Is a Good Lab One That Produces Good Results?「失敗の多い研究室」から強い人が育つ理由

on

(by Wuled Lenggoro)

Our Most Valuable Product Is Our Alumni

I have been the sole faculty member of this lab since its founding in 2007.

So what sustains students’ growth is not the number of faculty, but the intellectual interaction among the students themselves.

When people think of a lab’s achievements, they usually think of the number of papers, citations, patents, and grants. Of course, these matter too.

But after running this lab (empatlab.net) since 2007, what I’ve come to feel is this: the most valuable product of a lab is its alumni.

Graduates from my lab have gone into remarkably diverse fields — chemical & materials, heavy industry, semiconductors, automotive, construction, food, government, and academia. Toray, Mitsubishi Heavy Industries, SCREEN Holdings, Toyota Motor Corporation, JGC Corporation, Itoham Foods, Daikin Industries, Japan’s Ministry of Defense/Self-Defense Forces, and even alumni who became professors at universities in Malaysia and Indonesia.

Yet when you look at their undergraduate and master’s thesis topics —

  • Particle deposition on plants
  • Measurement of thermophoretic phenomena in air
  • Detecting particles in raindrops
  • Soil coating
  • Molding of artificial feed
  • Application of soot to heat exchangers

— almost none of it bears any direct relation to their current jobs. In fact, our lab’s own website states plainly: “The relationship between a student’s research topic and their job afterward doesn’t seem to be very strong.”

So why can graduates thrive in fields so unrelated to their research themes?

At first, I believed that research output and the number of conference presentations were what mattered most. But after watching students for more than fifteen years, that belief has gradually changed.

What Fifteen Years Taught Me

Here’s the surprising part: students with unfinished results are often stronger in job interviews.

When I first started running the lab, I believed it was important to get students to produce results as quickly as possible. But after fifteen-plus years of watching students go through job hunting, I’ve noticed something. A student who has “already presented at a conference” or has a “completed success story” isn’t necessarily the strongest candidate. What matters more is being able to explain: “What’s the problem right now?” “Why do you think it’s happening?” “What will you try next week?”

One graduate, for example, joined Dai Nippon Printing (DNP) and later moved into consulting at Accenture. What was valued in his final interview at DNP wasn’t the outcome of his materials synthesis research — his thesis had been on synthesizing secondary battery materials, though what he was actually wrestling with day to day was a coolant water leak in the equipment. What mattered was his process of framing problems and making decisions. Today, he doesn’t use that specific knowledge; he uses that same process of problem-framing and decision-making as his weapon in the consulting industry.

Why I Let Students Choose Their Own Research Themes

In my lab, I almost never assign a theme to a student. More often, a student looks at the parts and equipment already sitting in the lab and designs their own theme around them. Just looking at recent student themes:

  • Synthesis of zinc oxide using green tea extract
  • Separation of proteins from milk
  • Water-repellent surface by soot
  • Aerosol collector without using a suction pump
  • Water movement in soil with polymer added

— they’re all over the map. And yet, scattered as these themes look, there’s a common language underneath: Materials Transfer — or Transport Phenomena, the movement of matter, heat, and fluid.

In fact, one undergraduate thesis student once discovered, quite by accident, an unusual kind of soot particle from a candle flame that dispersed in both water and alcohol. Over the following decade, that discovery reappeared in different forms — an ultrasonic cleaning method, a hydrophilic/hydrophobic switching membrane, and eventually a study on tomato seed germination. Themes that look disconnected end up connecting in unexpected ways.

What I do as the supervisor is simply keep asking, “Why are you doing this?” The very fact that the themes are scattered is what frees students’ thinking from the professor’s own expectations.

Our lab is housed in a building filled with agricultural science labs. I sometimes encourage students toward a fusion of engineering and agriculture — but the decision is always the student’s own to make.

What I Do in Weekly One-on-Ones

In our weekly meetings, what I ask is nearly always the same:

Why that parameter? Why that temperature? Why that duration? Why do you need a comparison group? Why does it matter? Why did you find it interesting?

With one undergraduate, for instance, we went back and forth repeatedly: “Why compare deionized water and tap water?” “Why five minutes on, five minutes off for the ultrasound?” “Why 90°C?”

