Substack Notes Analytics: Benchmarks From Real Data

You post a Note, it gets six reactions, and you have no idea whether that's good. Substack doesn't publish benchmarks. Your stats page shows you the six and nothing else, and six measured against nothing is not information. Most writers end up comparing themselves to whichever Note they remember doing well, which is a terrible baseline, because memorable Notes are by definition not typical.

Substack Notes analytics only start earning their keep at the moment your numbers get held up against somebody else's: the median reaction count across thousands of Notes, the share that land in total silence, the restack rate that counts as ordinary. We pulled a sample of recent Notes activity and built that table. Underneath it sit the numbers from our own 20,000-Note analysis on what format actually changes performance.

One number worth having before you start: WriteStack has close to 100 verified reviews from Substack writers across multiple platforms. The other analytics tools in this category have none that we could find anywhere. That gap matters more here than usual, because a benchmark is only as good as the dataset behind it, and the dataset is the one thing you cannot inspect from a pricing page.

Table of Contents

• Why Your Own Dashboard Can't Tell You Whether a Number Is Good
• How We Built This Benchmark
• Reaction Count Benchmarks: What's Normal
• How Many Notes Get Zero Engagement
• Restack Benchmarks: How Often Notes Get Reshared
• What Format Does to Performance: Our 20,000-Note Numbers
• How to Read Your Own Numbers Against These Benchmarks
• The Layer That Turns a Snapshot Into a Direction
• What This Data Doesn't Tell You
• What Changes When You Stop Guessing

Why Your Own Dashboard Can't Tell You Whether a Number Is Good

Every analytics screen you have ever opened for Substack, including Substack's own, is built on one publication's data: yours. It knows what you posted, when you posted it, and what came back. Ask it whether Tuesday beat Monday and it answers instantly. Ask it whether six reactions is a good result for a publication your size and it has nothing to work with, because it holds no rows about anyone else's publication. That limit is not a matter of polish or effort. It is what is in the database.

So Notes analytics splits into two questions that sound alike and are not:

1. Did my number move? Your restack rate this month against your restack rate last month. Any per-account dashboard can answer this, and most do it well.
2. Is my number good? Your restack rate against the restack rate of publications roughly your size. No per-account dashboard can answer this at all, no matter how many charts it draws, because the comparison set does not exist inside it.

The second question is the one that changes behavior. A restack rate of 12% climbing to 14% reads as progress until you learn that publications your size average 26%, at which point the real story is that you have been slowly improving inside a hole. The same 12% climbing to 14% reads very differently if the field average is 9%.

This is the reason WriteStack collects Notes performance across publications instead of only inside your account. Your restack rate gets reported against other publications in your size band, and the direction it is moving gets reported as a trend across months rather than as this week's snapshot. The benchmark tables below are the static version of that. The product is the version that stays current and knows which row is yours.

Start a free 7-day trial with WriteStack and see where your Notes land against the field.

How We Built This Benchmark

The numbers below come from a sample of 5,000 recent Substack Notes, pulled and aggregated directly rather than estimated. We are publishing the methodology alongside the numbers because a benchmark you cannot verify is not much better than a guess: sample size, what is included, and what got left out all change how much weight a number deserves.

A few honest limits. The sample skews toward more recently active accounts, since it is ordered by recency, so it likely overrepresents writers who post often relative to Substack's full base. It also cannot separate a Note posted by a brand-new account with twelve followers from one posted by a writer with fifty thousand, so the benchmarks describe the Note-level landscape rather than any single account tier. Treat them as a broad reference point.

That last limit is exactly the one a live product solves and a blog post cannot. A published table has to lump every account size together. A benchmark running inside your own account already knows your subscriber count and can compare you to the band you actually sit in, which is what WriteStack's cross-publication benchmarks do with the same underlying kind of data.

Practical rule: a static benchmark tells you the shape of the landscape. A benchmark that knows your account size tells you where you stand in it. Use the first to calibrate, the second to decide.

Reaction Count Benchmarks: What's Normal

The average number of reactions per Note in our sample was 20.4. The median was 3. That gap is the most important number on this page, because it means the average is being pulled way up by a small number of Notes that did very well, while a typical Note performs far below what "average" suggests.

Percentile	Reactions per Note
25th percentile	1
50th percentile (median)	3
75th percentile	8
90th percentile	25
95th percentile	59

If your Note gets 3 reactions, you are sitting at a typical outcome rather than an underperforming one, even though 3 feels small next to an average of 20. Consistently clearing 8 puts you ahead of most Notes posted. Clearing 25 puts a Note in the top 10 percent of what we sampled.

