Looking back on four years at AstraZeneca through the lens of the book "Feedback-Driven Management" (Nikkei BP)

Pausing to put it into words #03
English follows

The other day, I introduced Masamune Mimura's new book, "Feedback-Driven Management." As CASE 5 in that book, he highlights the initiatives I led during my four years as President of AstraZeneca.

Dialogue is the key to innovation, driven from the top

The book begins as described above. The organizational engagement score in my fourth year, taken in November 2025, showed improvements in 7 out of 8 categories (24 out of 27 items) compared to when I first took office. I had not imagined such a clear score increase in an organization of over 3,000 people, and I was surprised myself.

The reality behind the "neat story"

If you read CASE 5 of this book, it might look as if I strategically drove the VoE cycle as a top leader and brilliantly transformed the organization. Thanks to Mr. Mimura's ability to put it into words, I was honestly a bit surprised myself, thinking, "Was it really such a well-organized story?" The actual four years were not that orderly.

This kind of reform is not something where you take one measure and immediately get a clear reaction from the organization. You listen to voices, take small steps, and listen again. I believe it was something that could only be done by steadily repeating that back-and-forth process over time.

And this was by no means a battle I fought alone. The management team and communication team, who shared the same vision, supported each other, and above all, the sales branch managers and office heads, who act as the bridge to the front lines, embodied the attitude of "honestly communicating what cannot be solved immediately, but sincerely talking about the efforts behind it" on the front lines. The organization changed because of that accumulation of efforts.

The first donut meeting I attended after becoming President of AstraZeneca overwhelmed me with the sheer volume of information. Based on materials neatly organized by members regarding events that occurred the previous month, the person in charge would explain if a new product was approved. About 55 minutes of the hour were devoted to this one-way communication.

There was only a little time left for questions, but no one raised their hand. I felt a similar sense of discomfort when I first became a top executive in Taiwan and when I led the oncology business at Takeda Pharmaceutical (Japan). On the other hand, in Dubai, in contrast, I was impressed by how questions flew back and forth from the very beginning of my tenure.

This is not a question of superiority or inferiority, but rather a difference in educational styles. In the field of pedagogy, two traditions of learning are often contrasted: the "Socratic method," where one finds answers for oneself through dialogue and questioning, and the "Confucian method," where one accepts the teachings of a master and masters them through practice *α. I feel that many people in Taiwan and Japan have received education closer to the latter tradition and were not yet sufficiently "accustomed" to the act of asking questions itself. It is not a matter of which is better; it is simply that the foundations on which they were raised are different.

That is precisely why, in management in East Asia, I consciously spent time helping people realize that "speaking up is a modest contribution to the organization". Specifically, if there was a proposal that seemed feasible, I would make it a reality as quickly as possible. Then, at the next town hall, I would have the person who made the proposal—a sales representative or branch manager—announce it directly, rather than me. I repeated the process of having the fact that their voice was heard and that things actually changed be told in their own words.

What I valued in order to get people to speak up was what you might call "quick wins"—making what could be realized take shape as quickly as possible. However, that alone was not enough. For voices that could not be solved immediately—such as requests regarding compensation systems or IT/digital environments—I made it a point to show a "roadmap" indicating how and on what timeline we would improve things, even if I couldn't provide an immediate answer. In fact, these two items were among the areas that improved the most in the fourth-year score (a 7-point improvement).

And for things that were decided to be "not done" after consideration, I honestly explained the reasons. This book calls this "fair process" and explains that the sense of satisfaction of the person who raised their voice depends more on whether the judgment process was fair than on the result itself *β. Regarding the company-wide EV policy, many voices were raised from the field, but I did not change the policy itself and continued to sincerely explain why we made that judgment.

Quick wins, roadmaps, and fair process. I feel that the distinction between these three is the part that this book has organized most carefully. For more details, please be sure to pick up a copy of the book.

