Cargo query

A bar with green and red indicates that the visitor has not yet achieved the outcome. The amount of green indicates approximately how close the visitor is to achieving the goal. For more information on defining business outcomes, see Manage outcomes. For more information on how Genesys Predictive Engagement gathers and uses outcome data, see Outcomes overview.

For more information about Journey Outcomes, see the Journey Outcomes view in the Genesys Cloud CX Resource Center.

A column heading with an arrow indicates that you can click the arrow to sort the list by that entity. An upward-pointing arrow indicates that the list is in ascending order and a downward-pointing arrow indicates that the list is in descending order.

Now, the Outcome Score Service uses “sequence learning.” For each outcome configured, the Outcome Score Service manages a Machine Learning (ML) model that is trained exclusively on your organization’s customer journey data and tailored to your business objective. Each “sequence learning” model considers the full sequence of web events in the customer journey. These events include page views, form fills, searches, and your organization’s custom events. An event contains diverse attributes related to a customer (such as their geolocation and device type) and their website activity (such as search keywords). You can enrich events further with custom attributes that capture the unique properties of your data.

The Outcome Score Service transforms sequences of web events into an ML-friendly format and determines which features have the most predictive power. It uses customer journey data sampled from the previous week to train an ML model that identifies the patterns that are more predictive of outcome achievement. The Outcome Score Service monitors the performance of each model continuously, and retrains models regularly on the latest data. This process allows the system to react quickly to emerging trends and adapt to changes in the data.

By considering the full customer journey and managing a unique ML model tailored to each outcome, the Outcome Score Service can now identify more insightful patterns of behavior and provide a higher level of predictive power. Action maps backed by sequence learning are an even more powerful tool for offering the right engagement at the right moment.

We need to put this in plain English, but I want clarification before I begin...

Questions

Is this current-state, or tied to AI-23?

The best way to improve a model's predictions is to increase the number of achieved outcomes. In general, your dataset should contain several hundred positive examples for your model to be adequately trained to make reliable predictions.

Other factors can affect the accuracy of your model's performance, including:

• Total number of visitor journeys recorded
• Frequency that an outcome occurs in the data
• Richness of events that the visitor produces
  Tip

  You can exclude IP addresses to prevent events generated internally from influencing your models.

While the model is undergoing retraining, outcome scores are predicted using the previously trained version of the model. During the training process, agents can continue working as normal.

Each outcome score represents the likelihood that the visitor will achieve the particular business outcome, based on the actions the visitor has taken so far during the session or on other activities related to the visitor that are included in the appropriate events (for example, geolocation).

The model updates a visitor's score for each outcome in real-time, and a visitor's scores can change as they navigate your website.

To start the training:

1. Create an outcome.
2. Have visitors use your tracked website. To verify user activity, use Live Now.

Frequency capping stops web engagement offers when specific conditions are met during the same session. If your organization configured frequency capping at the global level, you can override those settings for this particular action map. When you override the settings, Genesys Predictive Engagement always offers the web engagement that this action map qualifies. All other action maps where the global setting isn't overridden are not affected.

• Some action maps use web actions. Web actions engage with a visitor directly. For example, you can create an action map that presents a web chat to visitors who go to a particular page for your current promotional campaign. The chat might ask the visitor whether they want to speak directly with someone in sales, who could then help them complete their purchase.
• Some action maps, such as Architect flows, use actions that react to visitor behavior, but do not interact with a visitor directly.

• Priority: The action map's priority determines which action map is offered in the event that multiple action maps are qualified for a visitor.
• URL: For web-based actions (web chats, web messaging, and content offers), you can apply an action map to specific webpages.
• Timing: When you select the action for an action map, you specify when the action works.
• Schedule: For action maps that offer web chats, you can select a schedule that indicates when the contact center is open to interact with visitors.
• Activation status: An action must be active for it to work.

When you create an action map, you define its trigger. The trigger determines when the action map is "qualified." When an action map is qualified, it means that Genesys Predictive Engagement could offer the action, depending on the other offer-related conditions.

