Skills vs Agent Groups

[cavagnaro]

Hi guys,
I wonder if some of you can share which points makes you decide to suggest to your customer to use or Agent Groups or Skills with VAG...
I'd like to learn a little more of experience on this issue.

Thanks!

[Tambo]

Hi Cav,

We use Skils_VAG for routing within the strategy which works well with our multi skilled environment. Each dept may have skills that over lap but certain ones which don't, this also ties in well with our reporting which is VQ_Subdept_Skill which again works perfectly with IEX Total  view for WFM.

We only use Dept VAG for CCPulse monitoring. This allows each dept to monitor their agents whilst seeing that they are maybe using their overflow skills to answer calls.

Is this of any help?

Tambo

[cavagnaro]

Hey Tambo, thanks for your input. But I guess I would like to know why you choose Skills, I mean, supervisors I guess constantly train the agents, right? I also guess they have several skills, not just one. And maybe the routing is based on a combination of 2 or more skills? That is why you use skills rather than simple Groups?
The supervisors actually refresh the skills from time to time?
These are some of my questions and doubts...I have found that here in my country as CC are quite smalls they don't use the skills as perse but just as agent groups...so that is why I would like to understand other markets

[Tambo]

Hey Cav,

I think it started from us historically being SBR led and us going with what we knew and were comfortable with when we took on Genesys.

We actually have lots of skills for agents and at times they maybe have too many (12) attached to them to be productive, this can happen at peak times.

For instance a customer calls for an invoice quiery, this goes to the correct dept and is answered by an agent who can deal with it with an advanced skill. A customer calls to make a payment this can go to anyone anywhere as payments are taken in the same way sales or service. We have lots and lots of different depts using different systems to deal with the customers and each dept is quite specialised in it's products (there really isn't too much overflow although we do this where we can)

The on the day skills are managed by Mission Control dependant on what is busy and what the business requirements are at that point.

A skills matrix is run each morning through putty tio set skills back to our agreed default levels for each agent.

Our call centres add up to around 2000 agents.

Is that better or have I just confused myself too lol ??

Tambo

[cavagnaro]

So you only route to something like Skill("invoices")>8 and nothing else? or goes increasing?

Skill("invoices")>8
30 secs
Skill("invoices")>6
60 secs
Skill("invoices")>4
60 secs
Skill("invoices")>4 & Skill("Support")<6

If you would choose to swap to agent groups only, would it "hurt", meaning, really hurt your business logic?

[Kevin S]

Cav -
We use an escalation similar to what you describe (with quicker escalation times, and only three tiers) for skills.
We use skills to manage who an agent "focuses" on for that day. For example, an agent may support 8 skills (partner relationships) . If we need them to focus on our 3 high touch relationships for a day, we adjust the other 5 skills to 0. If we then need them to be available for those other skills, but we want the call to go to them only if no other skilled agent is available, we will set them at a lower skill.

Fortunately, we do not support language, seniority or "proficiency" (beginner, intermediate, expert, etc for a specific feature or product line) skills , so our skill matrix is one dimensional.

[Tambo]

Hi Cav,

Yes in essence thats the basics of it...........

skill (invoices) >8 wait 5 secs
skill (invoices) >7 wait 5 secs
skill (invoices) >6 wait 5 secs
skill (invoices) >5 wait 5 secs

the theory being that your call should get answered by a highly skilled agent rather than an agent who is on overflow (level 3 or lower say) and is there just to help get to service levels of PCA.

we are looking to try 'agent groups' in certain depts but training for some of the call types is needed because historically sales and customer service have been kept seperate, however, there is a belief that if the sales agents are more aware of the issue they can cause when they take shortcuts in their calls then we would get less service calls.

Unlike Kevin we do have senior or proficiency skills and the matrix is very hard work to keep up to date and accurate. I am keen to see if this is better as SBR does have it's limitations. I would say that SBR is most effective when you are relatively busy. When you are quiet then the same agents get targeted bbecause they have a higher skill level and when you are really busy then whoever is in available gets the call because it's been waiting.

[Gulden_NL]

Cav,
The post about sizing is spot on, we only have large pools of agents, so using skills gives a lot more control over prioritization.  For example, due to seasonality, Skill 1 may be fairly low, but from Nov 15-Feb 1, it may be top priority.  All agents have multiple skills, and they could be across several (or many agent groups), so this allows for tweaking the skill without messing around with supervisors, groups and such.

One feature that I'd like to point out is that using the latest Genesys WFM, you can tie training into a "skill quiz" created by the business that once the agent successfully completes the quiz, their skill is automatically updated in WFM AND in Routing.  So perhaps the agent can handle Skill 1 calls but only at a basic level. Once they have passed the quiz, they are now a Skill 1A and can begin receiving calls that require more experience/training and nobody has to look at a report, go into Administrator and update their skills. 
***Verrrry slick!***

[cavagnaro]

Hey guys, thanks a lot for all this info, it is quite interesting to read about your experiences with it. So far my previous idea that this works better on big agent groups seems to be correct...on small groups like what I'm used to is not so helpful.
One question, when you change the skill level to a set of agents, how much time can you spend on this task?

[mduran22]

Cav,

We have scenario in which we use static and dymanic skill expressions for routing. I would imagine this could apply to a contact center of any size. We have certian states where rules are different as far as training and licensing to handle a customer. We have static expressions normally, but if we have a call from a state which requires special handling we dynaically update the skill expression to add or remove a skill. Say we have 9 different static skill expressions and 4 different location type calls which can affect 5 of those skill expressions, if we used agent groups we would have to create 20 different agent groups to accomodate for those combinations. With dynamic routing we add a "specialty or location" skill to the static skill expression dynamically to appropriately handle the interaction.

Mike

[Tambo]

[quote author=cavagnaro link=topic=6781.msg29523#msg29523 date=1320856514]
Hey guys, thanks a lot for all this info, it is quite interesting to read about your experiences with it. So far my previous idea that this works better on big agent groups seems to be correct...on small groups like what I'm used to is not so helpful.
One question, when you change the skill level to a set of agents, [b]how much time[/b] can you spend on this task?
[/quote]

Cav, that's difficult to answer as each day brings it's own pressures. If WFM has the correct agents scheduled and calls arrive as predicted then virtually no time at all, but if you get call spikes from an unexpected area then it can be very time consuming trying to balance the skill levels etc. I would imagine though that this would be similar if you needed to move agent groups when call spikes arrive? no ?

[cavagnaro]

Thanks a lot for all this nice info
Definitively I wish I could see some implementation to understand it more but with all this info now I have a better panorama
Thanks again!