“There are two kinds of forecasters: those who don't know, and those who don't know they don't know”
Looking at big data with the power of statistical analysis gives us the ability to forecast beyond our gut instinct. In the call centre we predict how many calls we’re going to have during the month, when to make and receive calls, and to make sure we have the right people on the phones at the right time.
With large amounts of data, we can analyse patterns to influence our decision-making going forward. If someone has the highest call quality with certain types of customers, and these calls occur mostly at a certain time, then we know we should have them available for these calls.
At TieTa, there are three general types of analytics we use:
Time Series Projection
- This focuses entirely on patterns and variances based entirely on historical data. Call centres can calculate call volumes, number of agents required and the length of calls for each day:
- An autoregressive integrated moving average (ARIMA) analysis can be used for forecasting average answer time as certain dates (such as month ends and last working days) where call volumes will be higher:
- This uses highly refined and specific information about relationships between system elements and shows how importance it is to get the first step as best and accurate as you can. This type of analysis is also structured on past events. The below shows how this model flows in the call centre:
- This technique uses qualitative data and information about certain events and may or may not take the past into consideration. This is relying on call centre managers to use observations for when scheduled staff are needed to be on the phones, what lines agents should be on or who needs to be dealing with emails.
Overall, each of these techniques can be applied throughout the call centre. Different statistical and information based methods will always be used but it’s important to make sure the correct analysis is being applied at the correct time.