An individual action maps can qualify only once for each session. If multiple action maps qualify during a given session, the number of actions that are offered depend upon the offer conditions, specifically the frequency capping settings.

The types of triggers that you can define are:

• Who the visitor is (what group or segment the visitor belongs to)
• What actions the visitor takes on your website (for example, searches for something or visits a specific page)

• If a visitor is progressing toward a business outcome, do not impede their progress by popping a chat window unnecessarily. Intervene only when a visitor's activity indicates that they are no longer on track to achieve the outcome that you want. To prevent this occurrence, use the drop in outcome scoring probability.
• Configure searching on your site to use the site search settings.
• Genesys Predictive Engagement tracks when a visitor does any of the following:
  • Completes fields in a form
  • Searches for something
  • Responds to a chat offer
  • Does not respond to a chat offer
  • Responds to a chat offer initially but then remains inactive until the offer times out
• Genesys Predictive Engagement can also track custom web events that you define. Then, you can configure the action map to trigger when one of these custom web events occurs. For example, you can implement a custom web event to track a visitor who adds an item to their shopping cart but then removes the item or abandons the shopping cart.
• The priority of an action map affects when it triggers. This priority is a relative setting, based on priorities of other action maps applied to the same webpage.

1. Prepare a Genesys Cloud CX integration and prepare the Genesys Cloud CX data actions.
2. In Genesys Cloud CX Architect, configure a workflow.
3. In Genesys Predictive Engagement, create an action map that uses an Architect flow action.
4. Monitor your Architect flow action with a report.

For specific integrations, see the following:

• Salesforce Lead Creation and Campaign Attribution integration

For more information about using Architect flow actions in a secure and compliant manner, see About security and compliance.

1. Import the necessary data actions for your integration:
   • Required data actions for the Salesforce Lead Creation and Campaign Attribution integration
2. Associate the data actions with your integration and configure them. The configuration options for the data action are specific to the data action itself. For more information about possible configuration options, see Data actions on the About Integrations page.
3. Save and publish your data actions.

• Required integration for Salesforce Lead Creation and Campaign Attribution integration

When you first install a Genesys Cloud CX integration, it has an active status. The integration must remain active for a Genesys Predictive Engagement Architect flow action to use it. To verify the status, check the setting of the Status toggle on the main Integrations page.

For each type of web-based offer, Genesys Predictive Engagement uses only one action map for a given segment, at a given time, on a given webpage.

• Example (multiple content offers or multiple web chats): If a visitor qualifies for multiple content offers or multiple web chats, Genesys Predictive Engagement chooses the action map with the higher priority.

• Example (one content offer and one web chat): If a visitor qualifies for both a content offer and a web chat, the visitor sees both offers.

In addition:

• If multiple action maps target the same segment at the same time, and they have the same priority, Genesys Predictive Engagement displays each action map a set percentage of the time. For example, if two action maps exist, Genesys Predictive Engagement displays each one 50% of the time. If three action maps exist, Genesys Predictive Engagement displays each one 33% of the time.
• If a visitor ignores an invitation to chat, but then goes to a web page where a higher priority action map is set to trigger, Genesys Predictive Engagement presents the visitor with the invitation to engage.

• Deleted or removed queues aren't available for selection.
• If a queue that you selected previously becomes unavailable, a warning appears next to the queue list box.
• If you do not select a queue and you configured a default or general queue, Genesys Predictive Engagement sends all engagements from this action map to the configured queue. Genesys Predictive Engagement doesn't make routing decisions; they are platform dependent.

• Action map conditions are met.
• URL conditions are verified.
• Throttling allows the chat. By default, throttling aims to have 80% of chats answered within 20 seconds.
  • For Genesys Engage on-premisess customers only, the Agent Pacing Service determines agent availability.
• Agents are on queue.
• Chat is within scheduled hours.

Genesys Predictive Engagement calculates the service level based on the queue's configuration in Genesys Cloud CX.