Practice came first, and theory followed later

For ten years, I think I have been walking through the fog, feeling my way. Takeda Pharmaceutical in Taiwan, Dubai, Oncology Japan, and now AstraZeneca.

The terrain of the destination was different every time. This was because the management team I worked with, the experiences and frameworks they brought, and the culture on the ground were all different each time. I had no choice but to move forward by trial and error, and I had never been conscious of a guidepost called "theory" in that process.

This time, when I read our company's case objectively through Mr. Mimura's book, I felt for the first time as if I had been handed a map. I hadn't been walking with the help of a map. After I finished walking, I realized that the path I had taken was drawn there with surprising accuracy.

For me, the theory of the VoE cycle was not a map showing the path I should take from now on, but a map that illuminated the path I had already walked. If I had had this map beforehand, there might have been times when I could have walked with a little less hesitation.

But at the same time, I also think that because I walked by trial and error without a map, I was able to learn the feel of each place with my own body. I realized once again that there is a way of learning where you don't learn the theory first and then practice, but rather the theory catches up after you have repeated the practice.

Since I first stood in the position of top management at Takeda Taiwan in 2016, I have been practicing interactive management, relying on my own experience and learning. However, I did not consciously grasp this as a clear framework. This is because the management members I work with change from time to time, and the experiences and frameworks they have cultivated so far are also different.

I don't give detailed instructions for everything, but rather proceed while flexibly accepting the members' ideas and ways of thinking. Therefore, the form of the initiative becomes slightly different each time. In the last 10 years, I have had four opportunities to work on organizational transformation: Taiwan, Dubai, Takeda Oncology Japan, and now AstraZeneca.

When I looked back on it, what I realized when I read our company's case, which was objectively systematized this time, was that the form of dialogue I had been groping for on the ground overlapped with the theory of the "VoE cycle" advocated by Mr. Mimura. It is close to the feeling that practice came first and theory caught up later.

It should be noted that while our company customarily called this annual quantitative survey a "pulse survey," according to the definition in this book (the 4-quadrant framework of organization/individual x quantitative/qualitative), this is classified as a quantitative survey targeting the entire organization = "engagement survey" *1.

Naturally, these numbers alone are not the success indicators for organizational transformation. Behind them was the accumulation of daily, sincere dialogues by frontline leaders whose names are not mentioned.

The significance of learning from success and failure experiences

What I strongly feel again through this book is that,

When something goes well, or when it doesn't, you should appropriately reflect on the factors and formalize the essence in a way that can be applied when other issues arise. Only then does it lead to true growth.

In fact, this perspective is one of the important yardsticks in the talent reviews of young employees conducted by the management team. I want young leader candidates to remember that we look at whether they can reflect on their successes and failures in their own words and continue to improve them in a reproducible way, rather than the magnitude of the success or failure itself.

Three shifts seen over the last 10 years

What is important for getting people to speak up is not just listening. Proposals that seem feasible should be put into shape as soon as possible. And, rather than having the management team say it, have the person who made the proposal—the sales manager, branch manager, or marketing person—speak about it directly at the next town hall. Have them convey the fact that their voice was heard and things actually changed to those around them in their own words. Whether or not this extra step is taken changes how the realization that "it is meaningful to speak up" spreads.

Respond to voices that can be answered immediately with quick wins. For voices that cannot be solved immediately, show a roadmap of how you will face them and on what timeline. And for things that are decided as "not to be done" after consideration, honestly explain the reason. This attitude, which the book calls "fair process," taught me that a sense of conviction resides in the fairness of the judgment process, not in the results. I believe that the attitude of facing even the voices that cannot be answered, rather than just picking and choosing the voices that can be answered, is what builds trust.

I have been practicing interactive management relying only on the experience and learning of each situation. I was not conscious of it as a theory. However, by reading our company's case objectively this time, I realized that the path I had walked by trial and error was already drawn on a map called the VoE cycle. I believe that by not letting experience end as just experience, but by reconnecting it with theory later, it becomes knowledge that can be handed over to someone else next.

