What I Learned from Analyzing note Article PVs and Likes | People, Mechanisms, and Followers Change How Content Reaches Others | Analysis | Data | note Operations | AI Utilization | Generative AI #282

Previously, I wrote an article that was
“What actions on note change how articles reach people?”
which was essentially a blueprint for an experiment.

At that time, I was mainly looking at these three things.

• Does follower count affect initial momentum?

• Does collaboration change the flow?

• Does the way an article is read change the reaction?

That article wasn't about providing answers, but rather
a session to share what hypotheses I have and what I am testing.

So, this is the follow-up.

Since I have gathered some data, I want this to be
a session to first collect what has become visible so far.

First, let me state something important.

This is the result that occurred on the note of one creator named KITAcore. Therefore, I don't think this applies to everyone as is.

But conversely,
even with the logs of one person, there are things that become visible when you look at them properly.

And this time, this is what I thought most of all.

What is often said in the note community seems to be true after all.

You all might have already known this. But when you see it in your own numbers, the sense of conviction is completely different.

It really hits home, like, “I see, so that’s what it is.”

Moreover, this isn't just an impression; I hope it can also serve as
a guideline for what direction of articles I should write from now on.

*Since this is data as of the March 5, 2026, aggregation, it changes moment by moment. Please bear with that.

Alright, let's go.

First, the conclusion.

The conclusion that has become visible through this experiment is as follows.

• Articles involving people and those that spark conversations like on social media are strong

• Articles where the format itself acts as a hook are also strong

• When the 'foundation for reach' known as followers grows, the average reach also increases

To put it bluntly, what was strong this time were articles with either
'people' or a 'mechanism' that were present.

1. Articles involving people were indeed strong

First, this was the most obvious point.

Looking at the articles that were strong at the time of aggregation,

• 7th OASOBI Project #266: 586 PV / 70 Likes

• Marco Exhibition Record #272: 520 PV / 64 Likes

• Inspired by someone's creation, I made a video too #279: 445 PV / 53 Likes

like these, articles where relationships with people and the flow of events are visible indeed had strong numbers.

Looking at this, I think it's not just that the article is complete on its own, but rather

“it is being read along with the flow of people”.

Doing something together with someone. Introducing someone. Someone's actions leading to the next creation.

Such articles are naturally easier to share and more likely to generate reactions.

I think this has probably been said in the note community for a while.
But seeing it in my own numbers, I really felt, 'Yeah, that's right.'

2. However, it wasn't just about collaborations

This was also quite important this time.

It wasn't just articles involving people that were strong.
Articles where the format itself acts as a hook were also quite strong.

For example,

• [383-character experiment] #277: 564 PV / 70 Likes

• The Story of How That 314-Character Article Became a Proper Experiment #276: 541 PV / 68 Likes

• [314-Character Experiment] #274: 404 PV / 59 Likes

Around this point, the numbers were quite significant.

In other words, what I saw this time is that
whether "people are involved" or
"there is a mechanism in the article itself,"
having one of these makes it much easier to create an entry point.

So, the conclusion this time is quite simple.

I found this to be quite convincing myself.

3. The "foundation for reach" known as follower count could not be ignored either

There is one more thing I was able to see clearly.

That is, when the "foundation for reach," such as follower count, grows, the way articles are viewed also changes little by little.

For example,
the follower count, which was 787 as of February 4th,
increased to 936 by March 4th.
In this one month, that is +149 people (approx. 18.9% increase).

Along with that, the average PV per article also grew to

• 222.22 → 260.00

In terms of growth rate, that is
an increase of about 17.0%.

Furthermore, the average likes per article also increased to

• 21.00 → 26.36

which is an increase of about 25.5%.

The like rate also rose to

• 9.45% → 10.14%

.

Of course, I want to look at this carefully.

I wouldn't go so far as to say
“everything grew just because my follower count increased.”

The article themes are different, the timing is different, and there's the factor of whether or not there was a collaboration. So, I don't intend to definitively claim causality.

But at the very least,
in my logs, during the period when my foundation grew stronger, the reach and reaction density of my articles also increased together.

I think this is a fact that cannot be ignored.

4. Moreover, articles don't just end with their initial performance

Another thing that was interesting is this.

Articles don't just end with their initial performance the moment they are published.

Looking at the recent incremental growth, past articles were also growing steadily.

