Manage Problems
Learning Objectives
After completing this unit, you’ll be able to:
- Describe how to create problem records and evaluate key data fields needed for root cause analysis.
- Explain how to use Agentforce IT Service to identify the root cause and drive long-term resolution.
Progress to Problem Management
Earlier in this module you learned the fundamental difference between incident management and problem management. You learned that often incident management restores services now, and problem management finds the root cause to prevent the issue from happening again. In this unit, you explore problem management with Agentforce IT Service.
Déjà Vu: Overlapping Operations
If you felt comfortable learning how to manage incidents in the previous unit, you’re already halfway there. Agentforce IT Service provides a unified console, and that means many operations work exactly the same way for both of these key service management processes.
For example, you can apply your existing knowledge to these shared capabilities.
-
Create and find records: You create a new problem or find an existing one using the same list views and New button logic. Of course, problem record details vary slightly from incident record details.
-
Swarming: Just like with incidents, you can spin up a Slack Swarm for a problem to bring in experts who collaborate to determine the root cause analysis.
-
Associations: You link related records—such as the incidents caused by the problem—using the same Associations tab.
-
Chatter and activity: The Chatter feed allows you to post updates, tag team members, and track field changes just as before.
Next, learn the functionality that’s unique to problem management.
Assess Impact and Risk
Unlike incidents, where speed is everything, problems require careful analysis of risk and impact. Sometimes, the cost of fixing a problem is higher than the cost of living with it, or there are limitations to implementing the fix. Agentforce IT Service gives you specific tools to document these decisions.
From the App Launcher (
) select Problems. Once you open a problem record, you can see that the details are very similar, but the goal here is different.
If you identify a root cause but determine that a permanent fix is too expensive or risky—such as replacing a legacy server that is due to be retired anyway—you can officially Accept Risk. Located in the action menu, this operation allows you to enter a business justification for why you are closing the problem without a fix. This formally acknowledges the issue and closes the record, ensuring your team doesn't waste time troubleshooting a known limitation.
Reopen Problems
Note that a closed problem doesn’t have to stay that way. If new information becomes available, you can simply click Reopen from the action menu. For example, if a VPN software upgrade didn't resolve an issue across all enterprise locations and remote workers, you might reopen the problem.
Fix the Root Cause
Analyze with Agentforce
Before you can fix it, you have to prove what is broken. Since the raw data lives on the incidents, you must start there. Navigate to a related incident and use the Proactive Assistance panel to generate a root cause analysis (RCA). Agentforce scans descriptions and resolution notes to draft a summary. Copy this finding, return to your Problem record, and paste it into the Root Cause Summary field to help document the issue.
Act with Change Management
Once you confirm the root cause (for example, a "Memory Leak in v14.2"), you need to implement a permanent fix. You can generate a new change request record directly from the problem to track the deployment of the fix (like "Rollback to v14.1"). By linking the change request to the problem, you ensure that once the change is deployed, the problem can be officially closed.
Document with Knowledge Management
Finally, make sure your team doesn’t have to rediscover this solution from scratch. You can capture the resolution—whether it’s a Known Error, a Rollback Plan, or a specific Troubleshooting Guide—as a new knowledge article to help future fulfillers.
Put It All Together
Here’s what the progression might look like for the example VPN issue. As you can see, it involves many incident and problem management operations.
Sequence |
Operation |
Who Performs It |
|---|---|---|
1 |
New incident created |
Employee |
2 |
Major incident proposed |
Incident fulfiller |
3 |
Major incident accepted |
Incident manager |
4 |
New problem created |
Problem manager |
5 |
Series of troubleshooting steps (Swarm, root cause analysis, and so on) |
Problem fulfiller, problem manager |
6 |
Accept risk |
Problem manager |
7 |
New change request |
Problem manager |
8 |
New knowledge article |
Problem fulfiller, problem manager |
9 |
Problem closed |
Problem fulfiller |
10 |
Problem reopened |
Problem manager |
11 |
Problem worked to resolution and closed again |
Problem fulfiller, problem manager |
You’ve seen how Agentforce IT Service extinguishes fires (incidents) and helps fireproof the building (problems). In the next unit, you head to the control room to see how to track it all using dashboards.