Best Time to Post Substack Notes: Data From 933 Real Notes

Every "best time to post" article I have read says something like "Tuesday mornings perform best" or "post between 9 and 11 AM." No source. No data. Just vibes dressed up as advice.

So I stopped reading those and started measuring.

I pulled data from 933 of my own Substack notes. All times below are the times I posted, in Europe/Berlin. The account writes about tech and careers, which matters, and I will come back to it.

The numbers, and how small they really are

The overall average across all 933 notes was 0.9 reactions per note. That sounds low, and it is.

Reactions on notes are not newsletter opens. Most notes get very little engagement. So when I say a slot averaged 1.1 reactions, that is not a spike. The gaps here are small in absolute terms, and it is worth holding that in mind for everything below.

My best hours were 21:00, 20:00 and 12:00, each averaging 1.1 reactions. Those are not thin samples either: 117 notes behind the 21:00 slot, 107 behind 12:00, 74 behind 20:00.

My best days were Sunday, Friday and Tuesday, each averaging 1.0.

Wednesday was my worst day, and my busiest

Wednesday averaged 0.7 reactions per note. Against averages this flat, that is a real drop.

Here is the part that actually surprised me. Wednesday was also my highest volume day. I posted 156 notes on Wednesdays, more than any other day of the week.

I was pushing out the most writing on the day that got the least back, and I had no idea.

I was posting whenever I sat down, and Wednesday is apparently when I sit down.

That is the real finding. Not that 21:00 is magic. It is that without measuring, you cannot tell whether your habits are working for you or against you.

What this cannot tell you

I want to be honest about the limits, because they are the reason most articles like this are useless.

• The gaps are small. We are talking about fractions of a reaction per note. Posting at the right hour does not double your reach. It might nudge things.
• The patterns might not hold. They are what happened, not a law. Next quarter could look different.
• I do not know where my readers are. These are the hours I posted, in my timezone. Substack does not tell me the timezones of the people reading my notes, so I cannot say whether 21:00 works because of when I post or because of who happens to be awake. That is a real hole in this.
• It is one account in one niche. Yours has different readers with different habits.

If you take 21:00 from this article and paste it into your schedule, you are doing exactly what the vague advice articles ask, just with my data instead of someone's guess.

The transferable thing is the method. Not my 21:00.

How to find your own hours

The annoying part first. Substack does not give you an export of your notes with their timestamps and reaction counts. There is no download button for this. Post analytics exist; notes are not treated the same way. So there are two ways to get this, and the honest version is that one of them is free and tedious and the other is a tool.

By hand. It genuinely works:

1. Start recording as you go. A spreadsheet with two columns: when you posted, and the reactions a few days later. One row per note. That is a minute a day, and in two months you have something real.
2. Add the back catalogue only if you want it. You can work backwards through your profile and copy timestamps in by hand. For a few hundred notes it is an afternoon.
3. Split the timestamp. Add a column for the hour and one for the weekday. In Google Sheets, HOUR() gives you the hour and TEXT() with a format code gives the day name.
4. Average by group. Group by hour, then by day. That is the same table I built.

Or have it recorded for you. This is what StackBuddy does: it schedules the note, then keeps the timestamp and the engagement, so the table above builds itself instead of depending on you remembering. Every account starts with 14 days of Pro and no credit card, and the analytics are part of Pro, so you can see whether your own numbers say anything before deciding it is worth paying for.

Either way, the thing that matters is starting. You cannot recover history you never recorded, which is why starting today beats designing a better system next month.

Check the count before you believe the average

One caution before you act on anything you find.

Look at how many notes sit behind each average. A slot with six notes in it can look brilliant or terrible purely by chance. A slot with a hundred is telling you something.

If a number surprises you, check the count before you believe it.

Then look for your version of the Wednesday problem: a day or hour where you post a lot and get little back. That is the pattern worth acting on first, because it costs you nothing to move.

The harder part is posting into the slot

Knowing your best hour and actually using it are different problems.

Most of us post when we finish writing, not when the clock says it is a good moment. That is normal. Writing is unpredictable.

This is where scheduling earns its place. If 21:00 on Sunday is good for you, you can write at 14:00 and have it go out at 21:00. You stop posting randomly and start running an actual test.

The bigger point stands with or without a tool. Most advice about posting times is recycled from platform to platform with nothing behind it. The only data that matters for your account is yours. And you might find, like I did, that you have been working hardest on your worst day.