#14 Why does work handover fail even when data remains?

In a previous article, I wrote about how I, having joined the company in sales, was suddenly assigned the job of breeding as a breeder.
The thing that confused me the most at that time.
It wasn't that I lacked specialized knowledge.
It was the state of being in a situation where,
'The data remains, but I don't know how to do the work.'
There were past test results.
There was data on varieties.
There were records of the fields.
And yet,
'So, what should I do next?'
When it came to that, I didn't know.
This time, I would like to write about the 'difficulty of handover' in the job of breeding, which I felt from my own experience of actually taking over the work.

'Data remains' and 'the work has been handed over' are different things.

When I took over the breeding work, it wasn't that past records were completely gone.
In fact, the data itself remained quite well.
Characteristics for each variety and line.
Results surveyed in the field.
Which individuals or lines were selected in the past.
Information as numbers was certainly there.
At first, I thought,
'If this much data remains, I'll manage somehow.'
However, when I actually started the work, I immediately hit a wall.
For example, suppose an evaluation of a certain line remained.
But even looking at those numbers,
'Why was this item surveyed?'
I didn't know.
Furthermore,
'Why was this line kept?'
I didn't know that either.
The data exists.
But the purpose for which that data was taken, and what was thought when looking at those numbers, did not remain.
In other words, what was left was,
The result of 'what was done'.
What was missing was,
The reason of 'why it was done that way'.
I only understood this difference after actually taking over the work.

Even with the same '7 points', the meaning is not necessarily the same.

In breeding, plants are evaluated on various items.
Suppose an evaluation of '7' out of 10 points remains for a certain characteristic.
Looking only at the number,
I understand that 'it's a 7'.
However, what I really want to know is beyond that.
'Why is it a 7?'
'Compared to what is it a 7?'
'Is 5 not good enough?'
'How much importance was placed on this characteristic?'
'What is the reason for keeping it even if there are other flaws?'
Without such information, even if you receive only the numbers, you cannot reproduce the same judgment as the predecessor.
In breeding, you don't necessarily decide simply by looking at one feature and saying,
'This is the best, so I'll keep it.'
Yield.
Shape.
Color.
Resistance to disease.
Vigor.
Ease of cultivation.
Taste.
Marketability.
While looking at various elements,
Sometimes you judge, 'There are flaws, but I'll keep it because it has this feature.'
That is why, more than the final number,
'What was being thought when that judgment was made at that time'
becomes important.

Breeding involves many judgments that are difficult to turn into numbers.

What is even more difficult is that breeding involves quite a few judgments that are difficult to express only with numbers.
For example, when walking in the field,
There are times when I think, 'For some reason, this plant bothers me.'
How the leaves are attached.
Plant habit.
How the fruit is attached.
Overall balance.
I can put it into words if I try.
However, when you have been looking at plants for many years, before you can verbalize everything one by one,
You sometimes feel, 'Something is different.'
And when you investigate in detail later, it turns out it really had an interesting feature.
Knowledge based on such experience is explained in management studies using the concept of 'tacit knowledge'.
Ikujiro Nonaka and Hirotaka Takeuchi show that organizational knowledge includes not only knowledge that is easy to express in words and numbers, but also knowledge rooted in experience that is difficult to verbalize, and they argue for the importance of sharing and converting these within an organization (*1).
When I took over the breeding work,
I think one of the reasons I felt, 'The data is there, but I don't understand,' was exactly here.
In the predecessor's head,
'At this time, I look at it this way.'
'If it's in this state, I keep it.'
'Here, I prioritize this over the numbers.'
There were judgment criteria like that.
However, not all of that had become data.

What I was really troubled by was actually the field work.

And what I struggled with even more was a part separate from the data.
That was,
'How to manage daily work.'
For example,
'This tool is here.'
'This machine is used this way.'
'If a problem occurs with this equipment, contact this contractor.'
I had received explanations like that.
However, when I actually started the work,
'When do I use that?'
I didn't know.
Even if I know the location of the tool,
I cannot use it if I don't know which task it is needed for and at what timing.
Even if I am taught how to operate the machine,
I cannot prepare if I don't know when in the year that machine is needed.
Greenhouse equipment.
Agricultural machinery.
Ordering materials.
Repair contacts.
Field preparation.
Work requests to part-time workers.
Work timing for each crop.
These things are complexly connected in the work of one year.
If I make a manual,
'How to use machine A'
can be explained.
However, what was really needed was,
'When it reaches this state around month X, prepare machine A next.'
That was the flow of work.

