As the manager of a customer service call center, you are evaluating the quality of the call center's operations. One of the most important metrics in a call center is time in queue (TiQ), the time a customer?waits before a customer service representative (CSR) comes on the line to help. To provide the best customer experience, you want your customers to wait to be less than the 2.5-minute (150 seconds) industry standard. You know that when they wait for too long, customers are more likely to have a negative experience or hang up before being helped.

Another metric you measure in the call center is the handling time, or service time (ST), the amount of time a CSR spends servicing the customer. Last month's average ST was approximately 3.5 minutes (210 seconds). Your industry experience tells you that the average ST can be influenced by a CSR's level of training, amount of experience, and whether the CSR can provide a resolution to the issue quickly. Recently, you tested a strategy to have callers identify the type of issue they are calling about and then route calls to CSRs with expertise in that issue.

The new protocol (PE) is being tested side-by-side with the current protocol (PT) to see if ST is improved with this strategy.?The regional director was hesitant to test the PE protocol for two weeks, already concerned about the customer experience, but agreed to the test. After the first few days, the regional director is anxious to know the results and asks you to send a report. You ask the user experience (UX) team to pull a report for the TiQ and ST for both protocols to analyze.

Preparation

Review thecall time analysisprovided by the UX team.

Assessment Deliverable

need to Address the following in a email:

• Analyze the impact the new protocol (PE) has had on time in the queue and service time.
• Determine if the PE protocol should be implemented widely in the call center with what you know so far.
• Identify what additional data and analyses would be helpful to determine if the PE protocol is working.
• Explain what is likely to happen to TiQ and ST if the PT protocol is kept.
• Explain how a sudden increase of 20% more calls might influence TiQ and ST.
• Justify whether the data is sufficient to determine whether the PE test succeeds.
• Suggest additional metrics and supporting data needed to determine the performance of the call center's operations.

University of Phoenix Call Time Analysis Time in Queue Test The team performed a test of hypothesis to determine whether the Rejection Region average TiQ is lower than the industry standard of 2.5 minutes (150 seconds). -1.44 A significance level a=0.05 was used. Normal distribution This was a test of mean against a hypothesized value of 150 seconds. 9000 Because the sample size was large, we assumed knowledge of the population's variance. Figure 1: TiQ Rejection Region The null and alternate hypotheses is: The test statistic is given by: Ho : p 2 150 x - Ho_147.9-150 -2.1 Zcalc 138.0 3.37 = -0.62 H1 : u u2 This is a right-tailed test. Because of the very large samples, there 0.62 z is no real difference between finding the critical value with a normal 2689 p-value (one-tailed, lower) distribution or the t distribution. The critical value with a significance level a=0.05 is t = 1.645. 141.313 confidence interval 95.% lower 154.532 confidence interval 95.% upper The decision rule becomes: Reject Ho if teak > 1.645. 6.609 margin of error Figure 2: Results of TiQ Hypothesis Test: Mean versus Hypothesized Value Average Service Time Test The team performed a test of hypothesis to determine whether the service time (ST) with new service protocol PE is lower than with the current PT protocol. A significance level of a=0.05 was used. This is a test of means for two independent samples with unknown variances assumed unequal.Hypothesis Test: Independent Groups (t-test, unequal variance) PE Protocol PT Protocol 149.28 212.16 mean 185.89 190.58 std. dev. 853 821 n 1665 df -62.883 difference (PE Protocol - PT Protocol) 9.206 standard error of difference 0 hypothesized difference -6.831 t 5.91E-12 p-value (one-tailed, lower) -80.940 confidence interval 95.% lower -44.826 confidence interval 95.% upper 18.057 margin of error Figure 3: Hypothesis Test of Independent Groups (t-test, unequal variance) The tcalc = -6.8 falls in the rejection region. We conclude that the new protocol (PE) results in a shorter average service time than the traditional protocol (PT) based on available data. End of document