Please Settle an Argument for Me

[BroadbandBilly]

Hi guys,

This is my first post (as you can see) & I'm hoping you can set the record straight here. Let me give you the scenario:-

There're 4 VDNs. Each VDN has it's own skill & all calls are queued at the same priority (medium). A pool of 20 agents are each skilled for all 4 lines (& nothing else) & all at the same priority (1). Call volumes to each VDN (& therefore skill) differ.

Now here's the question: assuming that the call volumes are sufficiently high (to lessen the effect of "rogue data", if you will), will each of these skills have approximately the same service level at the end of the day?

My logic (if you can call it that) is that "yes they will". They won't be exactly the same but, if they're calling exclusively on the same pool of agents then surely (despite a higher call volume on one line & a lower call volume on another) everything is relative & the S/Ls must be within a fraction of each other at the end of the day.

Does my argument stand up or have I got it wrong (& if so, please explain: I need to understand this fully)? Oh, & this is currently not a live design, which is why I've got no stats with which to test it myself. Thanks in advance.

Great forum, BTW: glad to be a member.


B'Billy

 

[DAVIDPAYNE]

I would say not necessarily. It would all depend on what you call volume is, and the average length of each of those calls. For example, you could have one or two skills that have an average talk time of 10 min, but all the others are around 2 min. you then get 20 calls on the long talk time skills. Then your SLA, for the rest are shot. Unless the Business unit can give you a better understanding of the traffic, you are better off running it live, and set them after you get some data. But I would hope the Business unit would have some idea of what the call volume, and durations should be.

 

[simreal]

If one skill got more calls durring a lower service peridod then another skill got more calls durring a time when the agents were asnwering faster .. then your numbers wont match.

If you assume that all 4 skills will get the exact same call volume all day then you should see similar numbers.

But even in the best case scenario you will have agentsanswer immediately for one skill ecause the call came in first .. but then the other skill didnt get answered as fast because the caller had to wait 20 seconds for an agent.


There are too many variables to really say one way or another. I think it's likely they will be close ... but I wouldn't risk my reputation arguing it.

 

[azphoneguy]

Why do you have separate skills if they are all staffed equally? If this is being done for reporting you could report at the VDN level and then the skill performance will be consistent.

 

[BroadbandBilly]

Many thanks for your replies, guys.

SimReal,

You've hit upon the very thing I've been wrestling with. You're correct: if one line takes substantially more calls than another, it will likely have more calls queuing at any given time. However, is this not all proportional? They have more calls queuing but also receive more calls so, proportionately, it should even out? Lines that receive fewer calls will have fewer calls queuing at any given time, but it should work out at roughly the same proportion of calls to that line, thereby evening-out the S/L, would it not?


DavidPayne,

Good point! Average talk / call handling time is something I hadn't considered. I accept that, where this differs, this can have a significant affect on S/L but, in this instance the business are expecting a uniform call duration across these lines. I know there're variables we can't control / accurately predict: there might be 20 successive calls to one VDN all within the space of 90 seconds, but this is not the typical volume pattern.


AZPhoneGuy,

The business are requesting a separate skill for each VDN as they believe the S/Ls will vary significantly. We can (of course) give them call volume from the VDNs, but can only give them S/L from the skill(s). I'm trying to avoid the creation of unnecessary skills & therefore suggested they can keep all the VDNs but a single skill should suffice.


If we assume (& I know this is an artificial scenario) that VDN 1 takes 100 calls p/day, VDN 2 takes 200 calls p/day, VDN 3 takes 300 calls p/day & VDN 4 takes 400 calls p/day, & all lines are subject to the same average talk time & S/L target (80/20), would we not expect to see an attained S/L across all 4 lines within a whisker of each other? I know the timing & quantity of inbound calls will vary but, again, would this not (more or less) even out over a given period (& the longer the period the more even it would be)?

Again, thanks for your advice. I'm just trying to get a thorough understanding of this before I make my case to the business.

Regards,


B'Billy