← Blog

Hatchbox raised its price from $10 to $15 a server, and that's a healthy sign for a small team

Railyard

October 24, 2026 · 8 min read

In early September 2026, Hatchbox founder Chris Oliver told customers the platform’s per-server price was going up, from $10 a month to $15. Hatchbox’s own pricing page now lists a single plan at $15 per server per month, unlimited apps, unlimited team members, included. The jump is confirmed: the question worth actually answering is what it means, and what it doesn’t.

It doesn’t mean Hatchbox did anything wrong. A 50% increase on a product that’s been priced at $10 a server for years, from a two-person team that also runs a video tutorial business (GoRails) under real revenue pressure as more developers lean on AI tools to write and learn code, is not a scandal. It’s closer to the opposite: a small, well-regarded shop charging what the product is actually worth, instead of quietly under-charging until the business can’t sustain the thing people depend on. One longtime Hatchbox user’s public reaction to the news put it plainly: it was “criminally underpriced before and still is now,” and he’d happily pay three or four times as much for it. That’s not the reaction a price increase usually gets, and it’s worth sitting with for a second before reaching for a contrast.

What actually changed

Hatchbox is a deployment platform for Rails, Ruby and Node apps that runs on servers you provide, your own DigitalOcean, Hetzner or AWS account, not Hatchbox’s infrastructure. That’s an important detail to hold onto, because it’s the same basic shape Railyard uses: the platform is a control plane, the server is yours, and the bill from your cloud provider is separate from the bill to the company running the deploys. Hatchbox has billed this control-plane layer per server since the beginning: one flat price for whatever you run on that box, as many apps as you want. For a long time that price was $10 a month per server. As of the September 2026 announcement, and reflected now on Hatchbox’s own pricing page, it’s $15.

Why did Hatchbox raise its price?

The reasoning Chris Oliver gave, as reported in the indie-hosting newsletter coverage of the announcement and echoed on his own GoRails podcast a couple of weeks later, was straightforward: a price that’s too low for too long starves the thing people rely on of the resources needed to keep improving it. That’s a real tradeoff for any small, bootstrapped operator, and it’s one worth taking at face value rather than looking for a hidden motive. A two-person team supporting a hosting product used by working developers has fixed costs that don’t scale down just because the price per customer stayed flat for years: support time, server automation maintenance, security patching across every customer’s infrastructure, the ordinary cost of keeping a deploy pipeline from silently rotting. If the price doesn’t move, eventually the service does, and that’s a worse outcome for the exact customers a low price was supposed to protect.

Why “per server” is the detail that matters

The interesting part isn’t that the number went up. It’s what kind of number it is. Hatchbox bills per server under management. If you run one server, you pay one $15 (previously $10) fee. If your app grows and you add a second server, maybe a dedicated worker box, maybe a standby database host, maybe a staging environment on its own machine, you now pay two. A team running five servers under Hatchbox’s new pricing pays $75 a month in platform fees before a single dollar goes to the cloud provider actually running the workload. Under the old pricing, that same five-server setup was $50. Either way, the platform fee is a linear function of how many machines you’re running, independent of team size, traffic, or how much of that infrastructure is actually load-bearing versus a cheap staging box that barely gets used.

That’s not a flaw. It’s a coherent, defensible model, and it maps cleanly onto Hatchbox’s own cost structure: more servers under management is more surface area for Hatchbox to automate, monitor and support, so a per-server fee scales roughly with the actual work the platform is doing on your behalf. For a team that runs one or two production servers for the life of the product, it’s also a genuinely cheap way to get a managed deploy pipeline on infrastructure you own. The 50% increase changes the specific number, not the shape of the bill.

What a flat per-team price does instead

Railyard’s pricing is built on a different axis. There’s no per-server line item at all. Free includes one application server plus a database server for three apps. Pro is $19 a month per team and covers up to five servers with unlimited apps. Team is $49 a month per team with unlimited servers, single sign-on and an audit log. The number that moves as you grow is the plan tier, not a per-unit meter: going from one server to five costs exactly the same $19, because the fee is attached to the team using the platform, not to the count of machines that team happens to be running that month.