• If the lead exists, the workflow associates it with the campaign.
• If the lead does not exist, the workflow creates the lead and associates it with the campaign.

You can provide the campaign ID to the workflow directly. If you do not provide the campaign ID, the workflow attempts to extract the UTM campaign from the session ID to complete the association.

1. Download the Salesforce-Lead-Creation-And-Campaign-Attribution.i3WorkFlow
2. Associate the data actions.
3. Publish the workflow.
4. Switch to Architect to prepare the workflow.

1. Prepare your web campaign.
2. Create a dedicated Salesforce user.
3. Prepare the Salesforce data actions integration.
4. Prepare the Salesforce data actions.
5. Get the Salesforce Lead Creation and Campaign Attribution example workflow.
6. In Genesys Cloud CX Architect, prepare a workflow.
7. Prepare an action map to use the Architect flow action.
8. Monitor your Architect flow action with a report.
9. Genesys Cloud CX users: (Optional) Use scripts to help agents view and create a Salesforce lead:
   • Learn about the Salesforce Lead Creation and Campaign Attribution script
   • Download the Salesforce Lead Creation and Campaign Attribution script
   • Configure agent accounts
   • Communicate with agents about how to access and use the Salesforce Lead Creation and Campaign Attribution script

Prepare the Salesforce data actions integration

1. In Genesys Cloud CX, go to Admin > Integrations > Integrations.
2. Click + Integrations.
3. Install a Salesforce data actions integration.
4. On the Configuration tab, configure the credentials to use the dedicated Salesforce user account that you created in your Salesforce organization.

Prepare the Salesforce data actions

1. Download the required Salesforce data actions. For download links and descriptions, see the table that follows these instructions.
2. Import the data actions and associate them with the Salesforce data actions integration.
3. (Optional) test the setup of your contract.
4. Save and publish the data actions.

For more information about scripts, see:

• About scripting
• About widgets for web chat
• Use scripts

Create the action map

Override the default variable values

To create the new Salesforce lead with a richer set of journey data, override the default settings of these variables and tell Genesys Predictive Engagement where to obtain the user-specific values.

1. Next to the variable to override, under Action click Edit.
2. Select the Override the initial value for this action map check box.
3. From the New value list, select the value to send to the data action.
   • Static value: Unique identifier that Genesys Predictive Engagement creates for every user.
   • Session: Unique identifier that helps track user activity over time. For more information about using the session ID with Architect, see Get Journey Session action.
   • External Contact ID: Unique identifier that maps a user across the Genesys suite of products.
   • Other values:
     • Event attribute: If you trigger the action map based on a form_submitted event, the form fields appear here.
4. If a second list appears, select the source of the value.
5. Click Override.
6. Repeat these steps for all the variables to override.

• Integrations> Action > Execute permission
• Scripter > Published Script > View (to view agent-facing scripts)

Agents use the same dedicated Salesforce user account that Genesys Predictive Engagement uses to view and update leads in Salesforce.

1. Click here to download the Genesys Predictive Engagement-Salesforce-View-Create-Update-Lead-Script.script
2. Download the required data actions. For more information, see the following table.

Emergency schedule group: Set of schedules that you can activate in an emergency situation. When you activate the emergency group, all the action maps associated with the schedules in the group stop triggering. For more information about activating emergency groups, see Activate or deactivate an emergency group.

Source: Icon that represents the system in which the segment was created. For example, Genesys Cloud CX or a third-party system such as Adobe Experience Platform, Segment.io, Tealium AudienceStream, Emarsys, or Pega. If there are no segments from a third-party system, this column is hidden.

A column heading with an arrow indicates that you can click the arrow to sort the list by that entity. An upward-pointing arrow indicates that the list is in ascending order and a downward-pointing arrow indicates that the list is in descending order.

Name: Segment name.

Description: Segment description.

Type: Segment type (such as session or customer).

Segment Matches: Number of visitors matched to the segment.