Even though there was 'dot' information, it hadn't become a 'line'.

Looking back now, I think the thing I was most troubled by during the handover at that time can be expressed with these words.
Even though there was 'dot' information, it hadn't become a 'line'.
Data.
Machines.
Tools.
Contractors.
Work methods.
I had been taught about each one.
But,
'When, what happens, and what do I do next?'
I didn't understand that flow.
Agricultural work moves significantly depending on the season.
Sowing seeds.
Raising seedlings.
Planting.
Managing plants.
Surveying.
Cross-pollinating.
Harvesting.
Collecting seeds.
Advancing to the next generation.
Moreover, schedules can shift due to weather and plant growth.
It is not work that proceeds according to the calendar.
Therefore,
I realized later that it is necessary to hand over not only 'work methods' but also 'timing for judgment'.

In the end, I only understood after experiencing it myself for a year.

At that time, I was working while asking the person in charge of research and development, who was enrolled as a contract employee,
'What should I do about this?'
'What should I do next?'
each time.
And after actually experiencing the work for one year, finally,
'So the machine I was taught about at that time is used here.'
'So this material was needed for this.'
'That's why it was prepared at this time.'
One by one, things connected.
It is the feeling that the 'dots' that were scattered at first finally became a 'line' over the course of a year.
I think this is also a difficulty unique to work like agriculture and breeding, which is strongly tied to the flow of the seasons of the year.
There are many things that you cannot understand just by being explained once, until you actually reach that season and that growth stage.

Why does handover become difficult?

Of course, this is not a problem only in the seed and seedling industry.
It is a problem that occurs in many companies.
The 2025 White Paper on Small and Medium Enterprises also introduces the importance of promoting skill succession to reduce person-dependent work, and cases where technology and know-how become person-dependent during the company's growth process, making succession and standardization a challenge (*2).
Only a specific person knows.
Only a specific person can do it.
Work cannot proceed unless you ask that person.
This state is generally called 'person-dependency'.
I feel that breeding and cultivation work is particularly prone to person-dependency.
Because,
Plants do not behave exactly the same way every year.
The temperature is different.
The rain is different.
The amount of solar radiation is different.
The occurrence of pests and diseases is also different.
Plant growth is also different.
Therefore,
It is not necessarily the case that 'if you follow the manual, you will always get the same result'.
In the end,
Human judgment of 'if it's this year, I'll do this' becomes necessary.
Experience inevitably involves itself in that judgment.
That is why I think complete manualization is difficult and knowledge is easily accumulated in the individual in charge.

And, surprisingly, there is not much time for handover.

Another thing I felt was,
The time available for handover is not as long as you think.
That is.
The person in charge is transferred.
Retires.
A new person in charge is decided.
Handover starts from there.
However, due to things like taking paid leave, there are cases where the period during which the predecessor and successor can actually work together becomes shorter than imagined.
Furthermore, in the case of breeding,
Even if 'if you do it together for one year, you can teach one year's worth of work',
In reality, it is not always the case that such a long handover period can be taken.
In a few weeks,
Spring work.
Summer work.
Autumn work.
Winter work.
You cannot show everything in practice.
Here too, there is a difficulty unique to seasonal industries.
That is why I,
Came to think that it is too late to create handover materials after retirement or transfer is decided.

Because I experienced it myself, I now make sure to leave behind the 'reasons'.

Because I struggled with the handover myself, I now change the way I leave records a little.
For example,
Not just 'I kept line A',
But adding a word about 'why I kept it' as much as possible.
Not just 'I worked on month X, day Y',
But also leaving behind 'why I worked at this timing'.
Not just numbers,
Writing down the background such as 'because the temperature was high this year, I judged X days earlier than usual' as much as possible.
You cannot record everything perfectly.
Even so,
I think that just by leaving behind 'result + reason',
The understanding of the person who looks at it next will change considerably.

