Designing Proactive Web Messaging Journeys based on Cart Abandonment Signals
What This Guide Covers
This masterclass details the implementation of a Proactive Web Messaging strategy designed to recover lost revenue from cart abandonment. By the end of this guide, you will be able to integrate the Genesys Cloud Journey SDK with your e-commerce platform to detect high-value abandonment signals (e.g., dwelling on the checkout page for >60s with >$200 in the cart) and trigger a personalized messaging invitation to a specialized “Conversion Agent.”
Prerequisites, Roles & Licensing
Proactive messaging requires a combination of web-side tracking and Genesys Cloud predictive engagement.
- Licensing: Genesys Cloud CX 2 or 3 with Predictive Engagement enabled.
- Permissions:
Journey > Action Map > View/EditJourney > Segment > View/EditArchitect > Flow > View/Edit
- OAuth Scopes:
journey,messaging. - Infrastructure: Access to your e-commerce website’s
<head>tag for snippet injection.
The Implementation Deep-Dive
1. Deploying the Journey SDK and Custom Events
The first step is to ensure Genesys Cloud can “see” what is happening in the shopping cart. Standard page tracking isn’t enough; you need Custom Web Events.
Implementation Pattern:
When a user adds an item to their cart, your website should emit a custom event to the Journey SDK:
// Example: Tracking a Cart Update
genesys('track', 'cart_updated', {
cart_value: 249.99,
item_count: 3,
cart_id: 'abc-123'
});
2. Defining the “Abandonment” Segment
In the Genesys Cloud Admin UI, create a Segment called “High-Value Abandoner.”
Segment Logic:
- Condition 1: Current Page URL contains
/checkout. - Condition 2: Custom Attribute
cart_valueis greater than200. - Condition 3: Idle time on current page is greater than
60 seconds.
Architectural Reasoning:
Do not trigger on every abandonment. Triggering too early or on low-value carts (e.g., $5) can overwhelm your agents and decrease the ROI of the proactive strategy. Focus your high-cost human resources on the “Silver” and “Gold” abandonment opportunities.
3. Orchestrating the Action Map and Architect Flow
Once a user falls into the “High-Value Abandoner” segment, an Action Map triggers a Web Messaging Invitation.
Architect Implementation Pattern:
- The Invitation: The user sees a subtle toast: “Need help completing your order? Chat with a specialist now.”
- The Flow: When the user clicks “Chat,” they are routed to a specialized Inbound Message Flow.
- Data Injection: Use the
Get Journey Dataaction in Architect to fetch thecart_valueandcart_id. - Intelligent Routing: Route the interaction to a queue of “Sales Closers” with the
Salesskill and a high proficiency.
4. Handling the “Conversion” Participant Data
To track the success of the recovery, you must tag the interaction.
The Trap:
Losing visibility into whether the chat actually led to a sale.
The Solution: Use Participant Data to store the cart_id and initial_value. After the chat, your backend system should cross-reference this cart_id with its completed orders. If the order is completed within 24 hours of the chat, attribute the revenue to the Genesys Cloud interaction.
Validation, Edge Cases & Troubleshooting
Edge Case 1: The “Spammy” Invitation
- The failure condition: A customer is browsing multiple items and gets 5 chat invitations in 10 minutes, causing them to leave the site in frustration.
- The root cause: Lack of Frequency Capping in the Action Map.
- The solution: Configure the Action Map to “Do not show again for X hours” if a customer has already seen or declined an invitation. Respect the customer’s space.
Edge Case 2: Out-of-Hours Abandonment
- The failure condition: A customer abandons their cart at 2:00 AM, clicks the chat invitation, and gets a “No agents available” message.
- The root cause: The Action Map is active 24/7 but the sales queue is not.
- The solution: Add a Schedule Check to the Action Map or the underlying Architect flow. If it’s after hours, trigger an Asynchronous Messaging experience instead: “We’re away right now, but leave your email and we’ll send you a $10 discount code to complete your order.”