You will read everywhere that three to five Notes a day is optimal and that thirty a day gets you shadowbanned. I went looking for the source on both. Here is what actually exists.

Documented, from Substack: writers who post three or more thoughtful notes during their launch week gain 50% more total subscribers than those who do not. That is a real published figure from July 2024. Note that it is three notes in a week, not three a day. That single sentence is almost certainly where the three-to-five-daily rule came from.

Not documented anywhere: the shadowban threshold. Substack’s support pages state no posting frequency guidance, no rate limit and no character limit. Their content guidelines prohibit spam in comments and threads and prohibit importing lists without consent, and say nothing about posting volume. I could not find a single creator who has publicly tested at thirty a day. So treat the number as folklore, because that is what it is.

The one real experiment I could find: a writer who ran his own volume test reported roughly 150 to 400 new subscribers a month at one note a day, 400 to 600 at three to five, and a plateau around 600 at ten a day. He attributed the plateau to reader fatigue and said he saw no evidence of throttling. One person, one publication. Useful, not proof.

What I would actually do. Two to four a day, most days, because that is enough to stay in the feed without turning your profile into a firehose. If your numbers stop improving as you add more, that is your ceiling and it has nothing to do with a hidden penalty.

A Note is not a teaser for your post. It is a complete thought that happens to be short. The ones that travel work as writing on their own, and would be worth reading even if there were nothing else to click.

Say the whole idea. The instinct is to hold something back so people have to click. That instinct is wrong on a feed. Withholding reads as marketing, and people scroll past marketing.

Lead with the specific, not the category. “I learned something about pricing” is invisible. “I raised my price by $200 and got more clients, not fewer” is a note somebody stops for.

Make it easy to disagree with. Notes that state a position get replies. Notes that state a truism get likes at best. Replies are worth more, because a reply puts you in front of that person’s followers.

Short lines, real breaks. The feed is read on a phone. A dense block is a scroll.

Here is the tactic, and then here is the honest accounting of why it might work.

The tactic. Post the interesting thought with no link. Let people reply and ask where to read more. Then drop the link in a comment reply rather than in the note itself.

The reasoning people give is that Substack suppresses notes containing external links. Substack has never said that. The engineer who built the Notes ranking model has described the inputs publicly, and they are things like location, language, what you subscribe to and follow, your interests, audience overlap between publications, how often the creator posts, and how people engage. Links did not appear on that list in either direction. Substack’s own Notes documentation actively encourages links and shows you the rich preview.

What there is evidence for. A March 2026 analysis of 3,230 notes across nine creators found that notes with a link attached consistently earned less engagement, and that the creators who posted mostly plain text averaged the highest likes per note in the group. That is a real study with a stated method. It measures likes rather than reach, it is one niche, and correlation is not mechanism.

My read. The tactic is worth using, but the reason is probably simpler than an algorithm penalty. A note built around a link is usually a worse note. It reads like an ad because it is one. Take the link out and you are forced to write something that stands up alone, and that is the thing that actually travels.

The other half nobody mentions: a reply is a second impression. The person came back to your note, and now they are getting the link from a human answering a question rather than from a banner. That converts better for reasons that have nothing to do with ranking.

Substack’s Notes analytics do show you the real numbers. You get impressions, which is how many times the note was displayed on screen. You get a breakdown of who saw it, split between people who do not follow you, people who follow you, and people who subscribe. And you get interactions including likes, restacks, shares, profile visits and link clicks.

A number from one of my own notes: roughly 1,700 impressions produced 57 link clicks. That is a little over 3%.

I have no idea whether that is good. And neither does anybody quoting you a benchmark, because Substack has never published one and I could not find a credible third-party study of Notes click-through rates. Anyone telling you what a normal Notes CTR is has made it up.

What to do instead of chasing a benchmark. Track your own. Your third month against your first month is the only comparison that means anything, and it is a comparison you can actually act on.

The thing that makes a daily cadence survivable is not discipline. It is having formats you do not have to invent from scratch each morning.

A recurring prompt. Something like a standing question your readers answer, posted the same day each week. People start expecting it, and the replies are the point.

A standing observation slot. One thing you noticed this week in your subject area. No conclusion required.

The receipts note. A number from your own work with the story behind it. These consistently do the best for me, because a specific number is the one thing nobody can generate generically.

An evergreen rotation. Your best notes from six months ago will outperform whatever you would write tired on a Thursday. Almost nobody saw them the first time.

Now the scheduling, and an important 2026 update. Substack shipped native Notes scheduling in March 2026. On mobile you open the composer, tap the three dots and choose Schedule. On the web there is a calendar icon. They extended scheduling to Chat threads and added saved Notes in July 2026. So you do not need to buy anything to schedule notes anymore, and any article telling you otherwise is out of date.

Paid tools still exist and some are genuinely useful, but be clear about what you are buying. AutoStackNotes runs from $17 a month on annual billing, and what it adds over the native scheduler is a recurring weekly queue that refills itself, evergreen recycling and A/B testing. If you want those, it is worth a look. If you just want to line up a week of notes on Sunday, Substack now does that for free.

Notes are the top of the funnel and they are not the product. The mistake people make once Notes start working is letting the long-form slide, and then wondering why the subscriber count climbs while nothing else does.

One or two real posts a week is the anchor. That is what somebody sticks around for after a note got their attention. Notes without posts builds a following. Notes plus posts builds a publication.

The three-touch pattern. Somebody sees a note. Later they see another one and recognize the name. Then they check the profile and subscribe. It is almost never one note, which is exactly why cadence matters more than any individual note being great.

And once people are subscribing, the next question is what happens to them. That is a different subject and I have written it up separately: why Substack is not a blog, it is a conversion engine.