Updates: Conversation Assist enhancements including; Show or hide the Replies tab in the On-Demand Recommendations widget, Query Contextualization and more!
➡️ Exact delivery dates may vary, and brands may therefore not have immediate access to all features on the date of publication. Please contact your LivePerson account team for the exact dates on which you will have access to the features.
🚨The timing and scope of these features or functionalities remain at the sole discretion of LivePerson and are subject to change.
Conversation Assist
Show or hide the Replies tab in the On-Demand Recommendations widget
In the On-Demand Recommendations widget, the Replies tab displays predefined content to your agents. Predefined content is a subset of content that’s typically stored in your CMS or KMS. They’re ready-made responses that you’ve vetted and categorized to address common scenarios in conversations. Here are some examples:
- Greeting: Hello, how can I assist you?
- Conversational: Thank you for waiting.
- Research Required: Sure thing! Let me just grab a bit more info to give you the best answer. Hang tight for a couple of minutes—I’ll be right back!
Some of our brands want to use the Replies tab to offer predefined content, but others don’t. So, naturally, in the latter case, brands want the flexibility to hide the tab.
In this release, we add a new setting for this purpose on the Settings page.
Here are some important notes on using this setting:
- Select “Show” to display the Replies tab, and select “Hide” to hide the tab.
- If your predefined content is stored and managed via Campaign Builder, and you want to offer it to agents via the On-Demand Recommendations widget, set it to “Show.”
- (Recommended) If your predefined content is stored and managed via knowledge bases in KnowledgeAI, set it to “Hide.” There’s no need to show the Replies tab, as your predefined content is offered via the Answers & Bots tab according to the (highly flexible) knowledge base rules that you’re set up in Conversation Assist.
This immediately above is our recommendation, which many brands prefer. For the best agent experience, put your predefined content in KnowledgeAI. By doing so, you can offer it as answer recommendations in conversations and via the Answers & Bots tab in the widget. In the knowledge base rule within Conversation Assist, take advantage of both query contextualization and answer enrichment for optimal results.
Tip: The agent might need to reload the page, or log out and back in, to see a change made to this setting.
Query Contextualization is now available to all brands
Back in February we introduced the capability to contextualize the user’s query before searching the knowledge base for answers to offer as recommendations to your agents. Doing so can significantly enhance the quality and relevancy of the search results. (Read the February 26th release note.)
This capability remains in Early Access release. However, in this release we make the toggle visible to all brands. It is Off by default.
Don’t hesitate - check out this new feature today. And leverage the reporting enhancements described below to analyze improvements in the agent and consumer experience. Send us your feedback!
Analyze the effectiveness of recommendations benefitting from Query Contextualization
In our previous release, we launched Query Contextualization, which contextualizes the user’s query using the conversation context in order to enhance the accuracy and relevance of answer recommendations.
In this release, we enhance Conversation Assist’s reporting, particularly the KB Usage report to include the following new columns:
- Consumer Query
- Contextualized Query
- Query Contextualized?
Note: After you turn on the toggle that powers query contextualization, it takes the system 24-48 hours to start populating this data in the report.
Use the info in these new columns—alongside the existing columns for “Conversation ID,” “Recommended answer,” “Recommendation used,” and “Answer sent to consumer,”—to understand if the knowledge base is returning relevant answer recommendations that are used by your agents.
Taking things one step higher, use the Conversation Assist dashboard, Agent Summary report, and KB Summary report to determine if the answer recommendation use rate is improved.
Tips: Not familiar with Query Contextualization? Get acquainted here. Or, learn how to access and download reports.
