Forecasting is a part of many business situations; deciding whether to hire a new class for a call center, deciding how many resources need to be deployed to support an event, or predicting water or power generation needs for a large city. Regardless of the variables or time horizons, forecasting is the foundation to support effective planning. Forecasting varies widely in difficulty based on both the event we're attempting to forecast, and the time horizons involved; forecasts for what time the sun will rise tomorrow are extremely reliable and accurate, where forecasting the price of a stock 2 years in the future is far more difficult.
Determining how accurate a forecast on a future even may be depends on several variables:
1) the forecasters understanding of the factors that influence the forecast,
2) the quality and quantity of the data available,
3) whether the person(s) conducting the forecasts, or the forecasts themselves will directly or indirectly influence the forecast itself.
In contact centers, forecasters generally are concerned with variables 1 & 2. While the head of the federal reserve making a forecast on future interest rates is highly likely to influence the interest rates itself, contact center forecasters are unlikely to have influence on call volume, handle time or other attributes which determine staff required.
Call Center Forecast Factors
Call center forecasters must understand a number of factors when forecasting. When forecasting staff required to service demand, forecasters must work with call volume, handle time, shrink, occupancy and attrition inputs to accurately determine future staff required to achieve a company's targeted service level. Each of these variables have factors which influence the it. Call volumes are influenced by historic customer demand for services offered, the growth or decline of the customer base, seasonality, alternative options for customers self-serving their requests, or the availability of alternative channels for service, just to name a few. Handle time, a critical variable in forecasting staff required is influenced by the types of calls being offered (simple vs. complex inquiries), the tenure of the call center agents, the performance of the systems used to service the calls, and the process steps (and process compliance) required in servicing the customer. Likewise, shrinkage, occupancy and attrition also have many factors influencing these values as well.
The accuracy of a call center's staffing forecast is dependent on the forecaster's ability to understand and capture these factors.
Data Quality and Quantity
Even with an in-depth understanding of the forecast factors, the quality or quantity of the data itself may not be available to the forecaster. When a new product or service is launched within a company, there may be little or no historical data to leverage when determining customer demand. When data is available, it must also be inspected for the quality and relevance to the forecast itself. Historical call volumes from 3 years ago may have bias based on a product or service which is no longer offered.
Forecasters must take care to assess whether the quality and quantity of the data is appropriate for generating future forecasts.