Genesys Predictive Engagement (SL09) for Genesys Engage cloud

From Genesys Documentation
Jump to: navigation, search
This topic is part of the manual Genesys Engage cloud Use Cases for version Current of Genesys Use Cases.
Read this topic for other versions:
This use case is the subject of an Early Adopter Program (EAP). Please contact Lindsay Frazier, Product Management for more information. Customer Service applications of this use case is addressed by Genesys Predictive Chatbots (CE37).
Use machine learning powered journey analytics to monitor website activity, predict visitor outcomes, and proactively engage with prospects and customers

Use Case Overview

Story and Business Context

One of the biggest challenges for the modern business is learning to utilize all of the data available to them in a way that is both meaningful and actionable. The data generated by a website is often left unexplored, and as a result, the intentions and reactions of individual customers and prospects can be overlooked. Focus is often placed on the broad strokes–key metrics such as the number of page views this month–and we lose the ability to identify the potential customers who need engagement most. As a result, prospects who may be on the verge of signing up for a trial, completing a checkout, or any other desirable outcome, fall through the cracks.

The high volume of website traffic makes it challenging to identify the right individuals, best moments, and optimal ways to engage in real time. Expectations for time-to-respond are increasing but growing your inside sales staff is costly. Marketing dollars are spent on a wide range of demand generation activities but it is difficult to connect the dots from lead to contact to opportunity to closed. Genesys Predictive Engagement uses machine learning to track the progress of website visitors towards defined outcomes–purchase completion, requesting a quote–and enables the business to define rules to trigger intervention only at the points when it is needed most.

With Genesys Predictive Engagement, you can predict and prioritize high-value leads for your sales team to engage and proactively offer chat to better utilize your staff and reduce your costs. Genesys Predictive Engagement uses machine learning to track the progress of website visitors towards defined outcomes–purchase completion, requesting a quote–and enables the business to define rules to trigger intervention only at the points when it is needed most.

Use Case Benefits*

Use Case Benefits Explanation
Improved cross-sell and up-sell (Increase Customer Lifetime Value) Retain customers by increasing customer satisfaction with faster and more personalized service. Improve the ability to up-sell and cross-sell existing customers with data based on their current interests, online journeys and prior purchasing behavior.
Improved Employee Productivity Sales reps are empowered with real time customer journey data from your website. This visibility allows them to personalize and prioritize engagements with prospective customers. Productivity is improved when sales reps interact when they have the most impact. Our software predicts which prospects are most likely to buy or abandon based on outcomes from previous customers taken guesswork out of the equations for your sales teams.
Improved Net Promoter Score Improve CX and NPS scores and reduce Customer Effort by providing customers with a more timely and meaningful engagement online.
Increased Quality of Lead Conversion Follow individual customer journeys in real time on your website. Identify the moment of struggle or moment of opportunity and launch a chat or voice interaction with a sales agent at the right time to increase lead volume, improve lead qualification and reduce customer churn.
Increased Sales Conversions Accelerate sales cycles and lead conversion rates (MQL to SQL to conversion) by engaging prospects or online shoppers in real time—at the right time—as they browse your website.
Reduced Sales and Marketing Costs Reduce customer acquisition cost (CAC). Predict which prospects are most likely to buy based on successful outcomes from previous customers. Use these insights to focus marketing efforts and ad spend on a target profiles or sales sales activities.
*You can sort all use cases according to their stated benefits here: Sort by benefits


Genesys Predictive Engagement monitors individual customer journeys on your company website and applies machine learning, dynamic segmentation, and real-time outcome scoring to identify the right moments for proactive engagement with the right customer via chat or content offer. When the visitor interacts, the sales rep has the customer journey information at their fingertips.

Use Case Definition

Business Flow

Main Flow

The following diagram shows the main flow of the use case, from the point of view of the system.

Business Flow Description

  1. A visitor starts browsing the company website.
  2. Genesys determines whether visitor is new or returning to website, and associates data from previous journeys.
  3. Genesys continues to monitor website behavior and update segment attribution and outcome score as appropriate.
  4. The combination of segment and variations in outcome score as predicted by the machine learning engine can eventually trigger a chat to the visitor.
  5. An algorithm determines the availability of sales reps to handle the interactions.
  6. If sales reps are identified as available to handle chat, a proactive invitation to chat is presented to the visitor
  7. If the visitor accepts the invite, a popup registration window enables the visitor to enter their data.