For example,

• [314-character experiment] #274: +187 PV / +35 likes

• For those who don't feel refreshed even after sleeping: Why morning light and breakfast work #273: +62 PV / +23 likes

• Marco Exhibition Record #272: +47 PV / +9 likes

There are articles that are being read again later like these.

I think this is quite important, even if it's subtle.

In other words,
an article's lifespan isn't determined solely by the numbers right after publication.

Sometimes they are re-read due to the flow of people, and sometimes they grow later as the context connects.

Therefore, it's a bit too early to judge that it's
“over”
just by looking at the initial speed.

I think this is a very encouraging story for those of us who are creating content.

5. So, what will serve as a guideline this time?

This is the point I want to leave behind most of all this time.

If I were to use these results as a guideline for creating future articles, this is how I would think about it.

1) Articles where people are visible are strong

Rather than stories that are self-contained,
articles that show relationships or context with someone else are, as expected, easier to reach people.

2) Articles that are easy to share are strong

Articles that are easy to pass on by saying,
"Check this out," tend to create momentum.

3) Articles with hooks in their format are strong

Like having 314 characters or 383 characters,
having a mechanism that makes people slightly curious before they even read it serves as an entry point.

4) It is better to think of the quality of the content and the structure for reaching people separately

It is not that articles with dense content are bad.
However, in terms of how they spread, the design of the entry point is quite effective.

In other words,
writing good content and
creating a structure that reaches people should be viewed as separate skills.

This realization was a significant takeaway for me this time.

6. This is not the "correct answer," but "my primary information"

I will state this again.

The results this time are
the experimental results of one creator named KITAcore.

Therefore,

• I do not think this applies to everyone as is

• I have no intention of asserting that this is a law for all of note

• I am not saying this is a cheat code either

But I do believe there is meaning in logs like this.

The reason is that if you rely only on intuition,

"I feel like collaborations are powerful,"
"I feel like short posts perform better,"
"I feel like follower count really matters."

That's where it ends.

But when you actually record and compare them,
“I feel like” changes to “it certainly seems that way.”

This difference is huge.

And that becomes the basis for making decisions for your next article.

7. What I thought most about this time

Ultimately, this is what I thought this time.

What is said in the note community really was true to a certain extent.

People get involved.
Conversations are born.
It's easier to be introduced.
Context connects.
There are hooks in the format.

Articles like that are, after all, strong.

You all might have known that from before.

But this time, I was able to be quite convinced by seeing it in my own logs.

And because I was convinced,
it leads to the discussion of what kind of article to create next.

This is really interesting.

Looking at numbers just to be evaluated is exhausting.

But,when you look at them as material for thinking about your next move, it's really fun.

See? Isn't data fun?

In conclusion

This time, regarding the hypothesis I set in the previous article,
I summarized only the interim results first.

Roughly speaking, these are the three things I discovered this time.

• Articles involving people and those that feel like social media posts are strong.

• Articles where the format itself acts as a hook are also strong.

• As the foundation of follower count grows, the way content reaches people changes little by little.

From here on, I want to break things down a bit more specifically.

• How much did follower count affect initial momentum?

• Did collaborations really change the flow?

• How did reactions change based on how the articles were read?

I will look into these individually in detail later.

For now, I just wanted to put the answer to
“What did I actually see?”
out there first.

Whether you read the previous blueprint article or started reading from here this time,
if you have any thoughts like “I might look at it this way,” please let me know.

If everyone’s primary information overlaps with my primary information,
it will probably be even more interesting.

Previous article

Experimental blueprint is here

Topics planned for next time

• How much did follower count affect initial momentum?

• Did collaborations change the flow?

• Did reactions change based on how the articles were read?

You might be thinking, “Are you actually lying?”
So, I’ll also include the dashboard I’m currently building.

KITAcore-style dashboard: Under construction

If you take a look here, you should be able to see how all the numbers this time are connected.

※It's for my own use for now, but if there's anyone curious enough to want to try this out, let me know privately.

If you thought this article was
interesting
somewhat helpful
, I would be happy if you could react with a like, follow, or comment.

I quite enjoy reactions like likes and comments.

See you later.

Bonus

While making the thumbnail, I thought this one might be good too.

#note #daily-note #note-operation #note-analysis #creative-theory #communication-design #PV-analysis #like-rate #followers #data-analysis #diary #work #business #self #life #learning #record #experiment #consideration #continuation