Action Maps: Number of action maps configured to use the segment.

Outcomes Achieved: Number of outcomes achieved for the segment.

Active: Indicates whether the segment is ready to match visitors on your website.

Updated: Date the outcome was last modified.

Color: Background color to use when displaying the segment name (for example, in Live Now and the Action Map Report).

Options: Menu of options available for the segment, such as edit, delete, and duplicate.

For more information about Journey Segments, see the Journey Segments view in the Genesys Cloud CX Resource Center.

Updated: Date the segment was last modified in Predictive Engagement.

Description: Segment description.

Type: Segment type (such as session or customer).

Active: Indicates whether the segment is ready to match visitors on your website.

Color: Background color to use when displaying the segment name (for example, in Live Now and the Action Map Report).

Options: Menu of options available for the segment, such as edit, delete, and duplicate.

Segments appear in the order in which Genesys Predictive Engagement assigns them to the visitor.

For more information on defining segments, see Manage segments.

Removed custom sessions as part of AI-381 rollback.

1. Determine which snippet to use.
2. In Admin in Genesys Cloud CX, open the Global Settings page and then click the Tracking Snippet tab.
3. Copy the tracking snippet.
4. Paste the snippet. For more information about where to paste the snippet, see Deploy the snippet.
   

• uBlock Origin disables Genesys Predictive Engagement's tracking web beacons; this effectively prevents Genesys Predictive Engagement's tracking capabilities.
• uBlock Origin and uBlock Plus Adblocker block popular tag managers such as Adobe Tag Manager, Google Tag Manager, and more.

• About the data we track
• Predictive Engagement and GDPR
• Ad blockers
• About the tracking snippet
  • Types of snippets
  • Snippet recommendations
• Genesys Predictive Engagement and widgets
• Deployment overview

• If you use the traditional snippet, call ac('init') to initialize the Journey JavaScript SDK. For more information, see init.
• If you use the SPA snippet, use ac('load') to load the autotrackUrlChange module and initialize the Journey JavaScript SDK. For more information, see load and autotrackURLChange.

After you initialize the Journey JavaScript SDK, Predictive Engagement tracks pageview events when visitors arrive on your webpages. For more information, see Tracking methods.

• To customize how Genesys Predictive Engagement tracks activity on your website, see About event tracking with tag managers.
• To customize the Genesys Predictive Engagement tracking snippet with functions, see the Journey JavaScript SDK.

About deploying the snippet

• Deploy the widgets transport above the location where you deploy the Genesys Predictive Engagement tracking snippet.
• Deploy the Genesys Predictive Engagement tracking snippet on the main window of the website to track.
• Paste the tracking snippet in the head section of the website template page before the closing tag.
• Do not deploy the tracking snippet in an iframe.
• The Genesys Predictive Engagement tracking snippet loads JavaScript asynchronously without slowing down page loading. For more information about how the tracking snippet works, see About the tracking snippet.

Deployment instructions

• To deploy the tracking snippet manually, paste the copied snippet on your website.
• To deploy the tracking snippet using Google Tag Manager, see Deploy the tracking snippet with Google Tag Manager.
• To deploy the tracking snippet using Adobe Launch, see Use Adobe Launch with Genesys Predictive Engagement.
• To deploy the tracking snippet using another tag manager, see the tag manager vendor's instructions.

Web tracking

Use the Web tracking feature to specify the domains to track and configure web-related tracking settings.

Custom modules

Use a set of modules to enhance the functionality that the Journey JavaScript SDK provides.

Event tracking with tag managers

Use your tag manager to build custom tags and triggers that define how Predictive Engagement tracks activity on your website. For more information, see About event tracking with tag managers.

Domain and subdomain tracking with cookies

Use cookies to track domains and subdomains. For more information, see Cookies and Advanced tracking with cookies.

Website snippet

The Website snippet provides tracking support to traditional webpages that contain little or no dynamic content. Traditional websites load a new HTML page whenever the visitor navigates to a different URL. To provide tracking support to traditional webpages, the Website snippet calls ac('pageview') each time the web browser loads a new page.

