BUSI 3014 Week 10 One-Page Decision Memo Example

Memorandum

To: Owner, Kestrel Pest Solutions

From: Dispatcher and Operations Analyst

Date: April 24, 2028

Re: Recommendation: Saturday route trial, May 6 to September 30, 2028

Recommendation

Run a Saturday route from May 6 through September 30, 2028, as a trial, staffed by one technician on overtime and reserved for first visits to new residential customers.

Evidence

In April through September 2027, 270 of 1,226 new home requests, or 22.0 percent, went past the four-day mark before a technician arrived. In October through March, the share was 6.0 percent, 43 of 714. The long waits cluster in the very season the trial would cover. This spring looks the same: on April 24, 41 of our 148 open requests were past the four-day mark, although the dashboard's average wait shows green because it leaves out customers who have not yet been served.

Cost

A Saturday route costs about $405 a day: eight hours of a technician's overtime at about $33 an hour, plus fuel, truck wear and materials of about $140. Over 22 Saturdays, the trial costs about $8,900. At nine first visits a Saturday, the route could serve about 198 new customers who would otherwise wait into the following week.

What Would Reverse This Recommendation

Stop the trial after June 10 if either of the following holds over the first six Saturdays:

1. The Saturday route averages fewer than six completed first visits, which would mean there is not enough waiting demand to fill it.

2. The share of new requests waiting more than four business days in May has not fallen below 12 percent, which would mean the route is not reaching the customers who wait.

A Limit

Our records cannot yet show that waiting is why customers leave. The trial will record cancellations by wait time so that next spring's decision can test it, using a full season of our own data rather than the owner's impression.

Why Saturday, and Why Only New Customers

A Saturday route adds capacity without pulling technicians off weekday routes that existing customers expect. Reserving it for first visits targets the customers who wait longest, since existing customers' quarterly treatments are scheduled weeks ahead. If spare Saturday slots remain after first visits are booked, the dispatcher may fill them with callbacks, but first visits always take priority.

How the Trial Will Be Reported

Each Monday, the dispatcher will report three figures: Saturday first visits completed, the share of new requests waiting more than four business days in the prior week, and cancellations before a first visit. The owner will see all three on one page alongside the current dashboard, starting the Monday after the first Saturday route.

Basis

The measure, its definition and the data cleaning behind it are documented in the analysis files and are summarized here only as the single figure that decided the question. Provost and Fawcett (2013) describe data-driven decisions as ones grounded in analysis rather than intuition alone; this one rests on a year of our own records. The figures use waits adjusted for late logging, a correction the data audit justified because late logging was concentrated in the busy months (Wang & Strong, 1996). The measure should be used to judge the route, not individual employees, since tying it to someone's evaluation would invite gaming (Ridgway, 1956).

References

Provost, F., & Fawcett, T. (2013). Data science and its relationship to big data and data-driven decision making. Big Data, 1(1), 51-59. https://doi.org/10.1089/big.2013.1508

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

Wang, R. Y., & Strong, D. M. (1996). Beyond accuracy: What data quality means to data consumers. Journal of Management Information Systems, 12(4), 5-33. https://doi.org/10.1080/07421222.1996.11518099