Retention Schedule Template

This is a structure rather than a filled-in schedule, because the periods that apply to you depend on your jurisdiction and sector and cannot be supplied generically. What can be supplied is the shape that makes a schedule implementable, and the columns most drafts are missing.

The six columns#

Most published schedules contain the first two and the fourth. The third and the sixth are what make the difference between a document and a control.

Worked example rows#

These illustrate the structure. The periods are examples of how rows are written, not advice for your jurisdiction — every one of them needs local verification before use.

Note that the trigger differs in every row and that none of them is the creation date. That is the normal case rather than the exception, which is why a schedule without a trigger column produces wrong dates almost everywhere.

Filling in the trigger column#

Six triggers cover the great majority of record classes.

Creation for records with no lifecycle — correspondence, reports, published material. Closure or completion for matters, projects, cases and investigations. Termination or expiry for contracts, licences and agreements. End of relationship for customer and supplier records, which is later than the last transaction. End of employment for personnel records. And date of last exposure or the subject reaching the age of majority for the long-period classes in occupational health, education and safeguarding.

If a row does not fit one of these, it is worth checking whether the record class is defined clearly enough to have a rule applied to it.

Choosing the disposition action#

Default to review rather than destroy. A review costs minutes and produces a named person who authorised the disposal; automated destruction produces a system action nobody can stand behind.

Reserve automatic destruction for high-volume, low-risk classes where the period is uncontroversial, and treat it as a deliberate exception recorded in the schedule rather than the default setting.

Transfer to archive applies to the classes with permanent or very long retention, where the question is where they live rather than whether they are kept.

What has to sit alongside the schedule#

Three things, none of which belongs in the table itself.

A legal hold procedure, because a hold overrides every row here and must do so automatically. A disposal record format, capturing which records, under which row, authorised by whom, on what date, with confirmation that no hold applied. And a review cycle for the schedule itself, with a date and an owner.

The last is the one that fails quietly. A schedule approved once and never revisited becomes wrong as legislation and business practice change, and an out-of-date schedule applied confidently is worse than an acknowledged gap.

The completeness measure#

Track the number of documents with no retention class at all.

That figure, rather than the proportion correctly classified, is the honest measure of whether the schedule is working — and it identifies the material that will otherwise accumulate indefinitely because nobody can say what it is.

Before you circulate it#

Two sanity checks catch most of the problems that would otherwise surface at an audit.

Read down the trigger column and confirm no cell is blank or says "creation" by default. Then pick five rows at random and ask whether the system could identify a document belonging to that class without a person interpreting it. Any row that fails the second test is a class that will never have the rule applied, however carefully the period was researched.

Finally, check that the schedule and the document control procedure agree about where retention periods are maintained. Two copies of the same periods in two documents diverge, and the divergence is discovered by whoever relies on the wrong one.