SPA snippet

The SPA snippet can track activity on Single Page Applications (SPAs). SPAs load dynamic content on demand without requiring a hard reload of the HTML page. The SPA snippet can also track when visitors click relative links, which the Website snippet does not. The SPA snippet works on SPAs such as Angular, React, and Vue.js. For more information about the additional tracking options available with the SPA snippet, see autotrackURLChange.

Snippet recommendations

Troubleshoot the snippet

If the tracking snippet does not work properly when you load your web pages, check the following:

• Is the snippet on the correct page?
• Is the snippet in the correct place in the page header?
• Are the parameters defined correctly?
• Are there any typos, missing characters, or extra white space?
• Were any characters replaced with their printable equivalents?
• Is the domain on the list of allowed domains?

For more suggestions, see about deploying the snippet.

Error reference

Select the outcome for a visitor to achieve and then use the slider to specify the probability of the visitor achieving the outcome. Set likelihood to achieve outcome means that the likelihood of the visitor achieving the outcome is greater than or equal to the percentage that you set.

You can trigger action maps based on:

• Segment matches
• Visitor actions
• Business outcome probability

When a logged in user logs out and continues to use the website, Predictive Engagement considers the user as a new user and creates a new journey map for all actions of the unauthenticated user actions.

• In Genesys Predictive Engagement, is the action map that uses the Architect flow action active?
• In Genesys Cloud CX:
  • Is the Genesys Cloud CX integration assigned to the Architect flow action active?
  • Are the Genesys Cloud CX data actions published?
• In Architect, is the workflow saved and published?

1. Prepare an Architect flow, create an action map to use the Architect flow.
2. In the Set up an Action pane, click Configure Trigger Type.
3. Select the advanced orchestration action then click Architect flow.
4. Select the Architect flow to use with this action map. For more information about creating flows, see Prepare Architect components.
5. Override the default variables. Each data action that the Architect flow calls contains variables that are not set by default. These variables are placeholders that you can populate with visitor-specific data. For more information, see, Overrides for Salesforce Lead Creation and Campaign Attribution integration.
6. Specify when to activate the Architect flow action and the time period during which to use it.
   • Activation time
     • Immediately: Activate the Architect flow action as soon as a visitor arrives on your website.
     • On return to site: Activate the Architect flow action when a visitor returns to your website.
   • Time range: Time period to use the Architect flow action. To use the Architect flow action indefinitely, don't specify an end date.
7. Click Finish.

1. Prepare an inbound message flow or a bot flow and create an action map to use the inbound message flow.
2. In the Set up an Action pane, click Configure Trigger Type.
3. Select the advanced orchestration action and then click Inbound message flow.
4. Select the required inbound message flow to route the information.
5. Click Done.

1. Configure the report so that it displays the data that you want.
2. Download the report. The system downloads the report to your local Downloads folder.

About time values in exported reports

When you export a Visitor Activity report, the name of the exported file includes from- and to- timestamps in the standard ISO-8601 format (mmhhddmmyyyy).

If you import the CSV file into Excel, use the following formula to convert the ISO-8601 time formatted-cell to the Date type:

=DATEVALUE(MID(A1,1,10))+TIMEVALUE(MID(A1,12,8))"

(A1 represents the cell location of the first ISO-8601 time formatted cell in the Excel worksheet.)

1. Select "Device" and "Type."
2. Select the device type to filter:
   • If the device type is in the list, click it.
   • To search for the device type, type the first few letters of its name (not case-sensitive) and then click it.
   • To specify a device type not in the list, select "Other" and then type the device type (case-sensitive) in the box. For examples of device types, see Types of visitor attributes.
3. Filter the report.

You can filter by any combination of session segments and outcomes. You cannot filter customer segments.

1. Search for segments and outcomes by their names.
2. Select up to five session segments and five outcomes to filter.
3. To remove an item, click its X.