Automation Logic

Master the fundamentals of workflow automation with this guide to triggers, actions, and webhooks.

The "If This, Then That" Paradigm

All automation follows a simple pattern:

Core Concepts

Triggers

Triggers are events that start your automation. They answer the question: "When should this run?"

Common Trigger Types:

• Form submission
• New email received
• Calendar event created
• Database record updated
• Webhook received

Actions

Actions are what happen after a trigger fires. They answer: "What should we do?"

Common Action Types:

• Create record
• Send notification
• Update spreadsheet
• Call API endpoint
• Generate document

Conditions

Conditions filter when actions should run. They answer: "Should we proceed?"

Real-World Example: Typeform to Slack

Here's a practical automation that notifies your team when someone submits a form:

Configuration Steps

1. Trigger: New Typeform Submission

   • Select your form
   • Map response fields

2. Filter: Check Lead Score

   IF {{score}} >= 80

3. Action: Send Slack Message

   Channel: #sales
   Message: "New hot lead! {{name}} from {{company}}"

Webhooks Deep Dive

Webhooks enable real-time communication between applications.

Incoming Webhooks

Receive data from external services:

{
  "event": "form.submitted",
  "timestamp": "2024-01-15T10:30:00Z",
  "data": {
    "name": "John Doe",
    "email": "john@example.com",
    "score": 85
  }
}

Outgoing Webhooks

Send data to external services when events occur in your system.

Best Practices

1. Error Handling - Always have a fallback path
2. Rate Limits - Respect API limitations
3. Logging - Track automation runs for debugging
4. Testing - Use test mode before going live
5. Documentation - Document complex workflows

Debugging Workflows

When automations fail, systematic debugging saves hours of frustration.

Common Failure Points

Symptom	Likely Cause	Solution
Trigger not firing	Permissions expired	Re-authenticate the connection
Action fails silently	Missing required field	Check all required fields are mapped
Partial data	Field mapping changed	Verify source field names haven't changed
Intermittent failures	Rate limiting	Add delays between actions
Wrong data format	Type mismatch	Add a formatter step to convert types

The Debug Checklist

When an automation fails, work through this sequence:

1. Check the run history - Most platforms show exactly where the failure occurred
2. Verify credentials - Connections expire; re-authenticate if needed
3. Test with sample data - Use a known-good input to isolate the issue
4. Review recent changes - Did the source app update their API?
5. Check rate limits - Are you hitting the platform's usage caps?

Reading Error Messages

Error messages often contain the solution. Common patterns:

"401 Unauthorized" → Re-authenticate the connection
"404 Not Found" → Record ID is invalid or deleted
"429 Too Many Requests" → Add delays between actions
"500 Internal Server Error" → Try again later; not your fault
"Field 'email' is required" → Ensure the email field is mapped

Error Handling Patterns

Build resilient automations that gracefully handle failures.

Pattern 1: The Safety Net

Always include an "else" path for unexpected conditions:

Pattern 2: Retry with Delay

For flaky connections, implement automatic retries:

1. First attempt - Run immediately
2. Failure detected - Wait 5 minutes
3. Second attempt - Try again
4. If still failing - Alert a human

Pattern 3: The Dead Letter Queue

For critical processes, never lose data:

1. If the main action fails, write the record to a "failed" spreadsheet
2. Process failed records manually or with a separate automation
3. Include enough context to retry later

Multi-Step Workflow Strategies

Complex automations require careful orchestration.

Sequential vs. Parallel

Sequential: Each step waits for the previous one to complete

• Use when: Step B depends on data from Step A
• Example: Create contact → Wait → Add to campaign

Parallel: Steps run simultaneously

• Use when: Steps are independent
• Example: Send email AND update CRM AND log to spreadsheet

Managing Long Workflows

For workflows with many steps:

1. Break into sub-workflows - Modular workflows are easier to debug
2. Add checkpoints - Log completion of major sections
3. Use unique identifiers - Pass a tracking ID through all steps
4. Set realistic timeouts - Long-running tasks need longer timeouts

Platform-Specific Tips

Zapier

• Use "Filter" steps to prevent unnecessary runs
• "Delay" actions prevent rate limiting
• "Paths" enable complex branching logic
• Check "Task History" for detailed debugging

Make (formerly Integromat)

• Use "Error Handlers" on critical modules
• "Routers" enable parallel processing
• "Iterators" process arrays item by item
• Built-in "Break" and "Resume" for error recovery