Understanding Substack's analytics

An updated tour guide

In June 2023 I wrote an extensive post about Substack’s analytics, explaining what it all means and how writers can use those statistics to inform what they’re doing. As I’m sure you’ve noticed, June 2023 is literally years ago, so we’re long overdue an updated version of that guide.

Today I’ll be poking at the latest version of Substack’s stats, as of December 2025, as well as third party tools like Search Console and Google Analytics 4. This is, clearly, going to be specific to Substack, but if you’re using a different email platform it’ll likely have similar analytics — and if it doesn’t, you should ask the developers why not.

Before we get started, though, a quick note. When it comes to numbers and charts, remember that, unless you’re very specifically running a business, they’re optional. If you’re here to write, that’s your only important target. Don’t be distracted by big numbers and charts that go up: that way madness lies. They can help you get towards your goals, but don’t let them become your goals.

The dashboard overview

Right. A lot has changed in Substack’s dashboard since I wrote that 2023 article. Each change brought with it utter rage from some corners of the community, but I rather like what they’ve done.

When you hop into your Dashboard, you’re now greeted with an overview of three key stats and a nice big chart. What is shown depends on whether you have paid subs turned on or not.

Along the top I get Gross annualized revenue, Paid subscribers and Total subscribers, each complete with the main number and a percentage change. Clicking on one of these metrics displays the relevant data on the chart, and I can change the date range using the menu at the top-right.

One thing we all have to learn about publishing a newsletter is that it’s a long game. Your numbers will go up and down, and if you’re sweating over them every day, or even every week, you’re going to burn yourself out from the anxiety. Looking at a multiple-month timescale is more useful and healthier.

For example, here’s my 7-day chart for total subscribers:

Phew. People must have really hated that last chapter of my weekly serial Tales from the Triverse! Right?

Something you need to get used to when you start writing a newsletter is this:

Every time you send a newsletter, people will unsubscribe.

This is inevitable and not something to worry about. You haven’t done anything wrong, but it does take some mental gymnastics to look at it the right way. Each person that unsubscribes in that context is helping to make your list more relevant, more precise. The people remaining at the ones who really want to read your stuff. This compounds over time, and in that way your subscriber list is constantly being finessed, heading towards being more and more targeted towards people who enjoy your work.

Here’s my total subscribers chart over 30 days:

That doesn’t look so bad! Ups and downs, but generally heading in the right direction, I think? Though, it’s kind of plateaued for the last couple of weeks, which could be a bit worrying. Here’s the 90 day view:

You can see there are multiple plateaus, but if you draw the trend line it’s perfectly healthy. This is the main point to take away here: obsessing over short term changes is a bit pointless, because the long term is all that really matters.

Some individual newsletters might bring in a ton of new subscribers. Others will lose you subscribers. Neither matter as much as what happens cumulatively, over a six month period or more.¹

Growth reports

Next, let’s take a look at the Growth report, found under the Audience section in your dashboard. These charts debuted in 2025 and focus on three areas: Unique visitors, New subscribers and New revenue (if you have paid turned on).

Look at this colourful, funky thing:

This is my Unique visitors chart. In this case, a ‘visitor’ is someone who ended up on simonkjones.substack.com, which somewhat separates them from my subscribers who will be receiving posts more directly (to email or the Substack app).

The different coloured bands represent the top sources for my visitors. The big grey blob is ‘Direct’ traffic, which in analytics terminology is the system shrugging and muttering “I dunno.” More interesting are the other colours, and if you hover your mouse over your own chart you’ll get a handy legend:

‘Direct-to-app’ means someone clicked a link somewhere (probably in an email I sent) and wound up in the Substack app. ‘Google’ is search traffic. ‘Substack’ means they came to me via somewhere else on Substack, such as another writer’s newsletter or Note.

At the bottom of the pop-up you can see a summary of what I sent out via Substack on that day. This is useful for correlating activity on the chart with new posts or Notes.