The practical difference shows up exactly in the scenario that triggers Hatchbox’s per-server fee: adding infrastructure. A team on Railyard’s Pro plan that adds a dedicated background-worker server, then a standby database host, then a staging box, goes from one server to four without the platform bill moving at all, because all four are still inside the five-server Pro allowance. A team on Hatchbox’s new pricing doing the identical thing watches its hosting-tool bill climb from $15 to $60 a month, in $15 increments, for adding infrastructure that has nothing to do with how many people are on the team or how much the product itself has grown. Neither number is wrong. They’re measuring different things: Hatchbox is pricing the infrastructure footprint it’s managing, Railyard is pricing the team using the control plane, and infrastructure footprint happens to be free to grow inside a plan’s ceiling.

It’s also worth being honest about where Railyard’s model has its own version of the same tradeoff. The flat fee isn’t infinite: Pro’s five-server ceiling is a real limit, and a team that outgrows it moves to Team at $49, a jump rather than a smooth per-unit increase. A team that needs exactly six servers and nothing else Team offers is paying for headroom it isn’t using yet, in the same way a Hatchbox customer running one lightly-loaded server pays the same $15 as a customer running one heavily-loaded one. Every pricing model draws its lines somewhere; the honest comparison is about where the lines are, not whether one side has none.

Two small teams, two reasonable answers

It’s worth stating directly what this incident actually is: a well-run, respected hosting business recalibrating a price that had gone stale, explained candidly by a founder who’s been transparent about the pressures a small, bootstrapped team operates under. Nothing about the September 2026 increase suggests Hatchbox mismanaged anything or is in trouble. If anything, a founder willing to say “this was underpriced and that’s not sustainable for us to keep improving it” in public is the kind of transparency that’s rare in hosting, where price changes more often arrive as a quiet line in a billing email with no explanation attached.

What the incident actually illustrates isn’t a verdict on either company, it’s a clean natural experiment in the two defensible ways to price a BYO-server control plane. Bill per unit of infrastructure, and the price tracks server count exactly, which is simple to reason about and fair to the platform’s own costs, but means normal infrastructure growth (a worker box here, a staging server there) moves the bill in lockstep. Bill per team with a plan ceiling, and infrastructure inside that ceiling is free to grow, but the ceiling itself is a step function rather than a smooth line, and a team near the edge of a tier is implicitly paying for room it hasn’t grown into yet.

Both are honest pricing strategies for the same underlying idea, software that manages deploys on servers the customer owns rather than infrastructure the platform owns. The real signal to take from a Hatchbox price increase isn’t “per-server pricing is bad.” It’s that per-server pricing has a mechanical consequence (the bill moves with infrastructure, not team size or usage), and a team evaluating either platform should model their actual server count against both pricing shapes before picking one, rather than comparing the headline numbers in isolation. A one-server side project is cheaper on Hatchbox even after the increase. A team running four or five servers for workers, standbys and staging is very likely cheaper on a flat per-team plan. Neither fact makes the other platform’s pricing wrong.

FAQ

How much did Hatchbox’s price increase? Hatchbox’s per-server fee rose from $10 a month to $15 a month, a 50% increase, announced by founder Chris Oliver in early September 2026 and reflected on Hatchbox’s own pricing page since.

Does Hatchbox still charge per server? Yes. Hatchbox’s pricing is unchanged in shape, one flat fee per server under management, with unlimited apps and team members included. Only the per-server number moved, from $10 to $15.

Is Railyard cheaper than Hatchbox? It depends on how many servers a team runs. A single-server setup is cheaper on Hatchbox even after the increase. A team running four or five servers for workers, standbys and staging is typically cheaper on Railyard’s flat $19-a-month Pro plan, because Railyard doesn’t add a per-server charge.

Did existing Hatchbox customers keep the old $10 price? Coverage of the announcement indicates the new $15 price applies broadly rather than grandfathering existing customers at $10, though Hatchbox’s own pricing page is the definitive current source for any specific account.

Sources

See how the two models compare directly: Compare Railyard to Cloud 66, the closest architectural relative to both platforms, view the full pricing page, or see how Railyard works on servers you already own.

See the flat, per-team alternative.

One price for the whole team, not one price per server you add.

Free plan, no card, and 14 days of Pro for every new team. One flat price per team.