What is Flow? (automation pipelines)

This article explains what Flow is, how it differs from rendering designs one at a time, and what you can build with the visual pipeline builder.

What is Flow?

Flow is a visual pipeline builder that lets you automate a full batch of photos from a single canvas. Instead of opening each design, dropping in a photo, and sending it individually, you wire up a sequence of steps — and Flow runs every photo through that sequence for you.

The Flows tab appears in the top navigation when Flow is enabled for your organization. The tab carries a Beta chip — Flow is an early-access feature actively being developed.

How Flow differs from one-at-a-time rendering

Normally, creating a finished image in Templified looks like this: open a design, place a photo, configure the template, then send it to a destination. That works well for small batches and one-off edits, but it doesn't scale to dozens or hundreds of photos.

Flow flips the model: you describe the pipeline once, then run it against any number of photos.

One design at a time	Flow
Open a design, drop in one photo, send	Drop all photos into the flow at once
Repeat for every subject	Every step runs on every photo automatically
Manual review between steps	Steps chain together without stopping
Triggered by a person opening the app	Can trigger automatically via a webhook when a new photo arrives

The Flows tab and Beta chip

When Flow is enabled for your organization, a Flows tab appears in the top navigation bar. The Beta chip next to the heading signals that Flow is in early access — the core pipeline is fully functional, but the feature set is still growing.

If you don't see the Flows tab, your organization doesn't have Flow enabled yet. Contact your account administrator or reach out to support to request access.

The visual canvas

The Flow builder canvas showing a Photos start node connected to a Remove Background step and a Create Image step

Opening a flow (or creating a new one) takes you to a canvas where each step is a node you can drag and connect. The canvas is built around a left sidebar called Steps, which lists every available step type you can drag onto the canvas:

Remove Background — cuts out the subject and detects the face.
Text — derives a new per-photo field from an existing one (for example, extracting a session name from a longer event string).
Create Image — renders a template into a finished image.
Create Editable Image — creates an editable design in Studio instead of a finished render.
Create Team Photo — fans all the run's photos into a single team-layout design.
Send To — dispatches a finished image or design to a send preset or destination.

Every flow starts with a Photos node that you can't delete — it's where photos enter the pipeline. You can upload photos manually when you hit Run, or switch the Photos node to Webhook mode so an external system can send photos in automatically.

Running a flow

Once your pipeline is wired up, click Run in the editor header. Flow validates the graph (checks that all required fields are filled and that connections between steps are compatible), then processes every photo through each step in order. A results panel opens alongside the canvas and shows the status of each step as it completes.

You can revisit past runs from the Runs dropdown in the editor header — each run shows per-step status and links to any designs or images it created.

What you can build

A few common pipeline shapes:

Batch render: Photos → Create Image → Send To. The simplest pipeline: render every photo against one template and deliver the images to a destination.
Background removal + render: Photos → Remove Background → Create Image → Send To. Automatically cuts out every subject before rendering.
Team photo: Photos → Remove Background → Create Team Photo → Send To. Combine individual cut-outs into a team layout in one pass.
Per-photo text from a CSV: Upload a CSV on the Photos node; columns like name and jersey number become variables you can insert into any text field on a Create Image or Create Editable Image step.
Webhook-triggered pipeline: Switch the Photos node to Webhook mode and point your photo-capture system at the flow's URL — each incoming photo triggers a run automatically.