Flow editor | Yaplet

Where the builder is

Open a workflow from Brand → (your brand) → Workflows and you land on its Editor tab. The builder kept the web address it always had, so existing bookmarks and deep links still work — only the list you used to reach it from has moved. The back arrow at the top left returns you to that brand's Workflows page, and Save and Publish sit at the top right. The permission is Workflows.

Understanding Action Nodes

Action nodes are the building blocks of a workflow. Each node type performs a specific function and can be connected to create conversational pathways. Actions are categorized into two types:

Regular Actions are processed sequentially within a node - each action completes (either instantly or waiting for user input) before moving to the next action in the same node. You can include multiple regular actions in a single node.

Final Actions can only be selected one per node and will complete the node once executed. These actions typically branch the conversation or end the automated flow.

Regular Actions

These actions are processed one after another within a single node, allowing you to build sequential automation steps.

Message Node

Send text messages to guide users through your conversation flow. Perfect for providing information, asking questions, or delivering instructions.

Collect Data Node

Gather information from users with flexible input types. Collect emails, custom data, or survey responses to build customer profiles and improve targeting.

Ask Question Node

Ask open-ended questions to visitors and collect their natural responses. Perfect for gathering feedback, conducting surveys, or having free-form conversations.

Form Node

Launch interactive forms to collect structured data. Create comprehensive data collection experiences with validation, conditional logic, and custom styling.

API Action Node

Connect your chatbot to external services. Make HTTP requests, trigger webhooks, or integrate with third-party systems with automatic error handling.

Request Agent Node

Transfer conversations to human agents when automation reaches its limits. Ensure seamless handoff for complex situations.

Reply Time Node

Show visitors expected response times to set proper expectations for when they'll receive a reply.

Final Actions

These actions complete the node when executed and can only be used one per node. They typically create branching paths or end the automated conversation.

Buttons Node

Create interactive choices with clickable buttons. Each button can lead to different conversation paths, making it ideal for menus, surveys, or decision trees.

Conditions Node

Add intelligent routing based on visitor data or previous responses. Use conditions to personalize experiences and create dynamic conversation flows.

AI Bot Node

Hand a question to the brand's Vex AI agent, let it answer from the brand's knowledge, then carry on with the flow.

Custom Bot Node

Reuse an existing workflow as a component. Build modular automation by nesting one workflow inside another.

Link Node

Share clickable links with users. Direct them to resources, documentation, or external websites with customizable display options.

Deep Dive: Action Node Capabilities

Regular Actions

Message Node Details

The Message node is your basic communication tool. It sends text content to users at specific points in your conversation flow.

Common scenarios:

Welcome messages when workflows start
Instructions before data collection
Confirmation messages after actions complete
Help text and guidance throughout the process

Configuration options:

Custom message text with full formatting support
Outbound toggle to control message presentation
Support for both public and private messages

Pro tip: Use clear, concise language in your messages. Consider your audience and keep instructions simple and actionable.

Collect Data Node: Information Gathering

Transform your chatbot into a data collection tool with flexible input options and validation.

Data collection types:

Automatically collect and validate email addresses. Skips collection if the visitor already has an email on file.

Gather any type of information with customizable input types.

Available input types:

Text fields for names, comments, or open-ended responses
Number inputs for quantities, ratings, or measurements
Dropdown menus for predefined options and selections ::: ::

Configuration details:

Custom question prompts for each data point
Attribute names for data storage and reference
Input validation and formatting options
Integration with visitor profiles and CRM systems

dataType

select required

Choose between collecting customer email addresses or custom data fields

inputType

select required

For custom data, select text, number, or dropdown input types

question

string required

The prompt or question shown to users when collecting this data

attributeName

string

Name used to store and reference this data in other workflow actions

Ask Question Node: Open-Ended Conversations

Enable natural, free-form conversations by asking open-ended questions that visitors can answer in their own words. Unlike structured data collection, this node allows for organic dialogue and qualitative feedback.

Perfect for:

Gathering detailed feedback and reviews
Conducting open-ended surveys and research
Allowing users to express complex needs or issues
Building rapport through conversational interactions
Collecting qualitative data for analysis

Key features:

Custom question prompts that appear in the chat
Natural language responses from visitors
Flexible conversation flow without predefined options
Response storage for later analysis or follow-up
Seamless integration with subsequent workflow actions

How it works:

Configure your question text in the node settings
The question appears naturally in the chat conversation
Visitors respond in their own words without restrictions
Their response is captured and stored for workflow processing
Conversation continues based on your connected workflow paths

question

string required

The question text that will be displayed to visitors in the chat

Best Practice: Use Ask Question nodes when you want genuine, unscripted responses from users. Perfect for understanding customer sentiment, gathering detailed feedback, or exploring complex topics that don't fit predefined choices.

Response Handling: Visitor responses are stored and can be referenced in subsequent workflow actions, conditions, or passed to human agents for context.