The main points of this article are the following three.

• Voices have meaning only when they are not just received but acted upon and shared by the individuals themselves.

• Even for voices that cannot be addressed, engage sincerely through a fair process.

• Experience only becomes transferable knowledge when it is reconnected with theory after the fact.

Conclusion

I would like to take this opportunity to express my gratitude to Mr. Mimura and the entire team who helped me carefully articulate my own four years through this interview. You have given meaning and reproducibility to the days of trial and error.

If there is anyone who feels that their practice is currently outpacing their theory, there is no need to be impatient; you may simply be in the process of creating a map.

2025 (Domestic) Prescription Drug Sales: AstraZeneca Takes the Top Spot | Maintaining Over 500 Billion Yen

Looking back on the past four years, it is July 2026, and I am proofreading this article. By coincidence, an article in the industry newspaper Mix Online (published July 1, 2026) announced the performance of Japanese pharmaceutical companies for fiscal year 2025. Engagement is linked to performance. Congratulations to everyone working at AstraZeneca!

Sources and References

*1 Masamune Mimura, "Feedback-Driven Management," Nikkei BP, 2026, p. 404, Figure 22-3 "Four-Quadrant Framework for VoE Channels" (Chapter 22)
*2 Survey Methods and Notes: Performance was aggregated for companies whose results could be confirmed through Mix surveys or FY2025 financial results (fiscal years ending Dec 2025, Mar 2026, etc.).https://www.mixonline.jp/tabid55.html?artid=80480
AstraZeneca, Chugai Pharmaceutical, Otsuka Holdings, MSD, Novartis Pharma, Janssen Pharma, Pfizer, Sanofi, Eli Lilly Japan, Boehringer Ingelheim, GlaxoSmithKline, Bristol Myers Squibb, Bayer Yakuhin, AbbVie, Novo Nordisk Pharma, and Kyowa Kirin are based on results for the fiscal year ending Dec 2025. Nisshin Pharmaceutical is based on the fiscal year ending May 2025. All other companies are based on the fiscal year ending Mar 2026.
* An asterisk (*) next to a company name indicates drug price basis.
* Otsuka Holdings figures are for the healthcare-related business. * Eli Lilly Japan figures are on a financial basis including global alliance sales. * Shionogi & Co. results include Torii Pharmaceutical (consolidated from Sep 1, 2025). * Kaken Pharmaceutical results include medical device sales. * US dollar figures were converted at a rate of 1 USD = 148.71 JPY.

*α Gorry, J. (2011) "Cultures of learning and learning culture: Socratic and Confucian approaches to teaching and learning." Learning and Teaching, 4(3), pp.4-18.
*β Masamune Mimura, "Feedback-Driven Management," Nikkei BP, 2026, p. 340, Column on Fair Process

English starts from here
Reflect and Articulate #03

A few weeks ago, I introduced Masamune Mimura's new book, Feedback-Driven Management. In it, as Case 5, he featured the four years of work I led as President of AstraZeneca K.K.

Dialogue is the key to innovation — and it starts at the top

This is how the book's chapter opens. In the organizational engagement survey conducted in November 2025 - my fourth year in the role - 7 of 8 categories (24 of 27 individual items) showed improvement compared to when I first took office. For an organization of over 3,000 employees, I hadn't expected the score to move this clearly. It genuinely surprised me..

The reality behind a "tidy story"

Reading Case 5, you might picture me as a president who strategically ran the VoE cycle and transformed the organization in one clean sweep. Mimura-san's writing captured it so well that I found myself thinking, honestly a little surprised, "was it really this coherent a story?" The actual four years were nowhere near that tidy.

Change of this kind doesn't work by launching a single initiative and getting a clear, immediate response back from the organization. Listen. Make a small move. Listen again. That slow back-and-forth, repeated over time, was really the only way through it.