Business Flow


This diagram details the routing that takes place before and during the chat.

Business Flow Description

  1. Genesys routes the interaction to a sales rep based on the target suggested by Genesys Predictive Engagement (target expression and virtual queue).
  2. Sales rep and prospect are in conversation. The sales rep has access to full visitor context such as segment, journey information, and outcome score.
  3. After the conversation ends, the sales rep agent sets a disposition code within their desktop to record the outcome of the conversation.

Business and Distribution Logic

Business Logic

BL1 – Customer Identification

The system can use cookies to detect returning visitors and associate them with previous site visits. Identity information provided during the journey (such as email address or phone number) is captured when explicitly submitted from the web page and can identify the visitor even across devices. If a visitor uses a second device to visit the website the next day and provides a piece of this information, their visit can be associated to the previous journeys across devices. When visitor identity cannot be determined, the customer is handled as an anonymous user and all tracked data attached to them. Once the visitor is identified, all tracking data collected and to be collected is associated to that specific visitor.

BL2 – Segment and Outcome Configuration

Segments are a way to categorize visitors on the website into segments, based on common behavior and attributes. Segments are configured upfront during system provisioning. A segment is made up of two components:

  • Attributes, such as browser type, device type, location, marketing campaign they are associated with, UTM parameters, and the referral website.
  • Journey pattern, such as web browsing behavior, searches performed on the website, items clicked, returning users, cart abandoner, and high order value.

Outcomes or goals are specific tasks you want your visitors to perform on your website. As with segments, these are configured upfront. Typical outcomes include:

  • Proceed to checkout with a shopping cart
  • Submit payment
  • Download a whitepaper
  • Book a demo or appointment

Genesys uses predictive analytics to evaluate in real time the probability for a specific outcome to be achieved, based on segment and visitor behavior on the website (the outcome score).

BL3 – Action Map Configuration

Action Maps determine the way to engage with the website visitor. Within action maps, you define the triggers that will result in an action to the customer. These triggers include:

  • Segment
  • User activity
  • Outcome score - Typically, a drop in outcome score for a specific segment can trigger an action.

The following actions are part of this use case:

  • Invite to Chat

BL4 – Sales Rep Availability

Genesys Predictive Engagement can optionally evaluate sales rep availability, based on contact center statistics like estimated wait time and engage only when sales reps are ready to serve the interaction. If no sales rep is available for the requested queue, then no invite is sent.

BL5 – Customer Invite and Registration

Genesys Widgets is used for:

  • Invite messages for chat
  • Collection of visitor's contact details
  • Engagement over chat session booking

Distribution Logic

The distribution of the interaction is determined by the target expression and virtual queue configured in the Genesys Predictive Engagement rules.

Use Case Requirements

Customer Interface Requirements

  • Based on Genesys Widgets 9 with standard capabilities to adapt to customer corporate identity

Agent Desktop Requirements



Real-time Reporting

Interaction-related reporting is based on standard Pulse templates. Capabilities are Routing (CE18).

Historical Reporting

Interaction-related reporting is based on Genesys Interactive Insights (GI2). Capabilities are similar to Routing (CE18).

There are two analytical dashboards are provided:

  1. Visitor Activity on the website: Provides the count of visits filtered by time range, segments matched, and outcomes achieved.
  2. Action Map Performance: Provides the count of actions that were offered, accepted, and rejected filtered by time range.


General Assumptions

Genesys Widgets must be used. Customer must deploy both Predictive Engagement and Widgets code snippets on their website / web pages.

General logic for routing of interactions uses part of these cases. If CE18 is already deployed and customized, SL09 design and configuration must take it into account.

Customer Assumptions


All required, alternate, and optional use cases are listed here, as well as any exceptions.

All of the following required: At least one of the following required: Optional Exceptions


      None None None

      Premise Assumptions

      • N/A

      Cloud Assumptions

      Related Documentation

      Document Version

      • v 1.0.1