Below the chart is a table of the same data, although often with more granularity. There’s a useful categories menu to filter the results. For example, I can look at traffic that has come specifically from AI:

In my Substack settings I have AI scraping blocked, so I imagine I’d have higher numbers here if I was willing to jump into that particular pit of snakes.

You’ll likely have more visitors than subscribers, because only a subset of visitors will go on to subscribe. This is where we start to get into conversion funnels, but that’s a topic for another time.

Here’s my New subscribers chart:

As you can see, far fewer people are subscribing than visiting (though many of those visitors will already be subscribers). And while 7% of my visitors are attributed to Substack, 62% of my subscribers are attributed in some way to the platform.

In other words: Google brings me a helluva lot more viewers, but relatively few really engage. And think about your own Google behaviour: how often do you Google something and then go on to become a fully signed-up member of wherever you end up? More often you follow the link, get the info and scamper away into the night.

Substack, on the other hand, brings far fewer people to my newsletter, but they are vastly, vastly higher quality leads. These people (you may even be one of them, so - hello!) are more closely connected to what I’m actually writing, and that connection makes it more likely that they’ll be interested in subscribing and reading more from me.

None of us should rely solely on Substack for finding readers, but there’s no denying that it’s a very effective engine for building a readership.

Stats reports

Right, let’s get to the meat. 🥩

The Stats section is a collection of many different reports, some more useful than others.

Network

This report is essentially a promo piece for Substack. It’s them saying “you’d better never leave!”

I’m only half-joking. Look at this:

The various shades of on-brand orange represent my subscribers who were in some way already connected to the Substack ecosystem. The implicit suggestion here is that I might not have got any of those subscribers if I was using a different newsletter platform, like Buttondown or Ghost.

Maybe that’s true, or perhaps it’s simply that in 2025 who doesn’t have a Substack account of some sort? Whether as readers or writers, most of us read someone who uses the platform. Having my subscribers come from Substack-connected places isn’t as remarkable a thing now as it once was.

In other words, did these people subscribe because I use Substack and so do they, or was that largely incidental? It’s not really possible to tell.

Audience

Here you’ll find a completely different chart, with most probably a larger number than your subscriber number:

This is your Total followers. This combines your subscribers with people who ‘follow’ you via other means on the Substack platform, such as on Notes.

(if you’re not familiar, Notes is Substack’s version of Bluesky, Threads, old Twitter etc)

There’s been a lot of hand-wringing and general fretting about the idea of a ‘follower’ on Substack. The concern is that it’s a mode of engagement which we don’t ‘own’: we can’t take our followers with us if we leave the platform, for example. “It’s social media all over again!” cry the people.

Well, it is social media all over again. But do remember that this is an additional layer on top of/at the edges of your publication. It doesn’t really affect what you do with your newsletter one way or another.

What it does do is provide a light touch way for people to stay connected to each other without necessarily committing to reading everything they publish. For example, I follow lots of serial fiction writers but don’t necessarily subscribe to all of them. This is no judgement on their writing: I’d simply never have time to read all of those serials every week, but I am still 100% interested in what that writer is doing. Without the ‘follow’ option, I’d have no simple way of keeping tabs.

Crucially, a ‘follow’ has the potential to turn into a subscribe. It’s the very top of the funnel. Unlike other social platforms, a Substack ‘follow’ can very easily make the jump, because links between short- and long-form content are actively encouraged rather than penalised.

Further down the page you’ll find a map of how much of America reads your stuff, because America is the most important country:

You can, if you want, change the map to ‘World’, but this is very much a secondary concern:

Maybe one day Substack will allow us to pick our local country or continent.

For now, the choice is between the USA or the World, except the World option still includes the USA. This may all be an elaborate metaphor on Substack’s part.

Keep scrolling and you’ll find the audience overlap, which at least for me used to be far more interesting. Here’s what mine looks like today, in December 2025:

What an incredibly weird line-up! Audience overlap shows the publications that my subscribers crossover with. 14% of my subscribers are also subscribed to Letters from an American, for example, which is about US politics and history.