This gap between average and median is exactly what a raw benchmark hides when only one figure gets reported. A writer who sees "average reactions: 20" and then gets 4 on their own Note might conclude something is badly wrong, when 4 already beats the median. Reporting both numbers side by side is a deliberate choice here, because the average alone would set an unreachable bar for the overwhelming majority of Notes.

Practical rule: compare your Notes against the median, not the average. The average will make almost every normal Note look like it is failing.

How Many Notes Get Zero Engagement

About 20.9% of Notes in our sample got zero reactions. That is roughly one in five landing with no visible response at all. It is a useful number to sit with if you have posted a Note that got nothing and assumed something was wrong with it, your account, or the algorithm. One in five is the normal rate of complete silence.

Here is how the full distribution breaks down:

Reaction range	Share of Notes
0 reactions	20.9%
1-5 reactions	46.7%
6-20 reactions	20.4%
21-100 reactions	8.6%
100+ reactions	3.3%

Nearly half of all Notes land in the 1-5 range. That is the real center of gravity for Notes engagement, not the standout posts that make it into someone's "Notes that blew up" screenshot.

The number that matters for you personally is not this 20.9% but your own version of it, tracked across months. A zero-engagement rate holding at 22% is a publication running normally. The same 22% after three months at 12% is a signal, and you only catch it if something is keeping the monthly series for you.

Restack Benchmarks: How Often Notes Get Reshared

Restacks are the mechanism that pushes a Note beyond your existing subscribers, so they matter more for growth than raw reaction counts. In our sample, 37.4% of Notes received at least one restack. The average was 4.5 restacks per Note, while the median was 0, meaning more than half of the Notes in the sample got no restacks at all.

That combination, a meaningful minority getting restacked while the majority get none, matches what Substack's own description of Notes suggests about the format: it is built to let strong ideas travel through the network rather than guaranteeing every post reach. A restack rate in the high 30s as a share of Notes is a reasonable reference point for gauging whether your own restack activity is typical or unusually low.

Practical rule: if fewer than one in three of your Notes ever gets restacked, the fix is usually the Note's core idea rather than its formatting or posting time.

What Format Does to Performance: Our 20,000-Note Numbers

Benchmarks tell you where you stand. The next question is what to change, and that takes a different measurement: the same Notes sorted by structure instead of by outcome. We ran that across a separate sample of 20,000 recent Notes, and two structural factors came out clearly.

What we measured across 20,000 Notes	Effect on reactions	Effect on restacks
Multi-line Notes vs. single-line Notes	~11% more	~28% more
150-300 characters vs. under 50 characters	~33% more	Larger still than the reactions gap
Notes under 50 characters	Lowest band measured	Lowest band measured
Single-line Notes	Below the multi-line baseline	Furthest below it

The restack column is the one to read twice. Line breaks bought about 11% on reactions and about 28% on restacks, which means structure helps most on the metric that actually distributes your work to people who have never heard of you. A reaction is a nod from someone already following you. A restack is reach.

Both patterns point the same way: Notes carrying a small amount of structure, a line break, a second thought, a bit more context, do better than a single bare sentence. Longer is not automatically better. There is a floor below which a Note reads as too thin to react to, and multi-line Notes clear that floor more reliably than single-line ones.

Numbers like these are measurable because WriteStack reads live Notes data continuously rather than surveying writers or repeating advice. The same pipeline feeds the live inspiration feed, so what you see on a Tuesday reflects the formats working on Substack that week instead of a list somebody wrote once and left up.

Start your free trial and put your own Notes next to these numbers.

How to Read Your Own Numbers Against These Benchmarks

Pull your last 20 to 30 Notes and look at two figures: your median reaction count and the share that got zero engagement. A median near 3 and a zero-engagement rate near one in five means you are performing in line with the broader dataset. A zero-engagement rate running well above 20-25% is a more useful signal to investigate than any single Note's result, because it points to a pattern instead of a one-off.

Doing that by hand once is fine. Doing it every month is where writers quit, which is the practical reason to hand it to something that already has your history. WriteStack's heatmap lays out every Note you have posted by day and hour, so your median and your consistency show up as a picture rather than as a spreadsheet you have to rebuild. Most writers overestimate how consistent they have been until they see the gaps.

Check your numbers at more than one point in time before drawing a conclusion. A single bad week, three or four Notes that all underperformed in a row, falls well within normal variation given how skewed this distribution is. Look at a rolling month rather than a rolling week if you want a read that is not dominated by noise. Writers who track only their best or worst recent Note end up reacting to outliers instead of to their baseline.

