Solving Pain Points of Contact Center Forecasting

contact-center-male-operator-300Anyone who has ever been responsible for ensuring that the right number of contact center agents are in their seats at the right time can tell you that producing an accurate schedule can be complex.

An accurate forecast is crucial for accurate scheduling as well as reaching and maintaining desired service levels.  If you are experiencing pain points in trying to produce an accurate forecast, you need a strategy that will help ensure you are maximizing available agents with call demand.

Everyone knows that the volume of work arriving in a contact center is quite random.  If we look at a history of work arriving in a typical contact center, this appears to be true.  The volume of work does indeed vary from one interval to the next, day-to-day, and week-to-week. These are facts that may give the impression that accurately forecasting future work is an impossible task.

There are two simple effects of getting the forecast wrong and both cost you money.

Forecasting too high: Overstaffing creates a scenario for idle, unproductive agents to suffer from low morale or become bored or distracted and not focused on customer service.

Forecasting too low: Understaffing can result in lost customers due to abandoned calls or poor customer service.

It should also be remembered that both of the above could occur on the same day if the forecasted distribution of work is incorrect.

Four common problems involved with producing an accurate forecast

  • Using averages:  Forecasting an average is a safe bet (and the easiest to perform) but is unlikely to be the most accurate.  Very little that happens in a contact center is an average.  What happens next week is unlikely to be an average of what happened over the last few weeks.
  • The forecasting tool lacks enough data:  Generally speaking, the more data the forecasting tool has to work with, the greater chance of producing an accurate forecast.  If the forecasting tool cannot process more than a few weeks of data, its accuracy will be compromised.  A good rule of thumb is the more data, the better.
  • Having unrealistic expectations:  The forecasting tool’s predictions can be based only on what has happened historically and on what it is told will happen in the future.  It can never know more than this!  This may sound obvious, but don’t expect an accurate forecast for the coming year if you have only a few weeks data to forecast from.
  • Not understanding how your forecasting tool works regarding:
  1. How much data it can store/use
  2. If it takes into account inflation due to abandoned calls
  3. If it recognizes seasonal and growth trends
  4. If special event information can be input and correlation factors applied
  5. How all of this is accomplished

A poor forecast can result in high staffing costs and lost customer revenue, but forecast accuracy depends on many factors.  The key is ensuring that the forecasting tool has as much information about what happened in the past and what you expect in the future, and that it will allow you to properly utilize it.

An additional consideration for ensuring the forecaster has sufficient information is the quality of that information.  The nature of the data is important.  Validate your historical data by comparing new incoming data against a previously validated set of historical data.

Recognizing events. It is also important to recognize events which have an effect on the amount, and possibly the pattern, of work arriving in the contact center.  Besides holidays, other events may include billing cycles, mail or catalog drops, advertising promotions, new business activity, competitor activity, weather issues, or external factors (TV shows, sporting events, industrial actions, etc.).  The lifespan and seasonal trends of each type of event should be given consideration.

Correlating events. In the example of mail or catalog drops, similar events may occur on several occasions, but will affect work differently based on number of letters delivered.  The system must have the capacity to identify and appropriately weight these events to plan for future occurrences.

Critical components for accurate forecasting
Ensuring your forecaster takes these bullet points into consideration can help solve your forecasting pain points.

  • The amount of historical data available
  • The nature of the data
  • The forecasting period
  • An infinite number of different service objectives on one or more work streams
  • Algorithms that reflect real life customer behavior
  • Special events are treated differently, i.e., mail drops, campaigns, and special promotions can be quantified
  • Email and faxes have service objectives reflecting the way that work is handled

Bob Webb is the vice president of sales with Pipkins Inc.