Form Node: Interactive Data Collection

Launch comprehensive forms within your chatbot workflows to collect structured information from users. Forms provide a more sophisticated data collection experience compared to simple text inputs.

When to use Form nodes:

Multi-step data collection processes
Complex surveys requiring validation
Lead generation with multiple fields
Customer onboarding and registration flows
Feedback collection with conditional questions

Key features:

Full form builder with drag-and-drop field creation
Multiple input types (text, email, phone, dropdown, checkboxes, etc.)
Conditional logic to show/hide fields based on responses
Built-in validation and error handling
Custom styling and branding options
Progress indicators for multi-step forms
Data storage and CRM integration

How it works:

Pick a form from the list, shown as "Board name - Form name"
The form launches seamlessly within the chat interface
Users complete the form with guided validation
Form responses are automatically stored and can trigger subsequent workflow actions
Conversation continues after form completion

Where forms come from: intake forms are built on the board they feed, at Tickets → (board) → Forms, and the picker here lists the forms of every board you can see. There is no separate organisation-wide Forms page any more. See the Forms documentation for building one.

A form step blocks three channels. A workflow containing one cannot run on Facebook, on Instagram or on a phone call, because none of them can display a form. The Channels column on the brand's Workflows list shows this as greyed-out icons.

Best Practice: Use Form nodes for complex data collection scenarios where you need validation, conditional logic, or multiple related fields. For simple single-field collection, consider using the Collect Data node instead.

API Action Node: External Integrations

Connect your chatbot to the outside world with HTTP requests and web service integrations.

Integration possibilities:

CRM system updates and customer record creation
E-commerce platform interactions and order processing
Database queries and information retrieval
Webhook triggers and external workflow activation

Built-in reliability:

Automatic error detection and handling
Seamless fallback to human agents when APIs fail
Request/response logging and monitoring
Authentication header support for secure connections

API Configuration: Always test your API endpoints before deploying workflows. The system will automatically transfer to an agent if the API request fails, ensuring conversations never get stuck.

Request Agent Node: Human Handoff

Ensure your automation has a safety net by providing seamless transfer to human agents when needed.

When to use agent transfer:

Complex technical issues requiring human expertise
Emotional situations needing empathy and personal touch
High-value customer interactions
Fallback when automation cannot resolve queries

Transfer process:

Automatic agent assignment based on availability
Context preservation from the automated conversation
Optional custom transfer messages
Seamless transition without conversation interruption

Configuration options:

Pre-transfer message: Customize the message sent by the AI before requesting an agent. Leave empty to use the default message.

Best Practice: Use a custom pre-transfer message to set expectations or provide context to the visitor before the handoff occurs, creating a smoother transition experience.

Reply Time Node: Expected Response Times

Set clear expectations by showing visitors how long they can expect to wait for a response. This node helps manage visitor expectations during peak times or when agents are busy.

When to use Reply Time nodes:

During business hours when agents are available
Before complex tasks that require agent review
When transitioning from automated responses to human support
To manage expectations during high-volume periods

Display options:

Automatic calculation based on current agent availability
Custom time estimates for specific scenarios
Integration with business hours and timezone settings
Clear messaging about when responses can be expected

Best Practice: Use Reply Time nodes before Request Agent actions to set proper expectations. This reduces visitor frustration and improves satisfaction with your support process.

Final Actions

Buttons Node: Interactive Choices

Buttons create decision points in your conversation. Each button represents a choice that leads to a different workflow branch.

Perfect for:

Product or service selection
Yes/no confirmation dialogs
Multi-option surveys and polls
Navigation menus and help options

How it works:

User sees your message with clickable buttons
Clicking a button advances them to the connected workflow branch
Each button can lead to completely different conversation paths

Step 1: Configure Button Labels

Enter clear, descriptive text for each button option

Step 2: Connect Output Paths

Drag connections from each button to different nodes in your workflow

Step 3: Handle User Selection

The workflow automatically routes based on the user's choice

Step 4: Continue Conversation

Each path can have its own sequence of actions and responses

Conditions Node: Smart Routing Logic

Conditions add intelligence to your workflows by evaluating visitor data and making decisions automatically.

What you can check:

Visitor information (email, country, session history)
Widget status (online/offline agents available)
Previous user responses and collected data
Time-based conditions and patterns

Advanced features:

Visual query builder for complex logic
Support for AND/OR combinations
Multiple conditional branches (if/else if/else)
Fallback paths for unmatched conditions

::accordion

defaultValue: '1'

Check visitor attributes like email address, country of origin, session count, or first visit date. Perfect for personalized experiences and targeted messaging.

Evaluate whether agents are online, widget availability, or system status. Useful for routing to live agents during business hours.

Combine multiple conditions with AND/OR operators to create sophisticated decision trees for complex automation scenarios.

Create schedules and time-sensitive workflows that behave differently based on dates, times, or user behavior patterns.

AI Bot Node: Intelligent Responses

Step out of the script for one turn and let the brand's Vex AI agent answer in its own words.