For the broader picture of which Substack metrics actually deserve a place on your dashboard, our guide to Substack analytics covers the full set. This page is the Notes-specific half of it.

The Layer That Turns a Snapshot Into a Direction

A table like the one above is a photograph. It tells you what the landscape looked like on the day we sampled it. What actually changes your writing is the moving version: your own numbers, against the field, month after month, with the direction visible.

The part people don't expect

WriteStack benchmarks your restack rate against other publications your size, which is a comparison your own dashboard structurally cannot make. It runs trend analysis across months, so a restack rate sliding from 31% to 24% shows up as a direction rather than as two disconnected snapshots you never put side by side. And it tracks which of the two links inside a Note got clicked, going back to your very first Note, so "this one performed well" becomes "this one sent 40 people to the archive." Open the dashboard on a Tuesday and see that the Note you almost didn't post drove 40 link clicks, and that your restack rate is running well above typical for a publication your size. Not better than last week. Better than the field.

Here is what each layer can and cannot answer:

The question you are actually asking	Substack's own stats	A per-account dashboard	WriteStack
Is this restack rate good for a publication my size?	No	No, no data on other publications	Yes, benchmarked against your size band
Which direction is my restack rate moving month over month?	No	Partial, by hand	Yes, trend analysis across months
Which of the two links inside that Note got clicked?	No	No	Yes, across your full Notes history
Which formats work on Substack this week?	No	Static list, if any	Live inspiration feed from current Notes data
Can I schedule and then measure a chat post?	Not schedulable	Not schedulable	The only tool that schedules chat posts at all
What has my actual posting pattern looked like?	No	Partial	Heatmap of every Note by day and hour
Can drafting start from what already worked for me?	No	Prompt-based	Drafts from your published Notes, with model selection
Verified reviews from Substack writers	Not applicable	None we could find	~100 across multiple platforms

The pattern in that table is one thing repeated: every row WriteStack wins is a row that needs data from outside your own account, or history from further back than your last screenshot. That is the whole difference between a stats page and an analytics product.

Two other pieces sit alongside it. The Activity Center puts engagement work in one place instead of five browser tabs, and cross-posting through Buffer sends the same Note to 11 platforms, so the Notes you write for Substack stop being single-use. Publishing runs through a browser extension handing Notes to Substack's own scheduler, which means your session never leaves your browser.

What This Data Doesn't Tell You

This benchmark describes Note-level outcomes across a mixed sample of accounts rather than outcomes controlled for follower count, niche, or account age. We were not able to cleanly isolate how reaction counts scale with subscriber count inside a bounded query, so we are not reporting a number for that here instead of guessing at one.

We also did not separate niche categories, since Notes topics range from politics to poetry to software, and lumping them together in one table is the honest way to present a general-purpose benchmark rather than implying false precision.

One more gap worth naming directly: this is a snapshot, not a trend line. Substack's Notes feed, the mix of accounts using it, and the behavior the algorithm rewards all shift over months in ways a single sample cannot capture. Treat the specific numbers here as a snapshot from mid-2026 and expect them to drift.

Every one of those three gaps is the same gap: a published table is frozen, flat, and about everyone. Your account is live, sized, and about you. Closing that distance is precisely what a benchmark running inside your own analytics does, comparing you to your size band, refreshing as the field moves, and reporting the direction rather than the still frame. The table above is the free version. The version that keeps working is the product.

What Changes When You Stop Guessing

Run one audit by hand before you change anything. List your last month of Notes, mark the reaction count and restack count next to each, and sort by performance. The Notes at the bottom almost always share something: a format, a length, a topic, a time of day. That pattern is more actionable than any general benchmark, because it is specific to your audience.

Then look at what the second month costs you. And the third. The audit that felt clarifying in January is the chore you skip in March, and the writers who keep doing it are the ones who stopped doing it manually.

A month into having the numbers running on their own, the change is not that you post more. It is that you stopped arguing with yourself about whether a Note did well. Six reactions is no longer a question mark, it is a data point sitting slightly above the median with a restack rate you can see climbing against publications your size. The Notes at the bottom of your list get identified while you can still learn from them. The formats that work get repeated because you know which ones they were, not because you have a feeling.

Substack gives you your own six reactions. Benchmarks against publications your size, trend analysis month over month, link-click history going back to your first Note, and chat post scheduling are the layer built on top of that, and WriteStack is the tool that ships all of them together. Close to 100 verified reviews from Substack writers back that up, and no other tool in this category has any we could find.

Start a free 7-day trial. Put your numbers against the field. Stop guessing what six reactions means.