Pin Me

A Critical Analysis of Different Performance Measures Used in Call Centers

written by: Sidharth Thakur•edited by: Ginny Edwards•updated: 7/25/2011

Measuring the performance of call center employees can be tricky, even if you’re using some of the standard metrics. Here’s a critical analysis that will open your eyes to how even the topmost metrics can fail in real situations.

  • slide 1 of 8

    When measuring the performance of call center employees it’s important to focus on both quality as well as quantity. There’s a wide range of metrics that can be deployed for this purpose. However, before you decide on a set of call center employee performance measures you must understand what each of these metrics means and more importantly the influencing factors that can lead to misleading results from these metrics. Let’s take a look at the five most commonly used performance metrics and examine them critically to understand how their scores can be deceiving.

  • slide 2 of 8

    1. Average Call Length

    The name of this metric is indicative enough of what it tries to measure. This is one of the most popular metrics used for measuring the performance of a call center employee. This is computed as the total time an employee spends talking to customers divided by the total number of calls attended by him, within a specific time period. A shorter average call length earns a higher performance score. Despite its popularity, you cannot completely rely on it for the following reasons:

    1. The employee may bypass all the customary protocols to cut short the call length.
    2. The employee may avoid addressing all the issues raised by the customer, just to keep his personal performance on track.
    3. Some employees go to the extreme of dropping some calls midway when they seem to get too long.

    At times, the call may get stretched because of the customer; they may put the call on hold to attend to some other issue or may take a lot of time to provide the required information. Even one such call can make the average call time to score high.

  • slide 3 of 8

    2. On-Hold Time During Calls

    This metric measures the average length of time for which the employee puts the customer on hold during a call, for things like retrieving information or processing the order, etc. Lower on-hold time is considered better, since a customer is expected to get irritated if left on hold for too long. While it is certainly a good practice to cut down on the hold time, system related problems or complex queries from the customer can increase the hold time. In such cases the employee’s performance score will suffer for unwarranted reasons.

    Evaluating Call Centre Employee Performance Measures 

  • slide 4 of 8

    3. Call Escalation Rate

    Call escalation rate takes into account the number of times a call center employee forwards the call to a senior, because he is unable to handle the situation. If an employee is well versed with all the aspects of his job, there won’t be very many chances that he may need the help of a superior to resolve customer queries and problems. But can a low call escalation rate guarantee the employee's efficiency? There are times when the employee may simply deny the customer’s request to speak to a senior. Also, the employee may drop the call or may keep arguing with the customer till the customer drops the call, to avoid escalating it.

  • slide 5 of 8

    4. First Call Resolution Rate

    This performance measure is difficult to tabulate, because no one except for the customer can say whether his problem or query was resolved in the first instance. One option to measure this is to keep track of whether the customer called back, but this is no fool proof method as unsatisfied or irate customers may not return, preferring to seek solutions elsewhere. Collecting feedback over an IVR or post call survey, after the call, are other methods being adopted these days. But these again fail, as irate customers or those in a hurry may drop the call too soon or may not be interested in leaving feedback. Also, there are times when the call may be dropped midway by the customer, before a resolution is reached.

  • slide 6 of 8

    5. Leads Conversion Rate or Sales Realized

    This measure is mostly applicable to outbound call centers, but may also exist in some inbound call centers where the employees need to make soft sales during the call. Scoring well on this metric does acknowledge good selling skills and the ability to convince a customer. But a low score may not necessarily mean that the employee is not good at selling. Some customers may call up only to get all the information and may not place the order on the phone; they may instead place the order on the website. There are also situations where the customer doesn’t place the order during the call, but returns back later to place his order – for reasons like he was too busy at the time of the call or he wanted more time to think. In such cases, the employee who receives the order gets all the credit for making the sale, despite the fact that he really didn't put in any effort, whereas the employee who convinced the customer during the first call doesn’t earn a higher performance score.

  • slide 7 of 8

    Always Use a Critical Analysis When Selecting Performance Measures

    We’ve examined only a few of the popular call center employee performance measures, to tell you how you cannot completely rely on any particular measure. For all employee performance metrics, you plan to use at your call center, do carry out an evaluative analysis to find out if the employees can cheat on them or there are if there are other factors that make their results deceiving. The best approach in using these measures to gauge the performance of employees is to use a wider range of metrics and to assign each metric a certain weight, based on the importance of that metric to the particular structure and setting of the call center.

  • slide 8 of 8

    References & Credits:

    In addition to the personal experience of the author, the following sources have been referred to, during the creation of this article.


    Image Credit:

    Image by – Sidharth Thakur