Connect everything you already run. Publish it as one API.
Draw a flow on a canvas. Point it at the databases, files, APIs and AI models you already have. Publish it as a scheduled job, a secure endpoint, or a full app — on your own servers.
Free plan · No credit card · Your data stays on your servers
Works with the systems you already run
What you can do with it
- You have full control. Every step, every version and every permission is yours to set.
- You control everything. From the first read to the last delivery — on one canvas.
- You connect every system. Databases, APIs, files, email, directories and servers.
- You draw every flow. Branch, loop, run in parallel and merge back together.
- You map any format. Turn one system’s fields into another’s, visually.
- You schedule anything. Cron with time zones, webhooks, file drops and sync jobs.
- You publish your own APIs. Secure, documented endpoints — rate limits and keys included.
- You build the front end too. Forms, portals and multi-page apps on top of your pipelines.
- You put AI to work. LLM, document extraction, agents and vector search as steps.
- You talk to your customers. Chat flows on your website, WhatsApp and SMS.
- You see every run. The input and output of each step, painted on the canvas.
- You know the moment it breaks. Alerts by email and push — then replay from the failed step.
- You keep your data at home. Self-hosted. Nothing leaves your servers.
- You decide who sees what. Organisations, projects, roles and permissions.
- You ship changes safely. Drafts, published versions, rollback and environments.
- You have all the details. Payloads, timings, and an audit trail of every change.
- You own the whole thing. One platform, end to end — on your terms.
Someone in your company is the integration.
Every business has work that lives between systems — the POS and the accounts, the core system and the card processor, the supplier’s portal and the spreadsheet on the shared drive.
Nobody owns that space, so a person does: exporting, re-keying, reconciling, chasing. SpanFlow gives that work to a pipeline, and gives the person their day back.
Draw. Connect. Publish.
Draw the flow
Drag steps onto the canvas: read, check, transform, branch, loop, deliver. Start from a template if you like — or describe it and let AI draft it.
Connect your systems
Add connections once — databases, APIs, file stores, mail, directories, AI models — and test them right in the browser.
Publish it
Run it on a schedule, on a webhook, as a documented API, behind a form, or in a WhatsApp conversation. Watch every run live.
Everything connects through one canvas.
Automation, ETL and data integration, graph pipelines built from nodes, data mapping, webhooks and cron schedules, API publishing, front ends, AI and MCP — one platform instead of a toolbox of scripts.
From a nightly sync to a full product.
One canvas, one engine, one place to see what ran — instead of a different tool for every job.
Visual pipelines
Drag steps onto a canvas. Branch, loop, run in parallel and merge back.
Learn moreConnect everything
Databases, APIs, files, email, directories, servers — the systems you already run.
Learn moreTriggers & schedules
Run on a schedule, on a webhook, on a new file or record, or on demand.
Learn morePublish as an API
Turn any pipeline into a secure, documented endpoint with its own URL.
Learn moreFront ends included
Forms, portals and multi-page apps on top of your endpoints — designed, written or uploaded.
Learn moreAI built in
LLM, document extraction, agents and vector search as pipeline steps — with cost control.
Learn moreData mapping
Turn one system’s format into another’s with a visual mapping designer.
Learn moreChat flows & campaigns
Build conversational flows and reach customers on WhatsApp, SMS and web chat.
Learn moreMonitoring & alerts
See every run, live. Get alerts on your phone before your customers notice.
Learn moreSecurity & self-hosting
Runs on your servers. Your data never leaves. Built for multi-team organisations.
Learn moreBuilt for the work your industry actually does.
End-of-day reconciliation
The officer arrives to an exception list. Matched transactions are already filed; every break is in their inbox.
- 1Core ledger
- 2Settlement file (SFTP)
- 3Match & compare
- 4File the matched set
- 5Email each break
The nightly sales close
Cash and card variances are caught the night they happen, per location, not at month end.
- 1POS totals
- 2Card settlement
- 3Join per branch
- 4Flag variances
- 5Email the managers
Result routing
Every file is validated; normal results flow into the lab system, critical values reach a phone at once.
- 1File drop (SFTP)
- 2Validate identifiers & ranges
- 3Deliver to the LIS
- 4Critical → push alert
Track and trace
Carrier statuses land in one table in one vocabulary. Customers ask one endpoint.
- 1Each carrier API
- 2Normalise statuses
- 3Operations database
- 4Alert on exceptions
Quote comparison
One endpoint asks every insurer in parallel and returns a ranked list on the broker’s own criteria.
- 1POST /quote
- 2Insurers in parallel
- 3Join & rank
- 4One response
Usage file ingestion
Files are chunked, validated and loaded into billing; malformed records go to quarantine.
- 1Mediation file (SFTP)
- 2Split into chunks
- 3Validate records
- 4Deliver to billing
Order intake from several marketplaces
Each marketplace webhook maps into one order format; validate → reserve stock → notify runs as one unit.
- 1Marketplace webhooks
- 2Map to house format
- 3Validate
- 4Reserve stock
- 5Notify
A department portal
A published app with a multi-page front end, over the department’s existing database.
- 1Portal page
- 2Published app endpoints
- 3Department database
- 4Notifications
Friendly for business users. Serious for developers.
- JavaScript and C# code steps in a sandbox
- Auto-generated OpenAPI docs for every published endpoint
- Import / export everything as one reviewed bundle
- Pipelines and data exposed to AI assistants over MCP
- Client SDKs for web and .NET
Secure by design, hosted by you.
- Runs on your own servers with Docker
- Organisations, projects, roles and permissions
- Encrypted secrets, hashed API keys, audit trail
- Rate limits, IP allowlists and flood protection
- Scale out with more instances and worker pools
Start free. Grow when you’re ready.
Free
$0
For trying it out and small automations. Pipelines, apps and published APIs included.
Start freePro
Contact us
For teams running real work: AI steps, hosted pages, MCP, chat flows and more connectors.
Contact salesEnterprise
Contact us
For organisations: no limits, white-label domains, messaging channels, C# and web automation.
Contact salesQuestions, answered.
What is SpanFlow?
SpanFlow is a self-hosted integration platform. You draw a flow on a visual canvas, connect it to the databases, files, APIs, mailboxes and AI models you already use, and publish it as a secure endpoint, a scheduled job, a chat flow or a full app.
Do I need to know how to code?
No. Most pipelines are built entirely on the canvas with forms for every step. When you do need code, JavaScript and C# steps run in a safe sandbox.
Where does my data live?
On your servers. SpanFlow is self-hosted with Docker, so your data and credentials stay inside your network. Connection secrets are encrypted at rest.
How is SpanFlow different from Zapier or n8n?
SpanFlow is built for business data work: real database writes in batches, file stores like SFTP and S3, parallel branches, versioning with rollback, and the ability to publish any flow as a documented API with its own front end.
Can I try it for free?
Yes. The Free plan includes three projects with pipelines, apps and publishing. Upgrade when you need more projects or advanced channels.
Can AI build pipelines for me?
Yes. Describe what you want in plain English and the authoring assistant drafts it as a bundle you review before importing. AI models can also run as steps inside your pipelines.
Stop moving data by hand.
Start free with three projects, or talk to us about running it across your organisation.