A decision gets reopened three years after it was settled, and nobody can explain why the original answer was no. A supplier dispute turns on a variation agreed in meeting minutes somewhere, on a drive that has changed owner twice. Both create avoidable work: the organization pays again to establish what it already knew.
That forgetting has a name. Institutional memory is what an organization knows about itself, and it is more fragile than it looks, because it fails in two different ways.
What is institutional memory?
| Institutional memory is the accumulated understanding of why an organization works the way it does: the reasoning, context, and precedent behind its decisions. It is held largely in the experience of the people who were there. Records may form part of it, but the term usually describes knowledge that was never formally written down. |
That qualification matters. The Society of American Archivists, in its Dictionary of Archives Terminology, emphasises the informal layer: the reasoning behind a procedure, rather than the procedure itself. It does not exclude records, which carry much of the context that makes the reasoning recoverable.
Institutional knowledge is everything your organization knows, including what is set down in policy and systems. Institutional memory leans on the part carried personally, which researchers call tacit knowledge.
Why does institutional memory disappear?
Institutional memory disappears by two routes. People leave and take the context with them. Separately, the records that would have carried that context degrade through poor description and obsolete formats, whether or not anyone has left.
| How it fails | What it looks like | What addresses it | |
| Knowledge held by people | Leaves with retirement, resignation, and restructuring | Nobody can explain why a decision was made | Decision records, structured handover, knowledge transfer |
| The records that evidence it | Decays through poor description and format obsolescence | The file exists but cannot be found or opened | Appraisal, metadata, retention schedules, digital preservation |
The second route is the one that gets missed, because nothing announces it. A record nobody described is not lost; it is unfindable, and in a format that stops opening makes its contents unusable without further work. Either can go unnoticed until the material is needed, during a dispute or an audit.
How do you protect organizational memory in practice?
Five activities do most of the work: appraisal, decision records, description, retention and format planning. They overlap rather than run in sequence.
- Decide what is worth keeping: Archivists call this appraisal: the judgement about which records carry lasting value. Without it, retention gets decided by storage limits or by whoever is clearing a drive.
- Capture the reasoning, not just the outcome: A usable decision record states what was chosen, what was rejected, and which assumption the choice rested on.
- Describe what you keep: Description is what makes a record findable and intelligible later: who created it, when, for which decision, and what it supersedes. The origin and custody part of that is what archivists mean by provenance.
- Set a retention schedule and follow it: A schedule states how long each class of record is held and what happens at the end of that period. This is ordinary records management practice, reflected in ISO 15489. Whether the effort repays itself can be estimated rather than argued: calculating knowledge management ROI puts a figure on time lost searching for or recreating information.
- Plan for future format obsolescence, not only storage: Digital preservation is the separate discipline that keeps files readable, and their context intelligible, as software moves on. A backup restores a recent version. Digital preservation is about still being able to use and understand that file in fifteen years.
Is your organizational memory at risk?
Two patterns signal weakness: knowledge only one person holds, and records nobody can locate or open. These questions test for both.
- Can anyone explain the reasoning behind a decision made three years ago, without asking the person who made it?
- Does a named individual own each of your main shared drives?
- Do you have a retention schedule, and does anybody follow it?
- Could you still access and understand a file created in 2012?
- Does your departure process capture why someone did things a particular way, or only what they did?
Weak answers to the first and last questions point to knowledge transfer. The middle three point to records stewardship. The two need different remedies.
Frequently asked questions
Is institutional memory the same as institutional knowledge?
Not quite; institutional knowledge is everything an organization knows, including what is documented in policies, records, and systems. Institutional memory leans on the personally held part: the context and precedent people rarely write down. The two overlap.
How do you transfer institutional memory when an employee leaves?
Through exit documentation that records reasoning rather than tasks, captured in conversation while the person is still there. Transfer only holds if what you capture is then described and stored somewhere authorised colleagues can retrieve it.
How long should an organization keep its records?
It depends on the record type and your sector’s legal requirements, which is what a retention schedule exists to set out. Finance, employment and contractual records usually carry statutory minimums. Material of lasting evidential value is often kept permanently.
Can AI tools preserve institutional memory on their own?
No, although they can help. AI can assist with capture, transcription, tagging, and drafting descriptions. What it does not take on is accountability: deciding what is worth keeping, confirming a description is correct, and keeping formats usable remain organizational responsibilities.
The organizations that keep their advantage
Institutional memory that depends on who is still employed is not an asset; it is an exposure. What reduces that exposure is unglamorous: deciding what is worth keeping, describing it well enough to find, and keeping it in a format that still opens. Do that, and the reasoning behind your decisions stays available to whoever needs it next.