Thing is, all six of these overlap publications are simply massive publications. The crossover has far less to do with what I and they write, and more to do with them having a huge number of subscribers.

Many of my subscribers share the common trait of breathing. There’s a big overlap.

Correlation, not causation.

I mean, Story Club and In The Writing Burrow are both about writing fiction (more-or-less), so that does make a bit more sense. But it’s not like I’m about to email Margaret Atwood or George Saunders and ask them if they want to ‘do a collab’.

And Bari Weiss’ The Free Press? The politics in Tales from the Triverse would, I can only imagine, raise a few eyebrows from that readership.

None of those overlap publications are especially useful to me.

Anyway, my line-up back in 2023 looked like this:

George was still there, at least! But the rest of that line-up is far more interesting, and thematically much closer to what I write. I can confidently say that anyone who enjoys my writing would likely enjoy those six publications as well. The 2025 list? Not so much, and vice versa.

I’d be intrigued to find out if the audience overlay list is as generic for you as well, so do let me know in the comments.

Retention

If you have paid subscriptions turned on, the Retention report is by far the most stressful report on this list. It’s purpose-built to make us feel bad and despair that nothing we’re doing is working.

Or maybe that’s just me.

Anyway, you get a chart showing the horrendous tug-of-war between new paid subscribers arriving and old paid subscribers departing. It’s a never-ending struggle to stay afloat. Important if you’re running your Substack as a business, of course, but if you’re just here to write it’s the sort of pressure that doesn’t really help.

Slightly less depressing is the paid cohort analysis, which show how long people are sticking around after subscribing:

Look a those little affirmative ‘Good’ pats on the head. I wonder what it does if your numbers are not good? Is it a red cross and the word ‘BAD’?

All that said, it is quite useful to get a sense for how long people are sticking around.

How much you want to learn from this stuff is up to you. For example, I can look at my retention over time and note that the people who subscribed in May did not stick around:

The retention from those May subscribers was unusually lower than in all other recent months. Interesting! Why was that? I have absolutely no idea, but if I could be bothered I could go back and trawl through my output that month to see what happened. Maybe I wrote something accidentally controversial on Notes, or Google search brought in a bunch of people that liked one article, then realised they’d made a terrible mistake.

What I’m actually going to do, though, is nod sagely in the direction of that chart, then ignore it and carry on with some writing. Not to repeat myself, but unless you’re actively running a business I’d generally advise you do the same.

Sharing

This report is a simple table showing which Substack users have brought you views and subscribers via direct sharing of your posts. You can re-sort according to views, subscribers or paid subscribers.

I have to admit, it’s a rather heart-warming to spot names that you know on the list. Perhaps more than the ‘audience overlap’, this might be a good source for potential collaborations.

Referrals

Not to be confused with sharing, or indeed with referral traffic from other websites or campaign, the Referrals report instead relates directly to ‘gifts’ that your paid subscribers can give to their friends.

This is something that can be turned on or off in your publication’s settings:

It’s slightly obscure, IMO.

Use or not, depending on your predilection for turning your customers into unpaid marketing interns.

Notes

Confusingly, ‘subscriber notes’ have absolutely nothing to do with Substack’s Twitter-like Notes platform. 🤷

Instead, these are nice little notes that paid subscribers can leave when they upgrade their subscription. They’re really useful for a) making you feel warm and fuzzy inside and b) providing insight into why people are subscribing.

You can also (with the person’s permission) download the quote and share it as a testimonial. This is one of my favourites:

I have no idea why this is in the ‘Stats’ area, because it clearly isn’t statistical in the slightest. Qualitative, I suppose?

Traffic

Modern long-form newsletter platforms, like Substack, Ghost, Beehiiv and Buttondown, provide decent websites alongside the email functionality. As such, web traffic is an important part of the overall equation, unlike with more trad marketing platforms such as Mailchimp which were designed for single mailshots that don’t exist outside of a subscriber’s inbox.

