Occupancy, also known as utilisation rate, is the period of logged-in time agents spend on call-related activity. ‘Call-related activity’ covers talk time, hold, and after-call work (ACW).
An occupancy rate of 75% indicates that agents spent that proportion of their logged-in time on call-related activity. The remaining 25% of their logged-in time was spent waiting to take calls.
The simplest interpretation of occupancy suggests that the rate should be kept as high as possible, meaning agents use all the time available to them. While it is broadly true that centres with high occupancy rates are running more efficiently, such a simple understanding of this data falls short of representing the realities of call centre planning.
The most obvious consequence of occupancy rates that run close to 100% is agent burnout. Higher occupancy means little or no time between calls. This is known to drive down both agent satisfaction and performance, and may ultimately result in agents leaving the call centre.
Another serious issue with high occupancy is a lack of spare capacity. A high but sustainable rate of 70% indicates that traffic could increase further without seriously impacting service levels. When a centre approaches 100% occupancy, however, the resulting queues and low average speed of answer (ASA) represents a different and no less significant form of inefficiency.
How to manage occupancy
Occupancy requires maintenance. When it is too low, agents have little to do, and the contact centre is essentially wasting money. When it is too high, agents are overburdened, and the centre is operating too close to its capacity. Call queues will grow, and service level suffers.
The most direct way to influence occupancy is by changing the number of agents taking calls. How effective this will be depends on the number of seats at the contact centre. Smaller centres are more affected by changes in traffic volumes, but can more easily respond, as fewer agents represent a higher proportion of the total workforce.
Blended agents – agents who can take inbound or outbound calls as required – make it reasonably straightforward to adapt. When this is not an option, the more cumbersome approach is to ask agents to come in or send agents home.
Rather than being sent away, agents can be assigned to other activities like email correspondence or e-learning. However, while this does fill time, the inability to measure progress means it is not guaranteed to be any more efficient than having agents simply wait to take calls.