Email recommendation skill for CSM

Gianluca7
Tera Contributor

Hi,

I need to implement the CSM Now Assist skill for the ‘Email Recommendation’. What configuration options are available? Which inputs should I use to maximise the skill’s effectiveness? What steps should I follow to get the most out of the recommendation? I attach my basic configuration.

Regards

4 REPLIES 4

Tanushree Maiti
Tera Patron

Hi @Gianluca7 

 

Please refer: Now Assist for CSM - Email reply recommendations

Now Assist Email Recommendations will shock and amaze you

https://www.youtube.com/watch?v=0t2qk9WR5L0

Now Assist for CSM - Chat reply recommendations

Please Accept the solution if it assisted you with your question & Mark this response as Helpful.
Regards
Tanushree Maiti
ServiceNow Technical Architect
LinkedIn: https://www.linkedin.com/in/tanushreemaiti
Agents use Now Assist for CSM to instantly generate and refine personalized email replies based on case details and customer context. #servicenow #servicenowdemo # This video was recorded in the Zurich release. For more information, see: ServiceNow product documentation: ...

Gianluca7
Tera Contributor

Hi,

Thank you for the response. 

Just a point, at the moment I don't have a structured KB in my dev instance, can I start usino efficiently the email recommendation skill using the case context and previsious mail? In case how much should I customize the OOTB skill?

Hi @Gianluca7 

 

Without a KB, the LLM will rely strictly on case history, standard business logic, and previous emails.

Refer Servicenow documentation: Email response skill 

"Reply recommendations are generated based on KB articles, similar chats, and the context of the conversation."

Please Accept the solution if it assisted you with your question & Mark this response as Helpful.
Regards
Tanushree Maiti
ServiceNow Technical Architect
LinkedIn: https://www.linkedin.com/in/tanushreemaiti

Gianluca7
Tera Contributor

Ok thanks so should I start without KB or do you think it is very important to use the KB articles to use email recommendation very efficiently?