Three Collections Do Not Have to Start Together

State what the four minutes contain

For this exercise, the four minutes include the entire assumed handoff from its start to the maker becoming available again. No order needs additional production, payment troubleshooting, parking assistance or a correction. Those are restrictive fictional assumptions, not observations about an actual craft business. If one changes, the calculated queue can change too.

Call the three orders A, B and C. The order of service is fixed as A, then B, then C, so the comparison does not depend on an argument about priority. Waiting means time between the declared arrival and the start of that customer’s handoff. Completion time is a separate number. A customer can have no queue wait and still spend four minutes being served.

Plan one: everyone arrives at ten

All three customers arrive at 10:00. During 10:00–10:04, the maker handles A while B and C wait. At 10:04, A is complete and B begins; C is the only person still waiting for service to start. At 10:08, B is complete and C begins. At 10:12, all three handoffs are complete.

The start waits are therefore zero minutes for A, four for B and eight for C. Add them to obtain twelve customer-minutes of queue waiting; dividing by three gives a four-minute average within this fictional group. None of that says a real customer would consider an eight-minute wait acceptable. The calculation identifies where the wait comes from rather than deciding the social response.

The maker has twelve consecutive minutes of handoffs and no assumed idle gap within that span. Advertising a broad window does not, under these arrivals, distribute the customers across it. All three arrived together because that is what this plan declares.

Plan two: two windows with assigned starts

Now offer a first fictional window from 10:00 to 10:08, with A assigned 10:00 and B assigned 10:04. Offer a second from 10:12 to 10:16, with C assigned 10:12. Under the same exact-arrival and four-minute assumptions, A finishes at 10:04, B at 10:08 and C at 10:16. Each begins immediately on arrival.

The maker still performs twelve minutes of handoffs. The calendar from the first start to the last completion now spans sixteen minutes, including the four-minute gap from 10:08 to 10:12. That gap is not a saving in work, and the example assigns no additional productive task to it. The distributed plan exchanges a longer calendar span for the absence of the queue wait in this declared scenario.

A late arrival, an early arrival or one handoff taking six minutes would require another schedule. Two window names alone cannot promise the calculated result. Here the assigned start times, exact arrivals and service duration together create it.

One arrival shifts: keep the original windows closed on time

For a separate fictional case, keep A at 10:00 and C at 10:12, but make B arrive at 10:06, two minutes after the assigned 10:04 start. Keep the A–B–C service order and four-minute handoffs. Now explicitly require each whole handoff to fit inside one of the same advertised windows, 10:00–10:08 or 10:12–10:16. Neither closing time is extended.

A still completes at 10:04. When B arrives at 10:06, only two minutes remain in the first window; a four-minute handoff cannot finish before its 10:08 close. B therefore starts at 10:12 in the second window and completes at 10:16. B’s queue wait is six minutes, measured from the declared 10:06 arrival to the 10:12 start.

C arrives at 10:12 but, under the fixed order, waits while B is served. At the 10:16 close, C has waited four minutes and has not begun a handoff. That is waiting observed by closing, not a completed waiting time or a valid 10:16–10:20 appointment. Only eight handoff minutes have been completed, and one order remains uncollected. The available windows still contain twelve open minutes, but the four-minute block lost earlier cannot be recovered inside them. This version therefore cannot complete all three collections under its stated constraints; another agreed arrangement would need a separately specified model.

Choose which structure deserves a real trial

For an invented maker who can reserve the full sixteen-minute span and particularly wants to avoid having customers stand in a queue, I would investigate the spread plan first. For someone who can offer only a twelve-minute continuous collection period, the longer schedule would not meet that preference. The arithmetic exposes this conflict instead of calling either structure universally efficient.

SBA’s planning guidance asks businesses to describe how customers interact with them. This small example develops one such interaction with declared numbers; it does not supply a sales forecast or a tested appointment system. Before buying scheduling software, establish whether it can express the actual structure you need and whether the underlying handoff assumptions are realistic. No software feature can make three simultaneous arrivals become sequential merely because a window has been named.