I’d rather have students discover their own reasons than hand them an answer.

In Our Seminars, the Presenter Isn’t the Main Character

In many labs, a seminar is training for the presenter. In mine, it’s training for the audience.

At one undergraduate’s presentation, other students fired off questions like:

Why pH 4.6? Why 10 minutes? Why use a spray dryer? What does a 213% yield actually mean? How did you distinguish protein A from protein B? Why did you use the theoretical value?

Each seminar slot is officially one hour, but the Q&A alone often runs past an hour, sometimes continuing almost nonstop. The questioners get sharper training out of this than the presenter does.

Why the Seminars Get So Lively

The reason is simple: the students aren’t competing with each other. In my lab, nobody fights over data, ideas, or achievements — because everyone is working on a different theme.

A student working with milk gains nothing by criticizing a student working with green tea extract. So the questions come from pure curiosity, not competition or self-promotion.

Students aren’t really interested in someone else’s results. They’re interested in why that person made the decisions they did.

Lab Seminars That Run Without Professor in the Room

This is a little lonely for me as a professor — and, at the same time, deeply satisfying as an educator.

Sometimes I have to step out partway through a seminar for another meeting. Normally, discussion would stall. In my lab, it doesn’t. I sometimes catch up later by watching the Zoom movie recording.

It’s not the professor driving the discussion — it’s the students. That’s one of the things I’m proudest of.

What Matters in the AI Era

AI can produce answers. It still can’t produce interesting questions.

AI can now search papers, summarize them, write in English, and even generate polished graphs. In other words, the value of “knowing the answer” is dropping.

But:

  • Why choose this problem?
  • Which hypothesis do you believe?
  • What do you ask next?

— none of that can be outsourced to AI. That’s exactly what I want to cultivate.

In our lab, we frame this using the word “Failure.” Failure doesn’t mean “it didn’t work” — it means “it didn’t behave as expected,” and that’s information for designing the next question. The more unexpected the result, the more worth finding interesting.

Not the Job Offer — Strength Twenty Years Later

I’m not all that interested in whether a student lands a job offer. What genuinely interests me is what they’re like ten or twenty years after graduation.

Many of our graduates go into industries entirely unrelated to their research theme — chemicals, semiconductors, automotive, construction, food, consulting, government, academia, and more. What they share in common is the ability to think for themselves and explain their own reasoning.

Quite a few graduates have since changed careers. They don’t seem particularly afraid of jumping into a new field. I suspect what they actually learned in the lab wasn’t a specific body of knowledge — it was how to face an unfinished problem.

I don’t tell students, “Go to grad school” or “Get a job.” Instead, what I want to tell them is: “Become someone who, twenty years from now, can still dive into a new problem.”

Doctoral Students Play by Different Rules

Doctoral students don’t have as much freedom as undergraduate or master’s students. Because of ties to sponsors and projects, the direction of their research is often largely set from the start. I think of this as simply unavoidable.

Here’s my reasoning: by the time students reach the doctoral stage, I assume they’ve already gone through their “learning-mode” — a high-entropy phase where mistakes are welcomed — during their undergraduate and master’s years. From there, the doctoral phase moves into “performance-mode” — a low-entropy phase where smaller mistakes are preferable.

I don’t see this as taking away a student’s freedom. I see it as simply a different stage.

Connections With People Are Also Future Value

In 2008, I was involved in running a JSPS short-term program at TUAT as a host faculty member. One of the young researchers who took part — a master’s student from Indonesia — was T.L.

Honestly, I had forgotten about it. But in 2025, I ran into him again in Yogyakarta, Indonesia. He had become the head of the Graduate Studies, Faculty of Forestry at Universitas Gadjah Mada (UGM), now in a position to host forty TUAT graduate students.

It was a future that no one could have predicted seventeen years earlier.

Future value doesn’t only exist in research — it exists in human connections too.


So this is what I’ve come to believe: a good lab isn’t one that produces good results. It’s one where people grow who can generate good questions.

Our most valuable product is our graduates.

(Prepared by Wuled Lenggoro, 2026/8)

Note: As of this writing (2026), I have advised more than twenty doctoral students as the sole faculty advisor since the lab’s founding in 2007.

https://note.com/empatlab/ (日本語版)

https://researchmap.jp/blogs/