Skip to main content

Using Agent Assist to resolve requests faster

R
Written by Riya Sebastian
Updated over 3 weeks ago

Agent Assist equips service agents to act faster and with greater confidence. It provides AI-powered assistance directly within the Request details view, summarizing the issue clearly, performing root cause analysis, and recommending next steps. This minimizes the need for context switching and reduces time spent on investigation, improving the speed and quality of resolutions.

Whether you're working on incidents, requests, or service requests, Agent Assist brings all the information you need to one place.

Agent Assist is currently available only for requests and incidents, and not change requests.

How it works

Agent Assist generates intelligent insights by analyzing a wide range of data sources. These include request metadata including public and private conversations, historical resolutions across your service desk, and trusted, external knowledge sources.

Depending on the nature of the request, Agent Assist focuses more on the information that’s most helpful for resolving that type of issue. So, in a device-related request, it might emphasize metadata from past requests involving similar hardware. But while troubleshooting a software outage scenario, it would use system logs and resolution notes from comparable incidents.

This adaptive analysis results in rich, contextual guidance that enables agents to act with clarity. Agents can view these insights as:

  • Summary: Offers a brief, consolidated view of the issue by capturing the user's concern and any prior discussion or relevant notes.

  • Root cause analysis: Surfaces the most likely root causes based on similar tickets, web results, or known issues.

  • Recommendations: Provides next steps divided into short-term fixes and long-term solutions. These often include links to past resolved requests or trusted external resources.

Agent Assist references trusted external sources when suggesting root causes or actions, with all insights clearly cited and linked.

image.png

For example, a user reports they are unable to access Salesforce. The request contains no explicit mention of network issues, yet Agent Assist identifies patterns from previous cases suggesting VPN configuration as a likely cause. It generates a summary that highlights the user’s intent, points to the VPN setup as a probable issue, and recommends checking VPN logs as a short-term action, while also linking to a relevant configuration guide as a long-term fix.

All recommendations are clearly cited and linked, allowing agents to dive deeper as needed.

Intelligent attribute suggestions

In addition to surfacing summaries, RCAs, and recommendations, Agent Assist also helps agents fill out request attributes—intelligently. By analyzing request context such as subject, description, and comments, along with information about the requester (like department and assigned assets), Agent Assist can suggest the most relevant values for required attributes on any given request. Agents can review and apply all suggested attributes in one click, eliminating the need to guess or manually look up what to fill in.

Using Agent Assist

You can find Agent Assist within the Request details page of any request.

  • Navigate to any request, incident or problem

  • Click on the Assist tab in the right-hand panel

  • Review the summary, root cause analysis, and recommendations

  • Click on any cited request ID or knowledge source to open it in a new tab

  • Click on Attribute suggestions within the Details tab to see the list of suggested attributes

  • Review the suggestions and click on Update to fill them in

Did this answer your question?