And this was never a fight I waged alone. Executives and communication teams who shared the same conviction supported one another, and — more than anyone — the branch and office managers who served as the bridge to the front line embodied a simple stance at the ground level: "be honest about what can't be solved right now, but speak candidly about the effort behind it." The organization changed because that stance was repeated, day after day.

The first Donut Meeting I attended after taking office was overwhelming in sheer volume of information. Using materials the team had neatly prepared summarizing what had happened over the previous month, a product director would explain if a new product had just been approved. Of the hour-long meeting, roughly 55 minutes were devoted to this one-way flow of information.

Time was left for questions, but no one raised a hand. I felt the same unease when I first took the top leadership role in Taiwan, and again when I led the oncology business in Japan. Dubai, by contrast, was different from the start — questions came thick and fast from day one.

I don't think this is a matter of one culture being better than another — it's a difference in educational tradition. In education, a distinction is often drawn between the "Socratic" tradition, in which people arrive at answers through dialogue and questioning, and the "Confucian" tradition, in which people receive a teacher's guidance and internalize it through practice.※α Taiwan and Japan draw more heavily on the latter tradition, and I sensed that the habit of simply asking a question hadn't yet taken root. It isn't about which is superior — it's simply a difference in the ground people grew up on.

That's precisely why, in East Asia, I spent deliberate time helping people feel that speaking up was itself a small contribution to the organization. In practice, that meant moving quickly on any proposal that could realistically be implemented — and then, at the next town hall, having the person who raised it, a branch or office manager, someone from marketing, announce the result themselves, rather than having me announce it. I kept repeating this: let people hear, in the words of the person who spoke up, that their voice had actually changed something.

What mattered for encouraging people to speak up was moving quickly on what we called "quick wins" — implementing what could realistically be implemented, fast. But that wasn't enough on its own. For voices that couldn't be resolved right away — questions about compensation structures, or requests around IT and digital tools, for example — we tried to lay out a "roadmap": not an immediate answer, but a clear sense of the timeframe and the path toward improvement. Those two areas, in fact, were among the categories that improved the most by my fourth year in the role.

And for anything we ultimately decided not to do, we made a point of explaining why, honestly. The book calls this "fair process" — the idea that people's sense of being heard depends less on the outcome itself than on whether the decision-making process felt fair.※β Our company-wide shift to EVs drew plenty of pushback from the field, but we held the policy steady while continuing, honestly and consistently, to explain the reasoning behind it.

Quick wins, roadmaps, and fair process. If there's one thing this book untangled and organized more carefully than I ever had, it's the distinction between these three. I'd encourage you to read the book itself for the details.

Practice came first; the theory caught up later

For ten years, I feel like I was walking through fog, feeling my way forward. Taiwan, Dubai, Takeda's Oncology Business Unit Japan, and now AstraZeneca — the terrain was different every time. The executive team I worked with changed, along with their experience and the frameworks they'd built over their own careers. There was nothing to do but keep moving by feel; I never consciously reached for something called "theory" as a guide.

Reading my own organization's case laid out so objectively through Mimura-san's book, I felt, for the first time, as though someone had handed me a map. I hadn't been navigating by that map — I'd only been handed it after the fact. And once I looked at it, the path I'd already walked was drawn there with startling accuracy.

The VoE cycle, for me, wasn't a map pointing the way forward. It was a map that illuminated, in hindsight, the road already taken. Had I held this map earlier, I might have wandered a little less in a few places. But then again, it was precisely because I walked without one that I came to know the texture of each stretch of ground by feel.

Learning theory first and then practicing it is one path. But there's another: practicing first, and letting the theory catch up later. That's the kind of learning I found myself rediscovering this time.

Since 2016, when I first stepped into an executive leadership role at Takeda Taiwan, I've relied on my own experience and learning to practice dialogue-driven management. I never consciously framed it as an explicit theoretical framework, though — because the executive team I worked alongside changed each time, and each of them brought different experience and different frameworks they'd developed over their own careers.

