Service Level Maximizer. What is it in plain English?

[harryhoudini66]

Can someone explain to me in layman's terms what this distribution method does?

My call center has a goal of 80% of calls answered within 20 seconds. We are currently using EAD to distribute calls. Does it sound like we should be using SLM instead?

Service Level Maximizer (SLM) With SLM, the objective is to maximize agent utilization to meet "percent within service level" targets. The service level target is defined as X % of the calls for that skill are answered within Y seconds. SLM selects agents based on the target service level objectives assigned to the hunt group for each skill. The call is selected for the skill with the most negatively weighted service level relative to the administered target. The agent selected is the agent whose skills have the highest average weighted service level relative to the service level targets administered to the skills.

 

[avayafreakje]

in normal terms this means:

There is a pool of reserved agents which will be added/removed temporary from skills based on the service level defined on the skill form (x%call in y secs)

The result then is a balance of service levels between the skills. IF service level for one skill drops the system adds agents. If the service level then meets the reference (skill form) the temporary assigned agents are release of that skill.

Another interesting question is the difference between business advocate en slm. Basicly the advocate feature uses predictive wait times between the skills and the slm works with the skill reference.

 

[oclaudiox]

Please, there is a easy way to verify (to calculate) service leval, if I dont know "accepteds" ?

Claudio.

 

[BIS]

Is anybody using this? And if so what is your experience? We are also using ead-mia - would it make sense to change?

 

[harryhoudini66]

This sounds awesome. He have agents that have mutliple skills assigned to them. We only have goals specific to our general queue and we have not met these goals (80% of calls answered in 20 seconds) in two years. I am wondering if enabling this feature would benefit us.

We take an average of 5000 per day and have a staff of about 80-90 agents.

 

[avayafreakje]

this could be a benefit when you have several skills with service levels on which your agents can take calls for (as in EAS) but want automatic balancing of service levels between those skills

If you have the RTU for EAS-PHD then you also have the slm feature but i believe its turned off in the license file by default

As far as i understand does the setting on the skill form x% of calls in y secs only works if you use slm

@20060612

If you have an acd skill you can define service goals on vdn form and skill form. Those goals are used as reference in basic call management system(BCMS) and call management system (CMS)

There are also some tools. check this link:
http://www.math.vu.nl/~koole/ccmath/

 

[smurfyIAG]

Harry, I wonder whether you are over-complicating this?
The way I understand your last statement, your service level measurement is only on one skill.
But, because of the servicing of multiple other un-measured skill calls, you are failing to meet the service levels on that primary skill.

Have all agents been assigned the primary skill as their highest priority?
Are the vectors queueing calls to that skill at the highest priority level and the others at a lower priority?

You could even go overboard and include some basic vectoring in the lesser skills like:
Don't queue to skill unless number of available agents is greater than 2.

Sean Murphy
Call Centre Analyst
IAG New Zealand

 

[harryhoudini66]

The truth is I am just diving in to the phone system. This has been run by a few people in the past and I am very sure they are not doing a good job at it. In the little research and reading of manuals that I have done, I have come to find out I know more then they do. The sad part is these are people doing the job for 20 years. Nobody even looks at the vectors, we just trust that our Telecommunications guy is doing it right. Little do they know his interest is that it works not that it is running at optimum. We don’t even monitor the VDNs.

Yes, we have many agents with the primary skill at the highest level. However, we have some agents with up to 9 skill levels assigned to them. As far as vectoring, I took a look at it and it shows the caller queues with a medium priority. I will look in to having it changed to high. Could you explain how this would speed up the process? Does this make the system work faster as far as finding available agents?

Other things we have not used and I have pushed for are Abandoned Call Search, RONA, and Overflow Skills and Scripting.

 

[harryhoudini66]

Any other feedback on this feature?