Small-business outsourcing research
How long do delivery route exceptions wait before the customer is told?
A protocol for measuring the time between a recorded route exception and the first customer notice, and for separating internal delay from an unreachable customer.

Research question and scope
This protocol asks how long a delivery route exception waits before the customer is told what changed. The unit is one exception entry, not one route and not one parcel. The study measures the share of customers notified before the promised window closed, the median wait to notice, and the exceptions that were never explained. It does not judge whether the driver was at fault.
Set the scope to one business, one delivery operation, and one notice rule. Record the promised-window policy, the notice template, and the date the rule took effect. If the rule changes during the window, keep the original rule for exceptions already open and analyze later exceptions separately. A new dispatch tool can change the record without changing the route.
Define the population as every route exception recorded during the window, including exceptions resolved on the same day and deliveries the customer refused. State any exclusion, such as a test route or a training delivery, and apply it to every entry. Test the definition on ten closed exceptions before collection begins. The reader should be able to rebuild the population from the dispatch log.
Definitions that make the measure reproducible
Define a customer notice as a message that names the stop, explains the change in plain language, and states the new expectation. An internal status change is not a notice even when it is visible to staff. Define the clock start as the time the exception was recorded in the dispatch system, not the time the driver first mentioned it. Record both times when the system supports it.
Define the clock stop as the send time of the first customer notice. Use the promised window in force when the order was placed, and record the time zone. Treat a returned parcel and a refused delivery as separate outcomes. A definition that ignores the promised window will make every notice look early.
Evidence standard and data dictionary
For each exception, capture the exception identifier, route, stop, customer, promised window, record time, reason code, notice time, channel, owner, remedy, and any carrier claim. Keep the customer's order identifier linked to the entry. Record whether the customer replied and what the reply asked for. A complete entry lets a reviewer follow the whole event.
Capture the reason for a delay when it is documented: no notice template, a wrong phone number, a dispatcher backlog, or a customer who could not be reached. Do not infer a reason from an empty field. Mark unknown reasons as unknown and report the count. A guessed reason hides the step that actually needs repair.
Sampling and collection
Collect every exception in the window rather than a sample, because the notice rule applies to each one. If the business runs several routes or depots, collect each separately and record the group. A combined rate can hide one route with a persistent notice problem. Report the number of exceptions and the value behind every percentage.
Draw times from the dispatch log and the message log, not from a spreadsheet maintained after the fact. If either log is incomplete, say so and reduce the claim. Record the export date, the filters, and any exception that could not be matched to a notice. A missing log is itself a finding about the operation.
Analysis and reporting
Report the share notified before the promised window closed, the share notified later, and the share never notified. Show the median and range of minutes from record time to notice. Present the results by route, depot, and reason code. Use the same denominator for every rate.
Separate an unreachable customer from an internal delay. A wrong number is a different finding from a dispatcher who never opened the queue. Report both because they lead to different fixes. A business may accept a customer delay but should not accept silence from its own team.
Limits of the evidence
The message log can show that a notice was sent, but it cannot show that the customer read or understood it. A send receipt is not comprehension. The protocol also cannot capture a phone call that was never logged. State that limitation before presenting the rate.
A small operation may have too few exceptions to support a stable trend. Describe the exception count and the route mix instead of ranking drivers. Do not compare a business with a two-hour window with one that promises a full day. Report the window policy beside the rate.
What to change
Give the result to the owner and the dispatcher and ask them to choose one observable change. The change might be a notice template opened at exception entry, a named notice owner, a phone check at order intake, or a daily review of open exceptions. Record the owner, the effective date, and the measure that will show whether the change worked. Repeat the protocol with the same definitions.
If the on-time rate rises but complaints also rise, the notices may be accurate but unhelpful. If complaints fall but the never-notified share rises, the team may be closing entries without contacting anyone. Report the movements together before declaring success. Close the review when the owner accepts the finding and the next collection date is set.
Sources
- Federal Trade Commission, mail, internet, or telephone order merchandise rule, Federal Trade Commission; checked 2026-11-18
- U.S. Small Business Administration, manage your business, U.S. Small Business Administration; checked 2026-11-18
- Federal Trade Commission, advertising and marketing guidance, Federal Trade Commission; checked 2026-11-18
- U.S. Government Accountability Office, internal control standards, U.S. Government Accountability Office; checked 2026-11-18
Frequently asked questions
Does a favorable sample prove the routine caused the result?
No. The design is observational and supports a bounded operating decision, not a universal causal claim.
What should an owner review first?
Start with high-consequence defects, missing denominators, exceptions, and cases that required the owner to reconstruct the evidence.
Related research
What share of waitlisted customers are offered a slot within two business days?
A protocol for measuring how many waitlisted customers receive a specific slot offer inside the two-business-day window the shop states.
What share of vendor rebate credits arrive inside the stated rebate term?
A protocol for measuring how many vendor rebate claims are credited inside the term, and for keeping short pays and denials visible.
How often are seasonal shift call-offs recorded before the shift starts?
A protocol for measuring how many seasonal shift call-offs reach the register before the scheduled start, without collecting medical detail.