You are here:
Identify Contact Reasons with Einstein Conversation Mining
Einstein Conversation Mining uses advanced AI to discover insights from your conversational data. Use these insights to build new capabilities for your bot, update business processes to streamline common case reasons, and write knowledge articles.
Required Editions
| View supported editions. |
How Einstein Conversation Mining Works
Customers reach out to a contact center for a reason, whether to change an appointment, cancel a subscription, or get technical support. Einstein reviews conversation data and extracts the contact reason. Similar to a field that’s used to categorize a case, a contact reason can be specific, such as “fix the camera on my phone,” or general, such as “repairs.” The specific piece of customer dialog from the conversation used to identify the contact reason is known as an excerpt.
Across thousands of records, Einstein Conversation Mining reports show you your data grouped into these contact reasons. The reports also calculate key characteristics for each contact reason: the average conversation frequency and the average number of conversation turns to completion. Contact reasons with high frequency and low turns are the best opportunities for automation. Einstein Conversation Mining uses these metrics to prioritize contact reasons that are meaningful and actionable, giving you the contact reasons that give the best return on your time.
Conversation turns refer to the number of times a conversation moves back and forth between a customer and a service rep. Here’s an example of a multiturn conversation.
| Turn 1: Service rep | I can book a new appointment for you. Are you free next Tuesday, May 25, at 3 PM? |
| Turn 2: Customer | No, do you have anything else later that day? |
| Turn 3: Service rep | Yes, 4:30 is also available. Would that work? |
| Turn 4: Customer | Sure! |
| Turn 5: Service rep | OK, you’re confirmed for May 25 at 4:30 PM. |
How to Use Einstein Conversation Mining
Create a mining report with Einstein Conversation Mining. Your report can filter conversations based on date or on any field in the channel record. We recommend that you build multiple reports based on different parameters to see the results change, such as by product line, customer type, or location.
After the report is built, view the report and act on the results. In Einstein Conversation Mining, send excerpts to a bot intent. After they’re converted, excerpts become utterances tied to an intent. If you’re in Einstein Bot Builder, access Einstein Conversation Mining reports to build bots or add dialogs to bots.
Rights of ALBERT EINSTEIN are used with permission of The Hebrew University of Jerusalem. Represented exclusively by Greenlight.
- Enable Einstein Conversation Mining
Enable Einstein Conversation Mining in minutes. - Build an Einstein Conversation Mining Report
Create Conversation Mining reports based on specific dates or fields on your channel record. - Einstein Conversation Mining Report Results
Review your Conversation Mining report results and act on the top contact reasons by adding them to a bot. - Guidelines and Limits for Einstein Conversation Mining
Review the guidelines and limits for Einstein Conversation Mining.
