UCCE - skill levels

Unanswered Question
Jul 21st, 2009
User Badges:

Hi,


We migrated from UCCX4.0(5) to UCCE 7.2 last year. UCCX has a feature that lets you define skill levels for agent.


UCCE doesn't have such a feature, I now use a second skillgroup as an overflow, the LAA success node goes to say skillgroup SG.1 and the fail connection goes to 'queue to skillgroup' with groups SG.1 and SG.2 (agents only belonging to one of the groups).


I know this has reporting implications.


My questions:

1. What is the best way to emulate the skill level functionality from UCCX (e.g. what have you done?)?

2. What are the reporting implications exactly?


Tia

  • 1
  • 2
  • 3
  • 4
  • 5
Overall Rating: 4 (2 ratings)
Loading.
SAN J Tue, 07/21/2009 - 11:17
User Badges:

Make sure you use only Base skill groups as webview is built to report primarily on base skill groups. It will report on subskills too but there's no guarantee it will be accurate.

david.macias Wed, 07/22/2009 - 04:57
User Badges:
  • Blue, 1500 points or more

Another little hint about using multiple skill groups and reporting:


Symptom:

ICM, by design, counts abandons while a call is in queue against all the groups the call is queued to.

Conditions:

This situation will occur whenever a call is queued in ICM to multiple groups, and then abandons. Each group will be charged with an abandoned call.

Workaround:

There is no workaround in the existing system. This fix implements a new registry setting to control how this counting is done. See "Further Problem Description".

Further Problem Description:

This feature creates a new registry entry in the ICM router registry. This registry value should be set to 1 to enable the feature implemented by this change. This needs to be done only on one side of the central controller, and will be duplicated automatically to the other side. This registry value will not exist until the routers are restarted.

How abandons are counted based on this new registry setting

This change controls how the ICM accounts for abandoned calls when they are queued to multiple groups.

Currently, when a call is queued to multiple groups, and abandons, the abandon is changed against all groups. Specifically, RouterCallsAbandQToHalf will be incremented for all groups.

Also, if the call abandons within the service level threshold, ServiceLevelCallsAbandToHalf will be incremented for all groups.

This patch implements a new option, where the abandon will be counted against only one group. If this option is selection, RouterCallsAbandQToHalf will be incremented for only one group, and RouterCallsDequeuedToHalf will be incremented for all other groups.

Also, if the call abandons within the service level threshold, ServiceLevelCallsAbandToHalf will be incremented for one group, and ServiceLevelCallsDequeuedToHalf will be incremented for all other groups.

The group that is charged with the abandon is the group to which the call had been continuously queued to the longest at the time of the abandon. Normally, this would be the first group the script queued the call to, unless a more complicated dequeue, queue, or requeue scenario changes the order.

The term "continuously queued" means that if a call is queued to a group, later dequeued, and still later requeued, the earlier time in queue is not used in determining which group the abandon is charged against.

This change in accounting is optional, and is controlled by a registry flag. The new registry flag is:

Router\CurrentVersion\Configuration\Queuing\SingleSkillGroupAbandon

and is located on the router machines at the central controller.

The default value is 0, meaning that the original behavior is maintained, counting the abandon against all groups. If this registry entry is set to 1, the new behavior is enabled, and the abandon will be counted against only one group.

This change does not affect how abandons are counted against agents when calls are directly queued to multiple agents.

Because this change is delivered as an ES, some care must be taken if it is necessary to run setup after the ES has been applied. Running setup will delete this registry key. It will be recreated when the router in the ES is restarted, but will be reset to the default value of 0. It must be manually changed to 1 if the new behavior is required.

david

nurulchoudhury Wed, 11/10/2010 - 05:34
User Badges:

Hi David,





"Currently,  when a call is queued to multiple groups, and abandons, the abandon is  changed against all groups. Specifically, RouterCallsAbandQToHalf will  be incremented for all groups."


Abandon incrementation will be done for all groups of ICM or in all the groups where the particular abandoned call got queued earlier.



Regards,

Nurul

nurulchoudhury Wed, 11/10/2010 - 05:41
User Badges:

Hi Geoff,


Can you please let me know how it can be taken care during deployment.


Thanks and Regards,

Nurul Choudhury

jpsweeney77 Wed, 09/12/2012 - 13:40
User Badges:
  • Bronze, 100 points or more

Be aware that UCCE 8 automatically flips this reg setting equal to 1. Something to consider during an upgrade and it can be changed back in necessary. However,the key is completely deprecated in UCCE 9 and the behavior will always act as though the key is set to 1.

Actions

This Discussion