Even though my most recent banking position was in fraud (detecting it, not committing it--sorry, I know it's an old joke), I did my time in the call center, too.
I would say that the best numeric measure of your Call Center's effectiveness is not so much time-per-call as wait-time-in-queue. I say this because a call center (in some places still called the "Customer Service Department") is essentially about customer satisfaction more than bank operations. The bank may be tempted to evaluate based on time-per-call, because that's what the bank is paying for, but you have to look at it from the customer's point of view. Customers do not begrudge time they spend talking with a person who seems motivated and competent to help meet their banking needs. A customer with a particularly complex issue may come away from a 20-minute call feeling very satisfied. But even 30 seconds of the "All operators are currently busy" message is incredibly aggravating.
If the average customer's wait for a representative is under 30 seconds, you are probably overstaffed. But if customers are typically waiting over a minute, something may be wrong, and if they are waiting over three minutes, something is definitely wrong, whether with the number of staff, the training they have received, or the software they are using to investigate and resolve customer service issues.
Call Center work is a lot more demanding than some people imagine. Every time the phone rings, there's the possibility of a simple balance inquiry, an irate customer who has discovered a bank error on the monthly statement, a potential stockholder wanting to speak with the president of the bank, or an attempted fraud. I recommend that before you jump to any conclusions about staff levels, procedures, etc., you set aside a fair bit of time to listen to recordings of calls, short, medium, and long. Try sitting at your own computer as you listen, and go through the steps the representative had to do in order to answer the customer's query based on the information they provided. I think you'll find that time-per-call is only a useful measurement when averaged over hundreds of calls.
When I worked in the Call Center, our supervisor used to meet with us periodically for training, and would evaluate our performance on a randomly selected recorded call. Once, I walked in for my training session, and as soon as I heard the customer's name, I thought, "Oh, no! Not THIS call! I'm gonna be in big trouble." The call was over 20 minutes long! And you know what? It was the best evaluation I ever got. My boss said, "You were so patient with that elderly customer, and you were finally able to solve her problem."
_________________________
Opinions expressed do not necessarily reflect those of my employer or of my cats.