MSHR 6610 Week 9 Operational Metrics Note Example

What Each Measure Can Hide: A Metrics Note Testing a Utility Contact Center's Service Level, Occupancy and Attrition for Blind Spots and Proposing Interval, Threshold and Early-Tenure Versions

Student Name

Walden University

MSHR 6610: Aligning Human Resources with Business Operations

Instructor Name

Month Day, Year

What Each Measure Can Hide: A Metrics Note Testing a Utility Contact Center's Service Level, Occupancy and Attrition for Blind Spots and Proposing Interval, Threshold and Early-Tenure Versions

Why Test the Measures

Last month, Northfork Power & Light's contact center reported a daily service level of 81 percent of calls answered within thirty seconds, just above its 80 percent target. On the same days, evening callers between 6:00 and 8:00 p.m. waited an average of more than two minutes, and fewer than half were answered within thirty seconds. Strong mornings had averaged away weak evenings. Ridgway (1956) warned that performance measures can produce dysfunctional results when people manage to the measure rather than to the outcome it was meant to track; a measure that hides failing intervals invites exactly that. This note takes the three measures the center reports to its operations director and asks of each what it exists to catch, what it can hide and what version would catch the hidden problem.

The Three Measures

Table 1

Measures, What They Catch, What They Hide and Proposed Versions

Composite utility. Thresholds are proposals for discussion with the operations director.

Service Level

The daily average treats a call at 9:00 a.m. and a call at 7:00 p.m. as interchangeable, so a surplus in one interval offsets a deficit in another. The proposed version counts intervals instead of calls: of the day's 26 half-hour intervals, how many met the target? Reported by day part, morning, afternoon and evening, it shows at once whether evenings are failing. Gans et al. (2003) note that service targets are usually defined for each interval in staffing models, so measuring the same way makes reporting consistent with planning.

Occupancy

High occupancy looks efficient: agents are busy almost all the time. But sustained occupancy above about 85 to 90 percent leaves agents no recovery time between calls and is associated with stress and turnover in call center work. The average for the day can sit at a comfortable 82 percent while evening intervals run at 93 percent. The proposed version counts intervals above an 88 percent ceiling, making overload visible where it occurs. The ceiling is a judgment, not a research finding, and should be agreed with supervisors.

Attrition

The annual rate gives one number for the whole center. It hides two patterns that matter for operations. First, many departures happen early, within the first ninety days, often before or during certification, which wastes certification seats the constraint brief showed are scarce. Second, attrition is far higher on evening shifts. The proposed version reports ninety-day and one-year attrition separately and by shift, every quarter. Hom et al. (2017) observe that the causes of early and later turnover often differ, so separating them points to different remedies.

Reporting Together

The three proposed measures should be read together. Evening intervals failing the service target, evening occupancy above the ceiling and rising evening attrition are three views of the same problem: too few agents on evenings for the work that arrives. Reported side by side, they let the operations director see the connection without HR having to argue it. Behn (2003) argues that measures should be chosen for the purpose they serve; here the purpose is to show where evenings fail before the failure becomes turnover.

How the Change Would Be Introduced

The new versions can be produced from data the center already collects. The workforce system records calls and service level for every half-hour interval, and HR's records already show hire dates, shifts and termination dates. The first month's report should show the old and new measures side by side, so the operations director can see where they agree and where the daily average was hiding failing evenings. After a quarter, the old daily figures can move to an appendix. Supervisors should see the interval measures for their own teams' hours, so that they are judged on the same view of performance the director uses.

What the Note Does Not Propose

The note does not propose new targets. It changes how existing targets are reported, so that the center's current standard, 80 percent within thirty seconds, is measured where customers experience it. Changing the target itself would be a separate decision for the operations director.

References

Behn, R. D. (2003). Why measure performance? Different purposes require different measures. Public Administration Review, 63(5), 586-606. https://doi.org/10.1111/1540-6210.00322

Gans, N., Koole, G., & Mandelbaum, A. (2003). Telephone call centers: Tutorial, review, and research prospects. Manufacturing & Service Operations Management, 5(2), 79-141. https://doi.org/10.1287/msom.5.2.79.16071

Hom, P. W., Lee, T. W., Shaw, J. D., & Hausknecht, J. P. (2017). One hundred years of employee turnover theory and research. Journal of Applied Psychology, 102(3), 530-545. https://doi.org/10.1037/apl0000103

Ridgway, V. F. (1956). Dysfunctional consequences of performance measurements. Administrative Science Quarterly, 1(2), 240-247. https://doi.org/10.2307/2390989