Skip to content

Workflow AI

Describe the process. AI builds the draft.

Turn your idea into a visible workflow. Refine it in chat, review its settings and test before launch.

Here’s what you can do.

Three examples show how you can use Workflow AI.

01

Start in your own words

What should start the process, and what happens next? Describe a form submission followed by a task for your team, for example.

A chat request produces a sample draft connecting form submission to task creation.
Simplified product example · not a live product view
02

Change the steps you choose

Select a step and describe the change. Extend a wait, for example, without rebuilding the whole process.

A selected wait step is changed to 48 hours through a targeted chat request.
Simplified product example · not a live product view
03

Complete the draft before it goes live

Complete missing details such as the form, recipient or connection. Review triggers, conditions and actions, then test before publishing.

Before launch, the workflow shows an incomplete recipient selection in its settings.
Simplified product example · not a live product view

What this changes for your day-to-day work.

The practical benefits of Workflow AI.

Plan before you build

Discuss the idea in chat mode first. Switch to building once the trigger and next steps are clear.

Chat mode discusses a process before a draft is created.

Speak your workflow idea

Dictate your process through the microphone. Then review the transcription and generated draft.

A microphone transcribes a spoken idea into the workflow input.

Understand each step

Get explanations for supported triggers and actions inside the builder. Ask a follow-up when a setting is unclear.

AI explains a selected wait action alongside the workflow.

Change rules in your own words

Describe changes to supported settings, such as sending hours or re-entry. Check the values actually applied.

A chat instruction disables re-entry in the sample workflow.

Recover an earlier draft

Open and restore an earlier AI version when needed. If autosave is off, save changes before leaving.

Two AI drafts are shown in version history with a restore option.

Simplified examples, not live account data.

What you need to get started.

We check availability in your account and agree on rules, permissions and connected services. Messaging, additional actions and external apps can have separate charges or limits.

Workflow AI helps create and revise a process. The Workflow Builder runs the configured automation. AI actions that process data during a run are a separate feature.

Availability and any add-on or usage costs are confirmed for your ONE LOOP account before implementation.

Your questions, answered.

Do I need workflow experience?

You can start in everyday language. A clear trigger, intended outcome and specific timing help. The assistant can explain steps, but responsibilities, recipients and effects still need to be understood before approval.

Can I edit the draft manually?

Yes. The generated process opens in the normal Workflow Builder. Edit supported steps in chat or directly in the editor. For targeted AI changes, select the relevant steps or specify the scope explicitly.

Does AI automatically test that everything works?

No. Warnings about missing fields or connections do not replace testing. Use suitable test contacts and recipients; configured messages or actions may actually execute. Continue monitoring errors and results after launch.

Are saving and publishing the same thing?

No. Saving preserves your work. Complete missing settings and resolve detected blocking errors before going live. Publish only after review and testing. The absence of warnings does not guarantee an error-free workflow. Changes to published workflows can also affect contacts already progressing through them.

See it with your own use case.

Let’s explore your use case and what the setup requires.

Book a demo