Event-Based vs. Time-Based Retention

Every retention rule answers two questions: how long a record must be kept, and from what moment the clock starts running. The first question gets most of the attention, but the second is where retention programs most often succeed or fail. A retention period of “seven years” means nothing until you specify seven years from what. That starting point — the trigger — is what distinguishes the two fundamental models of retention timing: time-based and event-based.

Both models are legitimate, both appear throughout authoritative retention schedules, and most mature programs use a blend of the two. Understanding the mechanics, strengths, and failure modes of each is essential to building schedules that are both legally defensible and operationally workable.

Time-Based Retention

Time-based retention (sometimes called fixed-period or calendar-based retention) starts the retention clock from a predictable, self-contained date that is known the moment the record is created or filed. The trigger is the passage of time itself. Common anchors include the date a document was created, the date a transaction closed, or — very frequently in public-sector schedules — the end of the fiscal or calendar year in which the record was made.

The defining characteristic is that nothing external needs to happen for disposition to become due. If a policy says “destroy three years after the end of the fiscal year of creation,” the disposition date can be calculated automatically and immutably at filing time. This makes time-based retention the workhorse of automated records systems:

• Predictability. Disposition dates are computable in advance, enabling batch review and bulk disposition.
• Auditability. Because the trigger is a date, it is easy to demonstrate consistent, schedule-driven application.
• Low maintenance. No one has to monitor for an outside occurrence; the calendar does the work.

The trade-off is rigidity. Time-based rules assume the record’s usefulness and obligations are adequately approximated by elapsed time. For many transactional and reference records — routine correspondence, financial backup, logs — that assumption holds well. For records whose obligations depend on a relationship or status that has no fixed end date, time-based timing alone is a poor fit.

Event-Based Retention

Event-based retention starts the retention clock only when a specified event occurs. The event is the trigger, and the fixed period runs after it. Classic patterns are written as “destroy X years after [event]”: after a contract terminates, after a case is closed, after an employee separates, after equipment is decommissioned, after a grant is closed out, or after a person reaches a certain age.

This model exists because, for many records, the obligation to retain is tied to a condition rather than to a date. A personnel file should not be measured from the day it was opened but from the day the employment relationship ends. A contract file’s value persists for the life of the agreement, however long that proves to be. Event-based retention lets the schedule track the real-world lifecycle of the matter the record documents.

The strength of event-based retention is accuracy: it keeps records exactly as long as the underlying obligation lasts, no more and no less. The weakness is operational. The system cannot compute a disposition date until it is told the event happened, which introduces a dependency on people and processes outside the records system. Records held “until” an event can sit indefinitely in a suspended state — often the largest, oldest, and riskiest accumulation in an organization — if no one ever signals that the trigger fired.

How the Two Models Combine

In practice, most schedule line items are hybrids: an event trigger followed by a fixed time period. The structure is cutoff plus retention. The cutoff is the event (or the close of a period) that ends the active accumulation; the retention period is the fixed span that runs from the cutoff to disposition. Federal General Records Schedules and agency-specific schedules routinely express obligations this way — for example, “cut off at end of fiscal year, destroy three years after cutoff,” which is time-based, or “destroy 6 years after the case is closed,” which is event-based.

Seen this way, time-based and event-based retention are not opposites so much as two settings of the same dial. A purely time-based rule uses a calendar event (year-end) as its cutoff; a purely event-based rule uses a business occurrence. The discipline of separating cutoff from retention period — recommended in records management standards such as ISO 15489 — keeps schedules unambiguous and machine-applicable.

Implementation and Risk Considerations

The hard part of event-based retention is capturing the trigger. A schedule can be flawless on paper and fail completely if no business process notifies the recordkeeping system that the contract ended or the case closed. Effective programs address this by integrating retention triggers into the operational systems where events already occur — case management, HR, contract, or ERP systems — so that closing a matter there automatically signals the records system, rather than relying on someone to remember.

Several risks deserve attention:

• Orphaned holds. Event-based records whose trigger never fires accumulate indefinitely, defeating the purpose of a retention schedule and increasing discovery and storage exposure.
• Legal holds override everything. Whether the timing is event- or time-based, an active litigation hold suspends disposition until the hold is released. Triggers must be reconciled with hold status before anything is destroyed.
• Trigger data integrity. Event dates must be recorded accurately and consistently; a mis-keyed termination date silently produces a wrong disposition date.
• Auditable disposition. Both models require documented, consistent application of the approved schedule so that disposition can be defended as routine and good-faith.

It is worth noting that recordkeeping system requirements continue to evolve. NARA withdrew its endorsement of the DoD 5015.2 certification regime in 2022, shifting toward the Universal Electronic Records Management Requirements and the broader Federal Electronic Records Modernization Initiative (FERMI). Those requirements still expect systems to apply approved retention schedules with both time- and event-based triggers — the modernization is in how requirements are expressed and procured, not in the underlying retention logic.

Choosing between the models is ultimately a matter of matching the trigger to the obligation. When the duty to keep a record is well-approximated by elapsed time, use a time-based rule for its automation and predictability. When the duty depends on a relationship, status, or matter that ends on its own schedule, use an event-based rule — and invest equally in the process that reliably captures the event. For a fuller treatment of how these mechanics fit within retention rules and disposition workflows, see the retention and disposition topic hub.