Rather than directing every detail myself, I've tried to stay flexible, taking in the ideas and thinking of whoever I was working with. As a result, the shape of the effort has looked a little different every time. Taiwan, Dubai, Takeda's Oncology Business Unit Japan, and now AstraZeneca — over these ten years, I've had four separate opportunities to lead organizational change.

Looking back across all four, what struck me this time — reading my own organization's story rendered so objectively and systematically — was that the dialogue we'd been feeling our way toward on the ground turned out to overlap with the theory Mimura-san calls the "VoE cycle." Practice came first, and the theory caught up with it later. That's the closest way I can describe the feeling.

One clarification: internally, we've always called this annual quantitative survey a "pulse survey." However, under the book's own definition (see below chart), what we were actually running falls under "Engagement Survey."

Needless to say, this number alone isn't the measure of successful organizational change. Behind it was a steady accumulation of honest, everyday dialogue carried out by front-line leaders whose names never make it into the story.

The value of learning from both success and failure

What this book left me thinking hardest about is this: both success and failure only become an asset you can reuse — something reproducible in the next situation — once you've reflected on it, broken it down into its components, and put it into explicit, articulated form.

When something works, or when it doesn't, properly reflecting on why, and formalizing the essence of it in a way you can draw on when the next challenge appears — that's what it actually takes to grow, in the fullest sense.

In fact, this is one of the standards we use in talent reviews for younger leaders on the executive team. What matters isn't the scale of the success or the failure itself, but whether someone can put it into their own words, reflect on it, and keep improving in a way that's genuinely reproducible. That's a skill I hope every rising leader keeps in mind.

Three shifts this decade revealed

Encouraging people to speak up takes more than listening. Whatever could realistically be implemented, we moved on quickly — and then, rather than announcing the result ourselves, we had the person who raised it, a branch or office manager, someone from marketing, say so themselves at the next town hall, in their own words. That small extra step changed how far the sense that "speaking up matters" was able to spread.

We answered what could be answered quickly with quick wins. For what couldn't be resolved right away, we laid out a roadmap — a timeframe and a direction. And for whatever we ultimately decided not to do, we explained why, honestly. What the book calls "fair process" taught me that people's sense of being heard rests less on the outcome than on whether the process felt fair. Trust isn't built by only picking up the voices you can act on — it's built by facing the ones you can't, too.

For years, I relied only on experience and learning gathered along the way to practice dialogue-driven management. I never thought of it as theory. But reading my own case laid out objectively this time, I realized that the path I'd felt my way along had already been drawn on a map called the VoE cycle. Experience doesn't have to stay as mere experience — reconnecting it with theory, after the fact, is what turns it into something you can hand to someone else. In short:

Voice matters most when it's not just heard, but acted on — and told by the person who raised it

Even the voices you can't act on deserve to be met with fair process

Experience becomes something you can pass on only once you reconnect it with theory, after the fact

Closing

I would like to take this opportunity to thank Mimura-san, and the whole team whose work through this interview gave such careful, articulate shape to my own four years. To the days of trial and error — thank you for giving them meaning, and reproducibility.

If you're someone whose practice hasn't yet been caught up by theory, I don't think that's something to rush. It might simply be a map still being drawn.

I introduced the book "Feedback-Driven Management" itself in the previous article. If you haven't read it yet, I'd encourage you to take a look there as well.

Sources

※1 Masamune Mimura, Feedback-Driven Management, Nikkei BP, 2026, p. 404, Figure 22-3, "The Four-Quadrant Framework of VoE Channels" (Chapter 22)

※α Gorry, J. (2011) "Cultures of learning and learning culture: Socratic and Confucian approaches to teaching and learning." Learning and Teaching, 4(3), pp.4-18.

※β Masamune Mimura, Feedback-Driven Management, Nikkei BP, 2026, p.340, column on fair process.