Self-hosted Zapier alternative: unmetered workflow automation with Automatisch

Automation is the quiet glue of a working business -- the thing that moves a form submission into a spreadsheet, pings a channel when a payment clears, files a support email where someone will actually see it. Zapier figured out years ago that this glue is worth metering. Every task is a tick on a counter, every counter has a ceiling, and every ceiling has an upgrade button. Meanwhile the automations you built -- your logic, your API keys -- live on someone else's cloud. Automatisch is the same idea, on hardware you control.

What Automatisch is

Automatisch is an open-source workflow automation platform -- a self-hosted Zapier that runs entirely on your own server. You build *flows*: a trigger fires -- a webhook arrives, a schedule ticks over, a new record shows up in a connected app -- and then a chain of actions runs, each step handing its output to the next. It connects to more than 75 services, from Slack, Discord, Telegram and Gmail to Notion, Trello, Stripe, Salesforce, HubSpot and the major AI providers, plus a set of built-in utilities for the plumbing in between.

This isn't a stripped-down demo. It's the full builder: multi-step flows, multiple credentials per app, filters and formatters, and raw HTTP calls to anything with an API. Where Zapier gates multi-step Zaps, premium apps and higher task volumes behind escalating tiers, Automatisch simply runs. The only ceiling is the server it sits on.

The features that actually matter

• Visual flow builder. Chain a trigger to one or more actions, wiring the output of each step into the input of the next. This is Zapier's core editor, minus the task meter counting down in the corner.
• 75+ app integrations. Slack, Discord, Telegram, Gmail, Notion, Trello, ClickUp, Monday, Dropbox, Google Drive, PostgreSQL, Stripe, Salesforce, HubSpot, Zendesk, Typeform, GitHub, Twilio and more -- including OpenAI, Anthropic and Mistral for AI-powered steps.
• Webhooks, in and out. The Webhooks trigger hands every flow a unique URL that fires the instant an external system posts to it, and the HTTP Request action lets a flow call any REST API. Between them, even unlisted services are reachable.
• Scheduler. Run a flow on a cadence -- hourly, daily, or on chosen days -- without hand-wiring cron.
• Filter and Formatter. Filter halts a flow unless your conditions are met, so actions only fire on the records that matter; Formatter reshapes text, numbers and dates between steps. On Zapier both live inside multi-step Zaps, which sit behind the paid plans.
• Datastore and Delay. Datastore is a built-in key-value store that persists across runs -- deduplicate records, remember state between executions. Delay pauses a flow before the next step runs.
• Multiple connections per app. Store several accounts for the same service and choose which one a given step uses, so a single instance can serve every team and client you run.

Why self-host it instead of the hosted version

Your data stays yours. Every credential and every record that passes through a flow lives on your server, not on a third party's. Automatisch is explicitly built for teams handling sensitive data under regimes like GDPR, where "it's on a US SaaS somewhere" isn't an acceptable answer. Your API keys for Stripe, your customer emails, your internal payloads -- none of it leaves the box.

The cost doesn't scale per task or per seat. Zapier bills by the task, and a single busy flow can chew through a monthly allotment in days; adding a teammate adds a seat. Self-hosted, the automation is free to run as often as it needs to, for as many people as you like. Your bill is the server, and the server doesn't care how many tasks you fired.

You own the instance. It's open source, so there's no vendor deciding to sunset an integration, change pricing, or lock a feature you rely on behind a new tier. You update on your schedule, keep the version that works, and answer to nobody's roadmap but your own.

Who it's for

Agencies and consultants running automations on behalf of many clients -- one instance, multiple connections per app, and no per-client Zapier seat multiplying the bill.

Privacy-sensitive teams in healthcare, legal, finance or anywhere under GDPR, whose data genuinely cannot sit on someone else's cloud while it moves between systems.

Solo builders and small teams who started on Zapier's free tier, watched task volume creep up, and hit the point where the monthly invoice costs more than a small server that would run the same flows without a meter.

The honest setup notes

Automatisch is a real application with real dependencies, and self-hosting it means owning them. It needs a PostgreSQL database -- that's where your flows, connections and run history live, so it's the thing you must back up. It needs Redis for the background worker that actually executes flows on schedule. The official install path is Docker Compose, which bundles all of it, but you'll still handle updates yourself: pull the new image, redeploy, check it came back up. If you want email steps or notifications you'll configure SMTP, and you must set and safeguard an encryption key -- lose it and every stored connection becomes unreadable.

None of this is hard. It's just yours to keep alive, and the worker has to stay running for a scheduled flow to fire at 3am. That's the honest cost of owning it -- or you let us run it, and the database backups, updates and uptime become our problem instead.

Deploy it in one click

Rolling your own means standing up Docker, wiring Postgres and Redis, generating that encryption key, terminating TLS, and keeping the stack patched -- an afternoon of yak-shaving before your first flow runs, and a standing chore after.

On Caliber Node it's the opposite. You deploy Automatisch in one click, and the database, the Redis worker, HTTPS and the encryption key are all provisioned and managed for you. You get a running instance at your own subdomain with the credentials handed to you -- nothing to configure, nothing to babysit. It's still your Automatisch: your flows, your connections, your data, fully exportable whenever you want it. We just handle the parts that are infrastructure instead of automation.

First run

1. Open your Automatisch dashboard and create a new flow.
2. Choose a trigger -- try the Scheduler for a timed run, or the Webhooks trigger to get a unique URL you can point another app at.
3. Add an action step and connect an app -- pick Slack, say, and add your credentials as a connection.
4. Map the trigger's output fields into the action so the right data lands in the right place.
5. Switch the flow on. It's live, and it'll keep running until you turn it off.

That's a working automation -- from empty canvas to something firing on its own in about five minutes.

The bigger picture

Automation is leverage: the compounding advantage of a business where the boring things happen by themselves. That leverage is too central to rent by the task from a vendor who can reprice it, meter it, or read the data flowing through it. Owning the tool that runs your automations is the same instinct as owning your files or your inbox -- it's core infrastructure, and core infrastructure belongs to you. Automatisch gives you the whole capability with none of the counter, and hosting it yourself turns a recurring per-task bill into a flat, predictable one.

Deploy Automatisch in a few minutes →

Or browse the full app catalog to see what else you can own instead of rent.

One-click self-hosted apps, managed for you -- real logs, SFTP, and your data always exportable. Never a black box.