Lambda Workflow

The Lambda Workflow node lets you run complete workflows as single steps within other workflows. Think of it like calling a specialized helper that knows how to do a specific task perfectly every time.

This is incredibly powerful for building complex automations from smaller, reusable pieces. Instead of rebuilding the same logic over and over, you create it once and use it everywhere.

How it works

When your workflow reaches a Lambda Workflow node, it pauses, runs the specified sub-workflow with the data you provide, waits for it to complete, and then continues with the results.

graph LR
  Main[Main Workflow] --> Call{Lambda Workflow}
  Call --> Sub[Sub-Workflow]
  Sub --> Results[Return Results]
  Results --> Continue[Continue Main]
  style Call fill:#6d28d9,stroke:#fff,color:#fff

Setup guide

1. Choose Your Workflow: Select which workflow you want to run. This workflow must have a Lambda Input node to receive data.

2. Prepare the Data: Map the data from your current workflow to match what the sub-workflow expects to receive.

3. Configure Options: Set timeout limits and decide how to handle errors if the sub-workflow fails.

4. Connect the Results: The sub-workflow’s output becomes available for the next steps in your main workflow.

Practical example: Content processing pipeline

Let’s say you have a specialized workflow that analyzes any text content, and you want to use it in multiple different workflows.

Let’s say you have a specialized workflow that analyzes any text content, and you want to use it in multiple different workflows.

Main Workflow Data: Your main workflow has found an article and extracted some text:

• extractedText: “Artificial intelligence is transforming how we work…”

Calling the Sub-Workflow: You use the Lambda Workflow node to call a sub-workflow named “text-analyzer-v2”. You pass the extractedText as input.

The Results: The sub-workflow runs, does its magic, and returns a detailed analysis:

{
  "analysis": {
    "sentiment": "positive",
    "keyTopics": ["AI", "automation", "productivity"],
    "readingLevel": "intermediate",
    "wordCount": 847
  },
  "processingTime": 2300
}

Now your main workflow can use this data to make decisions or save reports.

Common use cases

Scenario	Why Use Lambda Workflow	Example
Repeated Logic	Same processing needed in multiple workflows	Text analysis, data validation, format conversion
Complex Operations	Break big workflows into manageable pieces	Multi-step data processing, content generation
Specialized Tasks	Reuse expert workflows across projects	AI analysis, web scraping, API integrations
Team Collaboration	Share workflow components between team members	Standard data processing, company-specific validations

Start with simple sub-workflows

Begin by creating sub-workflows that do one thing really well. You can always combine multiple Lambda Workflow nodes to handle complex scenarios.

Configuration options

Setting	Purpose	Recommended Value
Workflow ID	Which sub-workflow to run	Use descriptive names like “email-validator”
Timeout	How long to wait before giving up	30 seconds for most tasks, longer for AI processing
Retry Attempts	How many times to try if it fails	1-2 retries for network issues
Error Handling	What to do if the sub-workflow fails	Continue with default values or stop the main workflow

Troubleshooting

• “Workflow not found” errors: Check that the workflow ID exists and is published. Make sure you have permission to run it.
• Input validation failures: Verify that your input data matches exactly what the sub-workflow’s Lambda Input node expects.
• Timeout errors: Increase the timeout value or optimize the sub-workflow to run faster.
• Unexpected results: Check that the sub-workflow’s Lambda Output node is returning data in the format you expect.