Week of August 14th

Features

Routing AI agents - Now running on GPT-4o for better and faster performance

We’re still excited about the recent release of LivePerson’s Routing AI agent, but there’s even more good news to announce on this front.

In our commitment to providing you with the best possible solutions, we’ve just upgraded the LLM that supports the Routing AI agent from GPT-4.0 to cutting-edge GPT-4o. GPT-4o was recently made available, and our testing has revealed that it performs better and faster. What’s more, it’s cheaper. So, we’ve partially made the switch to GPT-4o as of Release 2024_4.0.

Accounts in North America: As of September 6, 2024, when you add a Guided Routing interaction to a bot, the default prompt in the interaction uses GPT-4o. In this case, there’s no action required by you.

But, if you have a bot that has a Guided Routing interaction that you added before September 6, 2024—and you want to make the switch to GPT-4o— manually delete the Guided Routing interaction and re-add it to the bot. This is a self-service approach.

Accounts in EMEA: As of November 9, 2024, the above also applies.

Accounts in APAC: GPT-4o isn’t yet available for accounts in this region, so your prompts can only use GPT-4.0.

Features

Signaled events

The Signaled Events capability facilitates smooth metadata transfer between a brand’s web page, unified windows, and third-party bots. By enabling the forwarding of SIGNALED events from the customer’s web page to third-party bots, this feature empowers real-time interactions. These bots can exchange signals with the brand’s page, creating a dynamic and responsive user experience. Brands can listen for these events and utilize the feature to allow third-party bots to assist users directly on the page, enhancing engagement and support.

Functionality  

• A new signal event implemented by LivePerson backend team
• The unified window can consume and trigger this event to allow metadata transfer from and to the web page.
• This functionality is aimed to be leveraged by Bots. Conversation Builder bots leveraging universal tile and Third Party Bots can send and receive signals.

For more information on how bots can send signals, click here.

For information on how bots can receive signals , click here.

To enable this feature, please contact your LivePerson Account Team.

Features

The Data Transporter tool has been updated with a new feature. Predefined Content (PDC) data has been successfully integrated into the reporting system, enhancing the "Messaging Interactions - Messages" report and introducing a new "Predefined Content Messages Summary" report.

Key Enhancements:

1. Expanded "Messaging Interactions - Messages" Report: LivePerson have added the following new fields to provide deeper insights into messaging interactions:

• predefinedContent
• predefinedContentEdited
• predefinedContentLanguage
• predefinedContentCategoryId
• predefinedContentId

2. New "Predefined Content Messages Summary" Report: This new report provides a comprehensive summary of predefined content usage, with the following fields:

• Account_ID
• Date
• Agent_ID
• Agent_User_Name
• Agent_Group
• PDC_Category
• PDC_Title
• Modified_PDC_Messages
• Unmodified_PDC_Messages
• Non_PDC_Messages
• Total_Messages

These enhancements, now available in the Data Transporter tool, will provide valuable insights into the utilization and performance of predefined content, aiding better decision-making and strategy formulation.

Features

LivePerson have some updates and improvements to our ‘Agents’ widget. Alongside the existing filters for time, skills, and agent groups, we now have additional filters that enhance the functionality and usability of the widget. Here are the list of filters available:

Filters:

• STATE: Filter agents by their current status.
• STATE DURATION: Filter agents by the duration of their current state.
• ONLINE RATE: Filter agents by the percentage of time spent in the online state.
• AGENT NAME: Filter agents by their name.
• ASSIGNED: Filter agents by the number of open conversations.
• ACTIVE: Filter agents by the number of active conversations at full intensity.
• MAX.OPEN: Filter agents by the maximum number of high-intensity conversations.
• LOAD: Filter agents by the total weight of assigned conversations.
• CLOSED: Filter agents by the number of closed conversations within the selected timeframe.
• TRANSFER RATE: Filter agents by their transfer rate.
• CSAT: Filter agents by their CSAT score within the selected timeframe.

Please note: The state duration should not exceed 14 days from the current date. 

Features

The new WhatsApp channel setup aligns the design with other channel setup pages, such as Apple Messaging for Business or SMS. The highlight of the release is the ability to get an out-of-the-box skill routing configuration for the new number added.

Features

AMB support rich links by default

Rich links are now supported by default in the AMB channel, allowing users to preview content directly within the conversation. Instead of displaying a basic hyperlink, rich links are presented as rich content cards, offering a more engaging and visually appealing experience for users. This enhancement greatly improves the interaction compared to the previous hyperlink format.

Check out the AMB quick start guide for more information.

Features

Picture In Picture feature added for Voice & Video support

The Picture-in-Picture feature has been added for voice and video support, allowing users, whether customers or agents, to continue their messaging conversations while simultaneously engaging in a video or voice call. This enhancement ensures that users can maintain seamless communication without interruptions, multitasking between the call and the chat window for a more fluid and efficient interaction.

Key Functionality  

• Upon launch, the video call webview will be displayed as full screen.
• Consumers can tap on the video webview to minimize the video window and message at the same time.
• If the video is displayed in full screen, tapping on ‘back’ (‘<’) button will take the user back to the messaging window. The video window will be minimized.
• The button modal will show:
  • Switch camera
  • On/off camera
  • Hang out call
  • Back to chat

To enable, please contact your LivePerson account team. 

Please note:  Picture in Picture is supported on Android N (API 24) and above. Previous versions will default to full screen call view.

Features

Support for hand-over Facebook and Instagram conversations between Third Party Platforms and LivePerson agents i.e. Third Party Bots Amazon Lex 

Support for handing over Facebook and Instagram conversations between third-party platforms and LivePerson agents. Conversations initiated with third-party bots, such as those built on Amazon Lex, may be transferred to LivePerson agents using Meta handover protocol. This ensures that customer interactions are smoothly passed from automated systems to live agents whenever necessary, providing a more efficient and responsive customer experience across these platforms.

Features

Android InApp SDK - Android auto close separator line

A separation line will now appear for conversations that are automatically closed, making them easily distinguishable.

Enablement:

For enablement, please contact your LivePerson account team.

Features

InApp SDK - Device OS Dictation Capability

The primary function of speech recognition is to convert spoken words into written text. This is achieved through sophisticated algorithms that analyze audio input, identify spoken words, and transcribe them into text format.

Key Functionality  

• Allows consumers with the ability to dictate to the device and have the text written into the messaging typing area.
• Simplifies the experience for consumers by removing accessibility issues
• Once enabled, a custom speech recognition button will appear in the input-text view area.

Enablement

See these links for enablement instructions:

• iOS - link
• Android - link