Photos and videos are also excellent handover materials.

There are many tasks that are difficult to explain with text alone.
In such cases, I try to leave behind photos or videos.
For example,
rather than just writing,
“Perform the task when it reaches this state,”
it is overwhelmingly easier to understand if there is a single photo of the actual plant.
For operating machinery as well,
there are times when taking a video of a few minutes conveys more
than writing pages of text.
If you have a smartphone, you can record it easily.
In the past,
when I thought about “creating handover materials,”
I had an image of creating impressive manuals in Word or Excel.
But now,
photos,
videos,
Excel,
notes,
past data.
I think it is fine to combine those things.
What is important is not creating beautiful materials.
It is that the next person can reach the answer when they are in trouble.
That is the most important thing.

Could AI also be used for technology transfer?

Recently, I have been thinking that AI could also be used here.
For example, trying to leave behind daily work content and reasons for decisions in writing becomes a burden in itself.
During busy periods,
even if you are told,
“Please write down the reasons for today's decisions,”
it is hard to keep up.
But,
if you speak into your smartphone,
“I kept this plant today. The reason is that the plant vigor was strong and the fruit shape was good,”
and then transcribe and organize it, the burden can be reduced considerably.
Furthermore, if you can search the accumulated records,
“What was I doing at this time last year?”
“How did I respond before when this symptom appeared?”
if you can search for these, you can make past experience easier to use.
Of course, I do not think that AI can be entrusted with all breeding decisions themselves.
Rather, what is important is,I think it might be using AI to ensure human experience is not lost,rather than replacing human experience with AI.

There is no need to create a perfect manual

When thinking about handovers,
one tends to think,
“I must turn everything into a manual.”
However, it is probably impossible to put all breeding and cultivation work into text.
I myself am not aiming for that either.
My current goal is simpler.
“Leave behind enough so that the next person does not struggle with the same things I did.”
That is all.
Something I worried about for an hour, the next person can solve with a one-line note.
If it took me half a day to find a vendor, I leave the contact information.
If it is something I noticed over a year, I leave a photo and a brief comment.
Even just that changes the starting point for the next person significantly.

Summary

When I first took over the work,
I thought,
“With this much data left behind, why don't I understand it?”
Now, I understand the reason a little.
It is because what was left behind was the “results” of the work.
What really needed to be handed over was,
what was done,
why it was done,
when the decision was made,
what was looked at to make the decision,
and what to do next,
which is the “flow” and “way of thinking” of the work.
Breeding is a job where it is difficult to completely replace human experience and intuition with numbers.
That is precisely why,
not just leaving data, but how to leave experience behind.
I think that will become increasingly important in the seed and seedling industry from now on.
I myself am still in the middle of trial and error.
Even so, I do not want the next person to go through the same handover struggles I experienced as much as possible.
For that purpose,
“Leave behind not just results, but reasons,”
“Use not just text, but also photos and videos,”
“Record daily work before retirement or transfers are decided,”
I am continuing these things little by little.
Leaving behind technology might not be about saving data, but about passing on “why it was done” to the next person.

*1 Source: Nonaka, I. & Takeuchi, H. (1995), The Knowledge-Creating Company: How Japanese Companies Create the Dynamics of Innovation, Oxford University Press.
*2 Source: Small and Medium Enterprise Agency, “2025 White Paper on Small and Medium Enterprises, Part 2, Chapter 1, Section 4: Human Resources Strategy,” etc. Examples of reducing personal-dependent work through skill succession and the personalization and succession of technology and know-how are introduced.
https://www.chusho.meti.go.jp/pamflet/hakusyo/2025/chusho/b2_1_4.html
▼Read more
・#11 The story of how I, who joined as a salesperson, suddenly became a breeder
・#12 Why do you become a “jack-of-all-trades” when working at a seed and seedling manufacturer?
・#13 What I realized by learning AI and DX in the seed and seedling industry where FAX remains