These capabilities reduce manual effort and improve consistency and case quality in documentation and communication. Paired with agentic orchestration, agents can also receive suggested actions or execute background workflows in real time all within the case context.
Virtuous Content Loop
Agent productivity features contribute directly to a continuous improvement cycle for content quality and resolution accuracy. By capturing structured and high quality outputs like AI-generated summaries, recommended steps, and knowledge/resolution notes generation with guidance from generated suggested steps, the system feeds insights back into both the knowledge base and automation design, increasing the health of the transactional data.
This loop ensures that:
-
Resolution notes and summaries are indexed for future AI Search and suggested answers
-
Suggested steps and article gaps, now resolved, are flagged to content authors via feedback loops
-
Agent usage signals inform which knowledge and templates are most useful
- Both customers and agents reduce the usage of tribal knowledge and start using better content
-
Usage data and resolution patterns help identify automation opportunities through Process mining, clustering and trends, enabling admins to define agentic workflows over time.
This dynamic reuse of agent-generated content reduces future handle time, improves deflection rates, and enhances both self-service and agent assist experiences over time.
Discover more by clicking on each of the skills:
-
Summarization
-
Recommendations
-
Content Generation
Measured Success
How do we usually measure success in this set of purpose-driven skills?
| Outcomes | Explanation (with applicable use case) | Success Metric |
|---|---|---|
|
Reduced average handle time |
Empower agents with auto-generated summaries and suggested responses to reduce time spent on manual documentation and content switching Use case: Case and Chat Summarization |
Avg. handle time per interaction (minutes) |
|
Improved documentation consistency |
Standardize output across agents using LLM-generated resolution notes, suggested steps, and wrap-up content Use case: Resolution Notes, Suggested Steps |
% of cases with AI-authored summaries or resolutions |
|
Increased agent satisfaction |
Eliminate repetitive tasks and reduce cognitive load by surfacing contextually relevant knowledge and actions Use case: Sidebar, NAP Q&A, Recommendations |
Agent satisfaction score (ASAT) |
|
Faster onboarding and ramp-up |
Accelerate new agent effectiveness by embedding AI guidance and reducing reliance on tribal knowledge Use case: Recommendations and Sidebar Assist |
Avg. time to proficiency for new agents (weeks) |
|
Improved customer satisfaction |
Deliver accurate, personalized responses faster using generative email and chat replies Use case: Email Reply, Chat Reply with Sentiment Awareness |
CSAT / customer feedback per interaction |
|
Reduced escalations and rework |
Present relevant steps and recommendations early, reducing back-and-forth or case reopening Use case: Suggested Steps, NAP Q&A, Sentiment Analysis |
% of cases escalated / reopened |
|
Increase CS agent retention |
Transform the agent experience by automating mundane activities to increase job satisfaction and mitigate service agent attrition |
Customer service agent turnover (%) |
|
Reduce effort to resolve customer cases (transfer / ramp up time) |
Reduce transfer portion of resolution time per case via auto-generated summarization of case history to date (applicable to live agent escalation & virtual to live agent transfers)Use case: Auto summarize case for transfer |
Time spent on transfer per case (in hours) |
|
Reduce number of customer cases worked |
Deliver actionable AI answers and conversational virtual agent interactions to customers, leading to fewer cases to the contact centerUse case: Promoted, narrative search results |
Customer case volume (#) |
|
Reduce effort to resolve customer cases (wrap-up time) |
Reduce wrap up portion of resolution time per case via auto-generated resolution notesUse case: Auto draft resolution notes |
Time spent on wrap up per case (in hours) |