My traffic chart has weird regular spikes, twice a week:

Those are when I send my newsletters. Inevitably, people click buttons in the email and end up on the website. At the same time, a new post has the potential to bring in new readers who have seen a link somewhere.

The date range can be adjusted at the top of the page, all the way back to when you created your publication. In my case that’s July 2021, and going all the way back looks like this:

You can see the long game at work here. For the first year and a half, web traffic grew slowly but steadily. In early 2022 it started to pick up, then jumped a year later in autumn 2023. That’s when I started making ‘how to’ videos, incidentally — Substack features like Recommendations and Notes also played a part.

You can filter the chart by traffic source, which is when things get really interesting. Check out this mad thing:

Up until mid-2022 I didn’t really get any traffic from search (by which I mean Google search). That started building in late-2023, after I’d built up a decent back catalogue of material. It could also be that Substack improved their own overall SEO game.

But look at those two crazy months in 2024: October and November. Hands up if anyone can remember a big political thing that happened in October and November 2024?

So, I don’t write about politics. My fiction is full of politics and social commentary, of course, but I’m not ‘a political writer’. And I definitely wasn’t writing about the US election. And yet…!

My best theory is that the US election and Twitter’s transformation after Musk’s buy-out caused a large exodus of people to different platforms, including Substack. Plus there was, of course, an awful lot of political commentary happening on Substack. At that time, Googling ‘what is Substack?’ often pointed people towards my video of the same name:

What actually is Substack?

I’d like to say that this was some devious grand strategy from me, but it really wasn’t. I’ve always enjoyed making tutorials, and that video was intended to help out other fiction writers like myself. Fast forward a year to 2024 and there were a lot more people trying to answer the question ‘what is Substack?’

Two things to note here, though:

1. Those spikes did not last. Traffic went back to normal afterwards.

2. The surge in subscribers and even paid subscribers that I had during this period was short-lived. I suspect people liked the video, subscribed happily, and then a week later I sent them the latest chapter of my weird science fiction and fantasy crime serial and they ran away screaming.

Below the chart is a table, where you can sort by views, users and subs. This is pretty handy if you want to know what people are doing — you might find that you’re getting a lot of view from one source, but all your subscribers are coming from somewhere else.

Email

This report is slightly confusing as it crosses over with the individual post stats, while claiming to be email-specific. I suspect it’s a legacy terminology issue more than anything.

Anyway, I like this table! It can be sorted according to any of the columns, which is hugely useful in this case. If I order by ‘free subs’ I can see which of my articles encouraged the most sign-ups. I can sort by ‘paid subs’ as well, which is even more useful. Likes, comments and shares are also here, making it easy to see which posts had the most engagement. It seems that some of the stats are specific to when the email was sent (such as opens) and others are about subsequent behaviour of the post on the website (e.g. views, subscribes).

What you do with those insights is up to you, but I do like being able to poke at it. It can also be a useful way to dig into the archive and re-share previously successful posts.

Unsubscribes

Never look at this report.²

Surveys

Did you know you can set up and run fancy surveys from within your newsletter? Well, you can! I have one that automatically pings out to new subscribers in their welcome email. Quite a nice way to get some insights into who is subscribing (though I desperately need to update it).

Anyway, this is an entire other topic for another time.

Talking of which, I think there are some follow-ups to today’s post that I should probably delve into at some point, namely:

• Individual post stats, of which there are A LOT

• Podcast stats

• Google Search Console

• Google Analytics 4

Let me know if any of those sound like they’d be useful.

---

I quite like data. I really enjoy having these stats at my fingertips. But not everyone does, and that’s totally fine. Numbers like this can drive you crazy, especially if all you want to do is write some stories or journal posts.

Funnily enough, Substack have actually recognised this, adding a setting to hide a lot of the numbers:

You can find that in the ‘Privacy’ section of your publication’s settings.

Regardless of what you do with the data, remember to stay true to your original purpose.

1. This also means that if you’re just starting out and don’t have months of data to look at, try to avoid stressing about those ups and downs.

2. My favourite feedback I received from a former paid subscriber was this: