The Conundrum of Events-Based Retention Periods

Is Schroedinger’s cat alive or dead? The answer in quantum mechanics is that both are true.  Is a record subject to retention or disposition? Again, both can be true at once and no more so than when records are existing within event-based periods. It is clear when a record has a defined time-based retention period such as 5, 10, or 20 years.  It becomes more difficult when a disposition is based on an event, such as termination of a contract, termination of employment, end of a patent, etc.  How can an organization manage records when there is uncertainty as to the legal and business needs for disposition?

There are obvious statutory requirements for retaining employee information such as right-to-work documents, pay statements, and medical and other benefits.  Similarly, organizations in certain industries may have reporting requirements according to local, state, federal, or international requirements. These generally prescribe specific retention periods.  However, there may be information owners who leave a position or a company, leaving the records in limbo as no one will have an idea if the key event has occurred, triggering the start of the retention period.

Although it may seem counterintuitive, for some organizations the best approach may be to disregard the event factor in the disposition of a record.  If your evaluation of your company’s records indicates that contracts are on an average of 10 years or your employees have an average of seven years of service you could base your retention periods around those time periods.  Of course, that would be accepting some percentage of over-retention or under-retention.

However, it would make retention and disposition simpler because records managers don’t have to rely on people throughout the company to be aware of their applicable event triggers and implement them efficiently. What if the person responsible for a given event trigger leaves the company or changes departments or jobs?

That institutional knowledge might not be taken up by others in their department or the person who replaces them.

This approach seems to imply having to have a certain risk tolerance. But, especially for companies with limited resources for information management, a good policy consistently applied could offer less risk than having the best policy applied inconsistently.  

Of course, there are many providers of records management software solutions that offer event-based disposition.  However, this involves integrating HR systems, contract management systems, intellectual property, and real estate records, among many other record categories with the vendor’s software. It is possible, but not all organizations will have the resources or budget to fully integrate their internal systems with those of a vendor.

Another possibility is the use of AI or machine learning to determine when a disposition event has occurred. For example, knowing when an employee has been terminated—which would be an event that impacts benefits, pensions, or other legal rights. This has become a very popular concept that would have real benefits but with little evidence of effectiveness. Again, this would involve the integration of multiple systems across different departments.

Straight time-based retention periods are much more easily realized by RIM vendors who rely on their customers to provide the event dates of their records. Event-based records require continued input and knowledge of the subject matter that the records manager may not have available.

Records managers should always work with their constituents to remove ambiguity because this achieves better compliance. Having a retention schedule that minimizes event-based periods can be the best option for organizations that would benefit from a more streamlined and clearer if less precise, company policy.  Schroedinger’s cat would agree.