Spojit
Branching & parallel

Branch on anything.
Run it all at once.

Real control flow on the canvas. Branch on any condition, loop over a list, and run independent paths at the same time with parallel branches that fan out and join back. Nest it as deep as the logic needs, no concurrency code required.

New orderfork
Check inventory
running
Reserve payment
running
Notify warehouse
running
Join · confirm order
How it works

Decide, fan out, join

01

Branch on a condition

Send the run down the right path with if/else logic based on the data in front of it.

02

Run paths in parallel

Kick off independent steps at the same time instead of waiting for each one in turn.

03

Join the results

Fan the parallel branches back in and carry the combined result into the next step.

Capabilities

Control flow without code

If/else branching

Choose a path based on a condition, so the workflow does the right thing for each case.

Parallel branches

Run independent work concurrently and cut total runtime, with no threads to manage yourself.

Loops

ForEach over a collection, or while a condition holds, to repeat a step as many times as needed.

Fan-out and join

Split into parallel paths and merge them back together when every branch has finished.

Compound conditions

Combine multiple checks into one decision, so complex routing stays in a single clear node.

Nest deeply

Put branches inside loops inside parallel paths, several levels down, and keep it readable.

Use cases

Real logic, real speed

Route by condition

Send each case down the path it needs, from approvals to tiers to regional handling.

Call APIs at once

Hit several services in parallel and gather the results, instead of one slow call after another.

Process a list

Loop over every item in a collection and run the same steps for each one.

Approve-or-reject paths

Take one path when a check passes and another when it fails, cleanly and visibly.

The difference

Parallel vs one step at a time

Run independent work concurrently and slash runtime, without writing any concurrency code.

Branch and parallelize when

  • The path depends on the data
  • Independent steps can safely run at once
  • You want speed without writing concurrency code

Sequential-only means

  • Every step waits on the one before it
  • Multi-API runs are needlessly slow
  • Branching logic hides in code, not on the canvas

Branch and run in parallel

Start free and build a workflow that branches on your data and runs paths at once. No card needed.