Support reps will continue to be at the core of any customer service organization. They provide the invaluable and irreplaceable human empathy and domain knowledge that help keep customer satisfaction at high levels. In this post we will discuss a few areas where machine learning can assist these reps so that they can add more value to their customers.
- Let’s say the user has placed a call and is engaged in a conversation with the support rep. This conversation can in parallel be captured by a system, transcribed in real-time and customer’s queries can be searched against an existing knowledge base and past case history. Any suggestions to resolutions can then be surfaced for the rep. The rep can analyze these suggestions and recommend best path to resolution. This helps the rep provide faster support – they no longer have to listen and also search manually for keywords. The system does that automatically while the rep is focused on the conversation. This type of functionality is possible today with Google Speech api that can transcribe in realtime and implementing powerful search capabilities to look for relevant cases and resolutions.
- Powerful product recommendation algorithms can be used to offer recommendations to users. These recommendations can be personalized based on the user’s preferences, past order history, current location, etc. These algorithms run in the background while the conversation is in progress and the rep can use his or her judgement to offer relevant suggestions to users.
- When a user calls in and even before the conversation begins an ML algorithm can be employed to surface a profile of the user that the support rep can use to offer a more personalized service. This could be based on generic information such as user’s location, date time and more specifics based on user demographics.