What it brings:

Natural language understanding, so the visitor can phrase things however they like
An answer drawn from everything the brand knows — knowledge bases, documentation, files, website pages and Q&A
A reply written in the visitor's own language
A way to cover the questions a scripted branch could never anticipate

Setup process:

Write the Question to ask the AI Model — a full sentence, question or instruction
Connect the node's output to continue the workflow afterwards
That is the whole configuration

There is no chatbot picker on this node. The question goes to the brand that owns this workflow, and is answered by that brand's Vex agent from that brand's knowledge. Nothing needs connecting, and there is nothing to choose wrongly.

Custom Bot Node: Workflow Composition

Build complex automation by combining several workflows into modular components.

Modular design benefits:

One workflow reused in several places instead of rebuilt each time
Simplified maintenance and updates
Consistent behavior across multiple touchpoints
Easier testing and debugging of complex flows

How it works:

Pick a workflow from the list, which holds every active workflow in your organisation — not only this brand's
Execute it as a step within your current workflow
Pass data between workflows seamlessly
Return control to the main workflow when complete

A nested workflow passes its channel restrictions up. If the workflow you call contains a form, the workflow calling it is also blocked from Facebook, Instagram and phone calls, even though its own steps are clean.

Share external resources and direct users to additional information or actions.

Link usage scenarios:

Product documentation and user guides
Support article references and knowledge bases
Download links for files and resources
External booking systems or appointment schedulers

Display options:

Open links in new browser tabs
Maintain conversation context while users browse
Seamless return to chatbot after link interaction

Which steps block a channel

A workflow always runs in its brand's chat widget. Whether it can also run on Facebook, on Instagram and on phone calls depends entirely on the steps it contains, and Yaplet works that out again every time you save.

Step used	Facebook and Instagram	Phone calls
Form	Blocked	Blocked
Link, Buttons, Reply time, Request agent, Collect data, Ask question	Fine	Blocked
Message, Conditions, API action, AI bot	Fine	Fine

The result shows up as greyed-out channel icons on the brand's Workflows list, with the reason on hover. Build a phone workflow from messages, conditions, API actions and AI bot steps alone and it will run on a call.

Workflow Architecture Principles

Starting Your Workflow

Every workflow begins with a Start node - this is your entry point, set off by the trigger you chose on the Settings tab.

A manually started workflow now runs even when a human agent is already on the conversation. Until recently every workflow, manual ones included, was blocked the moment an agent was assigned, so manual workflows quietly did nothing at all. Automatic triggers are still held back while a person is handling the chat.

Start Node

This node is automatically created and serves as the foundation for your entire workflow. You cannot remove or modify the start node directly.

Building Conversation Flows

Connect your action nodes by dragging from the output handles (right side of nodes) to input handles (left side of nodes). The system validates connections to ensure logical flow.

Connection rules:

Output handles create branching paths for decisions
Input handles accept connections from previous actions
Some nodes (like buttons and conditions) create multiple output paths
Workflows can branch extensively and merge back together

Completing Conversations

Workflows naturally end when they reach nodes without output connections. These terminal nodes represent conversation completion points.

End state considerations:

Clear user guidance about next steps
Data collection completion confirmation
Transfer to human agents when appropriate
Seamless conversation closure

Advanced Workflow Strategies

Error Handling and Fallbacks

Design robust workflows that handle edge cases gracefully:

Always plan for API failures: API Action nodes automatically transfer to agents when external services fail, but consider additional fallback messaging.

Validate data collection: Use conditions to check if required information was successfully gathered before proceeding.

Handle unexpected user responses: Include catch-all conditions and agent transfer options for scenarios your automation doesn't cover.

Performance Optimization

Keep your workflows efficient and user-friendly:

Minimize decision complexity: Too many conditional branches can confuse both users and workflow maintenance.

Reuse common components: Use Custom Bot nodes for frequently repeated processes rather than rebuilding them.

Test all paths: Thoroughly test every possible conversation path to ensure smooth user experiences.

Monitoring and Analytics

Track your workflow performance:

Monitor conversion rates at decision points
Analyze common agent transfer reasons
Track data collection completion rates
Identify popular conversation paths for optimization

Getting Started with Workflows

Ready to build your first workflow? Start simple and expand gradually:

Step 1: Define Your Goal

Decide what you want your workflow to accomplish - collect leads, provide support, or guide users through a process.

Step 2: Map the Conversation Flow

Sketch out the logical steps and decision points on paper before building in the editor.

Step 3: Start with Core Actions

Begin with Message and Button nodes to establish the basic conversation structure.

Step 4: Add Intelligence Gradually

Introduce Conditions and Data Collection nodes as your workflow becomes more sophisticated.

Step 5: Test and Iterate

Use the workflow preview to test different paths and refine based on real user interactions.

The flow editor gives you powerful tools to build personalised, intelligent conversations. Start with the basics and add more advanced steps as you get comfortable with the system. Remember - great automation is built on understanding your visitors and designing conversations that feel natural and helpful.