Small-business outsourcing research
How often are customer records kept past their scheduled disposal date?
A protocol for checking retention practice against the schedule, with tests for an expired record, an unexplained hold, and a disposal that left no trace.

Research question and scope
This protocol asks how often a customer record is still held after the disposal date in the business's retention schedule. The unit is one record set, identified by the customer, the record type, and the disposal date. The study measures whether the record is still active, whether a documented hold explains the delay, and whether a disposal was recorded when the record was removed. It does not evaluate whether the retention schedule itself is legally correct.
Set the scope to one small business, one customer record system, and one retention schedule. Record the record types covered and the people who may approve a hold. If the schedule changes during the window, keep the earlier definition for the full sample or analyze the change as a separate period. A new schedule can look like better compliance when it is only a new document.
Population and unit of analysis
Define the population as every record set whose disposal date fell before the sample date, including sets that were disposed on time and sets that are still present. The unit is the record set, not the customer, because one customer can have several record types. A record set that is split across systems remains one unit with a note about each location. A record set that is merged during the window keeps its original identity and the merge date.
Record the record type, the source system, and whether the record contains personal or financial information. These strata matter because a marketing list and a payment record carry different risks. Include records created outside the normal intake process, because the base rate is part of the finding. Test the unit definition on two ordinary records before collection begins.
A customer can appear in a support system and in an accounting system with different disposal dates. Record both locations and the date for each, and link them as one customer with two record sets. Do not merge them into a single unit, because one may be disposed while the other is still held. The link should show the relationship without erasing either schedule.
Definitions that make the measure reproducible
Define the disposal date as the date recorded in the retention schedule for that record type and creation date. Define an active record as one that is still readable by the team, even if it is archived. Define a documented hold as a written instruction with a reason, an owner, and an end date. Define an expired record as one whose disposal date passed without a disposal record or a valid hold.
Do not count a record that is merely hidden from the main view as disposed. Do not count a verbal instruction as a hold. A record that was disposed and later restored should keep both dates. Each definition should include a positive example and a boundary example so a second reviewer can apply it consistently.
Evidence standard and data dictionary
For each record set, capture the customer identifier, the record type, the system, the creation date, and the disposal date. Add whether the record is still present, the last access date, and the name of any hold. Record the disposal date and the person who approved it. Keep the evidence with the retention register rather than in a separate spreadsheet.
Also capture whether the record was included in a backup that has its own retention period. A record removed from the live system can survive in a backup, which changes the real disposal date. Record the backup system and its schedule. If the record was exported to a report, note the export and its retention rule. The evidence should show every place the record can still be read.
If the business uses a deletion tool, keep its log with the record set. A log that shows a job ran is not proof that the specific record was removed. Record the count before and after the job and the person who checked the result. The evidence should connect the action to the record, not just to the system.
Sampling and collection
Draw the sample from the retention register rather than from a list of records the team remembers. Record the export date, the filters, the row count, and the selection rule. Include every record type and system so the result is not dominated by the largest source. If the register is ordered by disposal date, a fixed interval spreads the sample across the window.
Collect the record and the schedule as they existed on the sample date. If a disposal happened during collection, record the date and analyze the state before it separately. Do not infer a disposal from a record that is no longer in the main view. Note the collection date separately from the disposal date.
Record the date the schedule was last reviewed and the person who approved it. An expired record may be measured against a schedule that is itself out of date. Use the version in force on the sample date rather than a newer draft. If the schedule is missing, mark the record set as unscored instead of counting it as both expired and covered.
Analysis and reporting
Report the number of record sets, the share disposed on time, the share held under a documented hold, and the share expired without either. Show the counts by record type, system, and whether personal information is present. Present the median days past the disposal date for the expired sets. Use the same denominator for every rate.
Inspect the expired sets and describe the pattern without naming the customer or the employee. A cluster in one system may point to a missing deletion job, while a cluster in one record type may point to an unclear schedule. Compare the result with the last schedule review. Report records that survive only in backups separately.
A useful additional view is the share of holds that have passed their own end date. That view separates a retention problem from a hold-management problem. Show it beside the expired rate and describe the difference. The owner can then decide whether to fix the schedule, the deletion job, or the hold review.
Report the result by record type and system so a reader can see whether the problem follows one schedule or one tool. Add the number of record sets that exist only in a backup, because they change the real disposal picture. State the sample date and the schedule version at the top of the table. A reader should be able to tell whether the finding is current.
Limitations and sources of error
The record cannot show whether a disposal was legally required or whether a hold was justified. It can show only what the schedule and the register recorded. A record held past its date may be protected by a litigation hold that is not in the system. The protocol also cannot capture a copy saved on a personal device or in an unmanaged inbox.
Systems can mislabel a creation date, and a merged customer can inherit the wrong schedule. A record that appears expired may belong to a newer record set. Record those cases and their treatment instead of dropping them silently. If the same person sets the schedule and approves the holds, note the independence limitation and analyze that subgroup separately.
Review cadence and next action
Give the result to the owner and the person who manages the retention register and ask them to choose one observable change. The change might be a monthly deletion job, a hold end-date review, or a required disposal note. Record the owner, the effective date, and the measure that will show whether the change worked. Avoid changing the schedule and the deletion tool at the same time.
Repeat the protocol with the same definitions after the change and compare the same fields. If expired records fell but holds rose, the team may be holding more rather than disposing better. If holds fell but expired records did not, the deletion job may be failing. Close the review when the owner accepts the finding and the next sample is scheduled.
Sources
- IRS, recordkeeping for small businesses, Internal Revenue Service; checked 2026-11-07
- Federal Trade Commission, business guidance, Federal Trade Commission; checked 2026-11-07
- U.S. Small Business Administration, manage your business, U.S. Small Business Administration; checked 2026-11-07
- U.S. Government Accountability Office, internal control standards, U.S. Government Accountability Office; checked 2026-11-07
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
How long do supplier bank-detail changes wait for independent verification?
A protocol for measuring the delay between a supplier bank-detail change and the independent check that confirms it.
What share of damaged-order claims include a usable photo and order reference?
A protocol for measuring claim evidence quality, with clear tests for a usable photo, a linked order, and a recorded delivery condition.
How often do support macros cite a policy version that is no longer current?
A protocol for auditing saved replies against the current policy register, with a test for stale wording, missing versions, and untested macros.