Escalation User Options and Impacts

Modified on Tue, 18 Jul 2023 at 02:20 AM

Purpose

The Escalations page provides several different sections presenting in a kanban format. This article provides guidance on what actions you can take within each column and what impacts those actions will have within SupportLogic.


Summary

Review the impacts of decisions made in the Escalations page and from working through the Escalation Workflow and what actions will and will not help train the Machine Learning engine of SupportLogic.


Escalation Page Kanban View

Within the Escalations page kanban view cases are segmented into five (5) different sections:

  • Cases reviewed - groups reviewed cases by Acknowledged, Snoozed or Disagreed decisions that have been taken on cases (by anyone in your organization)
  • Likely to Escalate - Machine Learning (ML) model prediction based on different LTE factors
  • Escalation Requests - Signals extracted from customer comments
  • Active Escalations - Cases escalated within customer CRM by the customer/agent
  • Resolved Cases - Cases that were predicted to escalate but are now in the Closed or Resolved status


Escalations Kanban View Sections


Next, let's look at the sections you need to take actions in.



Likely to Escalate 

These are the cases which are processed by Machine learning and predicted for escalation. The support cases which may get escalated are predicted proactively and grouped based on 5 factors which leverage 25 different case meta data to predict cases that are likely to escalate. The groupings of the meta data are:

  1. High customer urgency
  2. Customer history
  3. Poor support responsiveness
  4. Case activity
  5. Case age


In the case of Likely To Escalate (LTE), the predictions are done directly from SupportLogic which allows us to perform below operations after that LTE cases clears the queue and moves into the respective columns under the Cases reviewed section.


Acknowledge (I took care of this case )

Available

Disagree ( Disagree with prediction )
Available

Snooze (Snooze this case)

Available

Trains the ML model
Yes

Three columns within Cases reviewed



Escalation Requests 

These are cases where customers or your internal company members have requested for an escalation. SupportLogic processes the case comments and it detects the request for escalation. On the Escalation page, cases with customers requesting escalations are shown under the Escalation Requests column.


Acknowledge 

Available on the Support Hub/Case view

Disagree / declineAvailable

Snooze 

Not Available

Trains the ML modelNo




 



You can also take direct action on the case and change the status to Escalated if you wish to confirm the customer's request to escalate a case. 



Active Escalations 

These are escalated cases directly updated from the CRM, the SupportLogic application shows the escalations based on the value in the CRM .


When the case is escalated, the is_escalated field in the CRM database will be changed as True. When the field changes it gets updated and the ticket is marked as Escalated in SupportLogic. These escalated cases show up in the list until the case is closed. 


For Active Escalation cases you can:

  • Add escalation notes for the users handling the case. These notes are visible only to the case owner and the user who added the note.
  • Provide feedback to SupportLogic regarding the timing of the prediction of this case's escalation, by clicking on the lighthouse icon in the case history.


Actions on Active Escalation cases



Acknowledge

Not Available

Disagree Not Available

Snooze 

Not Available

Trains the ML modelNo



Permissions required

For Users to add notes, and take actions on the Escalations page will need specific permission. Admins with access to Manage Users can switch on/off the Manage Escalations toggle for the desired users. Review the User Permissions article for more detail.


 




Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select atleast one of the reasons

Feedback sent

We appreciate your effort and will try to fix the article