towards a common goal, they break down silos to improve cross-team collaboration.

For ICs

Instead of only focusing on your tasks, try to see the bigger picture, build relationships outside of your team, and proactively suggest ways to improve cross-team collaboration if you notice inefficiencies.

For example, if the department KPIs are not proactively shared by your leadership, you can proactively reach out to your superiors and ask how your own work connects with them. This not only will help you to understand the bigger picture and align with it to be more effective but I can assure you it will leave a great impression on your higher-ups since they will recognize your strategic thinking and willingness to contribute.

For people managers

Make sure that everyone in your team understands their responsibilities and how they contribute to the department OKRs and the whole company's goals. You can potentially create monthly meetings to make sure everyone is aligned. Make sure that your team is not working alone. Who are the main stakeholders of your team? Identify them and make sure your team is collaborating with them often.

Mobilizing leaders

CEOs need to delegate part of their leadership to other levels in the organization, hence, empowering others and fostering leadership at all levels is crucial. In this regard, both ICs and people managers can take action on their own to develop this leadership.

For ICs

Leadership is not just about leading people, is about displaying one of the traits that I find the most fascinating and difficult to develop, ownership.

To do this, you can look for small moments of leadership, for example by proactively mentoring someone new in your team, by owning the running of a meeting, or just directly asking your superiors how you could add more value.

I believe that leadership is made, one is not born a leader. However, it does require curiosity and proactiveness, two traits that every successful CEO possesses.

Also, giving credit to others is something that somehow can be expected from a manager and of course from a CEO, but it is not so natural for an IC to do this. By giving credit to others when needed, you are positioning yourself naturally as a leader, since you let go of your ego and recognize other ICs for their work.

For people managers

As a manager, your job is to enable your team to become the best version of themselves. The best way to do this is to cultivate an environment that fosters ownership, and this is not an easy task. Each employee's motivation to work is different, for some might be the money, and for others learning new things or get a promotion. It is our role to know each member and create a trusting environment in which members can share their motivations. From here, you will be a step closer to knowing how to wake up the ownership needed in each member to grow and become a leader.

Final thoughts

In Part 2 I will explore the last 3 CEO mindsets and how ICs and people managers can learn from them and apply them in their daily work.