I have been working on an automated system to schedule 20-30 staff for our call centre. Does any one have any comments whether automation is the best option or one shuld stick to the manual rostering . any comments would be appreciated.

Short answer is no it's not worth automating the rostering
However it is worth building an excel app for forcasting call volumes.

As a start there's a non Erlang version on my websiteHere.
You'll need to do a "Save Target As."

For an Erlang app see Here.

HTH

DaveA

I agree with Dave. Although such a system would reduce some work, for such a sized centre I think it would be difficult to justify.

a spreadsheet , such as Dave's is a useful low cost tool, with very positive benefits.

Also for those just starting a call centre team, its a great way to learn what's going on.




DaveC

Thanks a lot for the wonderful feedback!

I have to go thru the spreedsheet with a fine toothed comb as we have never worked with such a format. We just base out scheduling on the peak and the off peak times in blocks .For eg. if the peak is sum where between 1000-1300, we roster most of the staff for this period and then make a rotating scheudkle where half do the morning one week and the other half doing an afternoon shift. I might bother Dave off and on to get the explanation of few details on the spreedsheet, thogh.

rgds

Hi there,

Just to go to the next step. I have started looking at changing the way we do or rota's. I would like my boss to look at a WFM, as I hate having to spend 3 or 4 days sitting in front of excel forecasting and setting shift patterns.

Is it viable to suggest looking into a WFM for a contact centre of 82 and increasing or do i suggest an assistant?

Chris

Chris,

• How many teams have you got?


• How often are you forecasting?


• How are you measuring Adherence to rotas?


• Are you dealing with Intraday forecasting?

If you're on intraday and not doing intraday adherence
the answer is probably yes although something like Pipkins
will probably do you. You don't need a beast like IEX Totalview,
Teliopti or Aspect WFM yet.

Anything else give us a yell.

If you have a look around the board you'll find I have an opinion
or two {GRIN} on the matter.

HTH

DaveA

Hi there Dave,

Thanks for the swift response. In answer to yor questions

The centre is split into two groups each dealing with one of two types of customer, each of the groups is split into two teams.

I dont know what you mean by how often am I forecasting, I enter actual data the day after, so my forecaster is always producing an acrate 7 day forecast.

How am I measuring adhearance to the Rota, This in my opinion is for the Team Leaders to do with their teams, however I have been looking at log-in and log-out reports and reporting this to the team managers.

Are you dealing with Intraday forecasting? I dont know what you mean by this.

Thanks

Chris

Intraday forecating is where you look at the calls to date that
day Vs the forcast and re-roster off of the back of that.

All ou need at the base level for adherence is look at 'Agents available'
for each 1/2 hr Vs what you forecast.

As far as the forecasting goes using the previous weeks data has one slight drawback in that you are only looking at one data point for that day. I tend to
use the previous 6 weeks as a basic guide with the nearest 2 weeks from the previous year as a weighting factor to allow for seasonality.

This means you won't get caught out by a sudden annual / quarterly event you didn't see coming because no one told you!

HTH

DaveA

Hi there Dave,

You seem to be on a role.

My forecaster takes into account the last 5 weeks of data in 14min slots, so that I predict the call volumes.

I have been telling my line manager about the fact that we are currently over staffed, as we have enough staff on each customeer type to anser between 2 or 3 times the call volumes we currently get, the only reason we loose calls is becasue they dont adhear to the rota and meet their log on targets and AHT targets. Some thing that I can bring to the Team managers attention with the stats I produce, but for them to run with the ball on.