Substack takes 10% of every paid subscription. Ghost takes zero and charges a flat monthly fee for hosting. Both numbers are real. Both answers are right for different writers. Most Substack vs Ghost posts end there with a verdict and a referral link.
What the verdict skips: both platforms ask the same prior question, and both answer it the same way. Which surface is the primary one, the email or the blog? Substack says email and treats the blog as the archive. Ghost says blog and lets you bolt the newsletter on top. The question itself decides more than the pricing does.
The pricing crossover, in real numbers
Substack is free to start. The cost is a cut on every paid subscription: 10% to Substack, plus Stripe's 2.9% + 30¢ per transaction, plus a small recurring billing fee. On a $10 monthly subscription the publisher keeps roughly $8.36. Across a year of 1,000 paid readers at $10 each, the platform cut alone is about $12,000. We worked the same math at 1k, 5k, and 25k paid subs in The Substack tax.
Ghost flips the model. The managed hosting product, Ghost(Pro), runs on a flat fee with Starter at $15 a month, Publisher at $29, and Business at $199 as of mid-2025. There is no cut on paid revenue. Stripe still applies; Ghost itself takes nothing from subscriptions.
The crossover is mechanical. A publisher earning about $290 a month from paid subs is already handing Substack $29 in cuts, so the Publisher tier pays for itself. At $3,000 a month, Substack is taking $300 and Ghost is taking $29. The bigger the paid revenue, the wider the gap.
There is also the self-hosted route. Ghost is open source. If you can run a server, the software itself is free; you pay only for a small droplet and your own time on updates, deliverability, and backups. That removes the price floor entirely and adds a different bill in maintenance.
On math alone, Substack wins before paid revenue exists. Ghost wins after. The bend is at a few hundred dollars a month, sooner if the publication is on the Business tier for the wrong reason.
Where each one is honestly better
Substack's advantage is the network. Notes circulates posts between publications. Subscriber recommendations widen the funnel without a paid acquisition channel. A new writer on Substack inherits a small amount of audience momentum from being on Substack at all. None of that is fakeable on a self-hosted blog.
The editor is also simpler. Fewer settings, no theme repository to evaluate, no DNS to configure beyond a custom domain. A writer who only wants to write and send is live in twenty minutes.
Ghost's advantage is ownership and craft. Custom themes, custom domain by default, full SEO control, edits to old posts that update the public version without re-sending the email. The publication is on your domain, indexed at your domain, and the back catalog earns search traffic for years. That is the reason most journalist-owned independent publications sit on Ghost: 404 Media, for example, launched on a Stripe account and Ghost and was profitable inside six months.
Where Substack quietly forces a tradeoff, Ghost usually offers a setting. That cuts both ways. The writer who only wants to write can find Ghost dense. The publisher running a real archive will find Substack thin.
What both have in common, and why it matters
Strip the differences and what's left is the same structural choice, made for you in advance.
On Substack, the email is the product. The blog page is where past emails live. Every URL is shaped around the send. Search-engine traffic to old posts is treated as a side effect of the email, not the point.
On Ghost, the blog is the product. The email is a feature you turn on per post. The default reader path is somebody arriving from search, reading a post, and maybe subscribing on the way out. The send is one of the things the publication does, not the thing the publication is.
Both can perform the other side. Neither is built around it. The asymmetry shows up in the small details: which analytics get pride of place, whether tags refer to email segments or post categories, what happens when you edit an old piece. Each platform made a clear call about which surface gets the love, and the other surface gets adequate.
That call was right when the trade-off was real. Today, most independent publishers want both surfaces at full strength: a send that lands cleanly in the inbox, and a post that ranks at their own domain, from the same draft, without two copies of anything.
The frame both skip
A publishing OS is the category neither Substack nor Ghost is in. The frame is older than either of them and it's gaining ground for one reason: the blog and the inbox stopped being separate jobs.
The shape is plain. One draft becomes both the email and the blog post in a single publish, on the publisher's own domain, with one URL the search engine and the inbox both point to. One subscriber list, tagged at the page the reader came from. One analytics layer. No senior surface; both surfaces equal, on day one.
That frame answers a different question than "Substack or Ghost." It answers "what should a publishing tool look like when neither the email nor the blog is the senior surface." Nashra is one option in the category. A few others sit nearby.
How to decide
Three rules cover most of the choice.
Pick Substack if the network is the point. You write a personal essay, the Notes feed plus subscriber recommendations is a meaningful share of how new readers find you, and you accept the 10% cut as the price of distribution.
Pick Ghost if ownership is the point. You want the publication on your own domain, themed your way, with the back catalog earning search traffic, and you're fine running the email as a feature attached to the blog.
Pick a publishing OS if neither surface is meant to be senior. Half your subscribers find you through search; the other half forward sample sends to friends. You want one draft to publish to both surfaces at once, on your domain, with one subscriber list underneath. That's a different category, not better than Substack or Ghost at what they do, but different enough that the comparison title is misleading.
The short version
Substack and Ghost are both good at what they were built for. Substack's pricing wins until paid revenue is real. Ghost's wins after. Both leave you to pick which surface, email or blog, runs the publication.
That choice matters more than the pricing. A subscriber converts roughly 10× better than a follower. That part doesn't change with the tool you pick. What changes is how many of those subscribers arrived through a post the search engine sent versus an email a friend forwarded. If both numbers are real, the comparison worth running isn't the one in the title.
