Staffing-based vs. service level-based metrics

[TeamLead8]

Hi,

This is my first time here. I find this forum very informative. Thanks to the experienced members of this group. i hope i can be of help to most people in this community as much as you guys are of big help to me.

Im working on a report right now and part of which is to determine the difference in metrics between a staffing-based and service-level-based call center... Things to consider to meet these metrices in order not to be penalized. Would appreciate if someone can give me a general information regarding my conncern.

Thank you very much!

[lavarthisantosh]

Staffing based Vs service level - based metrics

Hi ,

Let me explain you the difference between staffing based call center and service level based call center

Staffing based call center requires a particular number of FTE's to be staffed for every half hour.

For example

09:00 - 25 FTE
09:30 - 40 FTE
10:00 - 45 FTE
10:30 - 60 FTE

Here in the above example irrespective of the call volume the call center should have the above mentioned FTE staffed in that particular interval with productive Auxes like (Handle time,hold time,wrap time or on avil mode)the FTE should not be neither on any breaks nor in any team meetings. Beacuse here the concentration is more on productive FTE numbers irrespective of what the call volume is .

09:00 - 25FTE .we might receive on 10 calls however we need to staff 25 agents as this is the main key metric.
09:30 - 40FTE .we might receive 250 calls in this sector which can be handled by 50 FTE's however the call center doesn't need to worry on the service level nor on the abandon rate because it just need to staff only 40 FTE's

Key metrics

FTE's staffed
Average handle time or productivity
Schedule adherence

NON key metrics

Average speed of answer
Service level
Abandon Rate


Service level based call center requires the no of calls to be answered in the predefined service time

Example :
Irrespective of the call volume in any particual hour,Irrespective of the no of FTE's a call center has it needs to answer all the calls which are coming through in that pre defined service level time (calls should be answered in less than 15 secs with an abandon rate of 1 % )

Key Metrics :

Average spped of answer
Service level
Abandon rate
Schedule adherence

NON key Metric

No of FTE's staffed

[livebean]

Well metrics in a CC should be ideal based around the SL, as SL is the basis for resource allocation in accordance to the level of service in other words for quality of service.

If you base your resource allocation on staffing then in almost all of your time intervals you would be quite over or under staffed. Thus having an absolute negative impact on your efficiencies and quality.

Use workload and not call volume.. very common mistake most of the self claimed WFM experts make.



Write to me directly in case you need any help

Mahesh Punia

CEO
LiveBean Consulting
maheshpunia@livebean.net

* We are CC specialists.

[johnysmith]

This thread is really great. And thanks for starting this thread...

[TeamLead8]

Thank you so much for the responses... they were very helpful... just a follow up question, In a staffing-based call center, is there a way where we can actually tell real-time if we are going to pass an interval? what we do is we get the average no. of agents staffed every 5th, 20th and 29th min of an interval (half-hour). Personally, i dont think this is effective because what actually counts is the no. of total hours of all agents logged in during a particular interval and not the actual # of agents. Is there really a standard way of checking if we need to add x no. of agents just to make the interval?

Thank you so much!

[shua79]

So we use interval based staffing metrics for one of our outsourcers at my company. What you want to do first is set up how many PRODUCTIVE FTE you need on the phone, not just how many FTE you want staffed. (Though, if you take AUX out of this chart I show, you can have staffed FTE variance just the same) This way you are measuring how much of your staff is on the phones when they need to be.

To figure up interval based success, parse your ACD system for staffed time, (we get ours in seconds, shown below) and aux time. In some cases you may have multiple kinds of aux (Break, planned, unplanned, technology errors, etc) and you may want to exclude some forms of aux from being included, while including others. Now take AUX out of staff, and derive how many hours of staff you have per interval, and then turn that into actual productive FTE... now use this formula...

1 - (required FTE / Productive FTE) = staff variance

Now then you just set up a metric... We allow for 5% under to 10% over in interval staffing for this particular outsourcer.

In Seconds
Req FTE Staffed Aux Staff - AUX Staff Hrs Prod FTE Var
9:00 AM 35 8769846 1753969 7015877 1949 32 -7.76%
9:30 AM 55 13565479 1532899 12032580 3342 56 1.27%



Now on a side note, I really thought you had originally been talking about scheduling accuracy... which is different, that measures how well you are SCHEDULED per interval vs required... to do this you take...

1 - (Abs(Req FTE - Sched FTE) / Req FTE)

our goal per interval is 88% or above and to hit that 88% of the time.

Here's an example... in this you can see you are going by interval requirements so a smaller interval requirement allots for fewer available variants than a larger interval... (In the 9:00 AM interval we are only 5 off required but miss the variance allotment while at 10:00 AM we are 8 over required and meet the variant requirement.)


Req FTE Sched FTE Variance
9:00 AM 35 40 85.71%
9:30 AM 55 53 96.36%
10:00 AM 102 110 92.16%


Let me know if you have any questions about these.