Select Page

Workforce management software will predict the volume and length of calls.  The reality is that the forecasts will not be very good without someone challenging the results.  Does this mean that the tools are bad?  NO!  It means that people who understand the business are necessary to add accuracy to the forecast.

First of all, the forecasts that workforce management tools offer are derived from history with some minor accommodations for changing business conditions.  Most tools will allow for a growth/contraction factor.  This is applied to historical patterns.

This logic will work if your business if fairly stable and the volume is predominantly normal and recurring.  Most contact centers receive 5% or more of the volume from special events (weather, marketing  campaigns, new product rollouts, quality issues, etc).  Those events are likely to recur at some time in the future – but not necessarily at the same time on the same day – and certainly not on every day of the week from this point forward.  For this reason, workforce management must be committed and precise in cleaning data.

The second major reason for failure of the tool’s forecasting ability is that the underlying logic for forecasting.  The software is going to use either a simple average or a weighted average of prior periods to predict volume.  This methodology might work sometimes for some centers but it probably won’t work all of the time.  Perhaps a Box-Jenkins forecast or exponential forecast would be more accurate for a business.  The most successfully planned contact centers have forecasters that are willing to create a number of forecasts using different methods and assumptions.

Personnally, I have created 20-25 forecasts for a long-term picture to see what impacts different methods will have on the forecast.  Many of the forecasts are clustered in the center with a few that predict very high or very low volumes.  I use these to determine what my contingency plans might need to accommodate.

Remember, a forecast is a prediction.  It is not going to be accurate.  This fact should not prevent us from trying to forecast.  It should encourage us to try different methods and create different forecasts.  Over time, we’ll get a feel for which method is more accurate for our business (or components of our business).  We need to be willing to try new and creative approaches in developing our forecasts.