Core Concepts

Learn the fundamental concepts that power Flint's workflow automation system.

What is a Workflow?

A workflow is an automated sequence of actions that mimics your business process. Think of it as a digital worker that follows the same steps you would, but faster and more consistently.

Key Components

Every workflow has three essential parts:

Trigger - What starts the workflow

A new email arrives
A file is uploaded
A schedule runs (daily at 9 AM)
Someone clicks a button

Steps - What happens next

Send an email
Update a database
Call an external API
Transform data

Connections - How steps link together

Do step A, then step B
If condition X is true, do step Y
Run steps A, B, and C at the same time

Building Blocks

Actions

Actions perform work - they do something:

Send Email - Notify customers, partners, or team members
API Call - Connect to external systems like Salesforce or Slack
Data Transform - Convert data from one format to another
Database Query - Retrieve or update information
Generate Document - Create PDFs, reports, or contracts
Delay - Wait for a specified amount of time

Conditions

Conditions make decisions - they decide what happens next:

If/Then - Simple yes/no decisions
- If invoice amount > $1000, then send to manager for approval
Switch - Choose from multiple options
- Route customer inquiry based on their region
Filter - Process only items that meet criteria
- Only process orders from premium customers

Triggers

Triggers start workflows - they initiate the process:

Webhook - External system sends data to your workflow
Schedule - Run daily, weekly, monthly, or custom timing
File Upload - Process files as they arrive
Manual - Start workflow with a button click
Event - System events like "new user created"

Variables: Connecting Your Data

Variables let you pass data between steps using simple {{variable}} syntax.

Common Variable Types

From Triggers

{{webhook.customer_email}} - Email from incoming webhook
{{file.name}} - Name of uploaded file
{{schedule.current_time}} - When the schedule ran

From Previous Steps

{{step1.api_response}} - Data returned from API call
{{email_step.success}} - Whether email was sent successfully
{{calculation.total}} - Result from a calculation

Global Settings

{{variables.api_key}} - Your organization's API keys
{{variables.sender_email}} - Default email address
{{variables.company_name}} - Your company information

Variable Examples

Email Subject: "Welcome {{customer.first_name}}!"
API URL: "https://api.example.com/customers/{{customer.id}}"
Condition: "{{order.total}} > 500"

Flow Control

Sequential Flow

Most workflows run steps one after another:

Receive Order → Validate Data → Update Inventory → Send Confirmation

Conditional Flow

Make decisions based on data:

Receive Invoice → Check Amount
                     ├─ If > $1000 → Manager Approval
                     └─ If ≤ $1000 → Auto Approve

Parallel Flow

Run multiple steps at the same time for efficiency:

New Customer
    ├─ Create Account
    ├─ Send Welcome Email
    └─ Add to Mailing List
            ↓
    All Complete → Next Step

Error Handling

Real-world workflows need to handle problems gracefully.

Automatic Retries

If something fails (like a network error), workflows automatically retry with smart delays:

1st retry: Wait 1 second
2nd retry: Wait 4 seconds
3rd retry: Wait 16 seconds

Manual Intervention

For issues that need human review:

Workflow pauses automatically
You get notified via email/Slack
Review the issue in the dashboard
Choose to continue, retry, or abort

Fallback Actions

Create alternative paths for common failures:

If primary email service fails → Use backup service
If customer data is incomplete → Route to data team
If payment fails → Send payment reminder

Organization Features

Custom Actions

Your organization can create specialized actions for your unique systems:

Connect to proprietary databases
Integrate with custom APIs
Implement your specific business rules

Access Control

Control who can do what with workflows:

Viewers - Can see workflow status and history
Editors - Can modify workflows
Administrators - Can create new workflows and manage settings

Resource Management

Keep workflows running efficiently:

Execution Limits - Prevent runaway workflows
Priority Queues - Important workflows run first
Regional Deployment - Run workflows close to your users

Audit Trail

Every workflow execution is completely tracked:

What Gets Recorded

When each step started and finished
What data was passed between steps
Any errors that occurred
Who triggered the workflow
How long each step took

Why It Matters

Debugging - Quickly find and fix issues
Compliance - Prove processes were followed correctly
Optimization - Identify slow steps for improvement
Accountability - See who did what and when

Best Practices

Workflow Design

Start Simple - Build basic version first, then add complexity
Test Early - Use sample data to verify logic
Handle Errors - Plan for what happens when things go wrong
Document Steps - Add descriptions so others understand your workflow

Performance Tips

Use Parallel Steps - Run independent actions simultaneously
Batch Operations - Process multiple items together when possible
Set Timeouts - Don't let steps run forever
Cache Results - Store frequently used data to avoid repeated lookups

Security Considerations

Secure Variables - Use encrypted storage for sensitive data
Limit Access - Only give permissions to those who need them
Regular Reviews - Periodically check who has access to what
Audit Regularly - Monitor workflow activity for unusual patterns

Getting Help

Built-in Guidance

Step Descriptions - Each action explains what it does
Parameter Help - Tooltips explain each setting
Example Values - See sample data for variables
Error Messages - Clear explanations when things go wrong

Learning Resources

Template Library - Start with pre-built workflows
Video Tutorials - Step-by-step guidance
Community Forum - Learn from other users
Support Team - Expert help when you need it

Core Concepts

On this page