Years ago, when I was still somewhat new to records management and I was struggling to form my own understanding of the information lifecycle, I was working as a consultant and technical sales rep for a small software vendor that created and sold an ‘enterprise records management application’. The bulk of our customer base were US Federal government agencies and I dutifully followed the government’s documented philosophy that some pieces of recorded information could be categorized as ‘records’, while some other pieces of recorded information could be categorized as ‘non-records’.

When I would demonstrate the email records management functionality of our application, I would often use an email invitation to lunch as an example of a ‘non-record’ that would not require preservation in our solution based on Federal records management regulations. Over time, the “Let’s Go to Lunch” email became something of an industry standard for understanding the concept of what type of Federal agency information constituted a ‘non-record’.

But as I began to learn more about the records management discipline, I became increasingly troubled with the notion of calling any recorded information a ‘non-record’. Labeling something a non-record implies that the information doesn’t have value and isn’t worthy of the rigorous business rules required of effective information lifecycle management. But all information is recorded for a reason and who’s to say that any single piece of information isn’t valuable at any given time?

Take the “Let’s Go to Lunch” email as an example. Sure, it sounds meaningless at first glance, but what if the invitation was sent by a male Corporate Vice President to one of his female subordinates? And what if it was the twentieth email he sent to her over the last few months asking her to go to lunch? And what if the subordinate eventually files a lawsuit claiming a long history of sexual harassment by the Vice President? Wouldn’t that ‘non-record’ email have enormous value to that woman and shouldn’t it be preserved for some commonly accepted period of time?

Legally speaking, whether you call a piece of recorded information a ‘record’ or a ‘non-record’ is meaningless because it is always evidence and it is always discoverable in a civil matter. But those people bent on destroying records management want you to believe that labeling information as a ‘non-record’ frees them of their responsibility to preserve and manage that information. This way they can protect themselves from prosecution when potential evidence of their misbehavior is ‘lost’ or ‘deleted’ by claiming it was a ‘non-record’, and so they did nothing wrong by not preserving it.  

____________________________________________ 

As I noted in a previous post, the Federal government defines a ‘record’ as “all recorded information.” This was a small modification to the definition made in 2014 that had profound ramifications for Federal agency records management.

How is it possible now to continue to separate Federal agency information into ‘records’ and ‘non-records’ based on this new definition? The truth is, you can’t. But sadly, at many agencies (and far too many organizations in the private sector), the practice persists. This policy has to stop if any organization is ever going to be able to effectively manage their recorded information.

But, perhaps surprisingly, this is actually good news for records managers. This is because managing the lifecycle of all of an organization’s recorded information is safer, more efficient, and less costly than dividing that information into ‘records’, to be managed, and ‘non-records’, to be ignored.

This is true for a long list of reasons. Here’s one. This is a decision tree used for many years at several Federal agencies that is intended to help an agency information worker determine if a particular piece of recorded information is a ‘record’ or a ‘non-record’ (redactions are mine, and made to prevent the identification of any single agency):

Amazingly, agency information workers were trained to pass every piece of recorded information they created or received through this complicated workflow, just to determine if the information was a ‘record’ or a ‘non-record’. This included emails, instant messages, text messages, unstructured content, even voicemail.

Think of the amount of information you generate at your job. How much time out of your day would it take you to run each of those pieces of information through this workflow? All that time would be wasted given that there is no real value in labeling a piece of information a ‘record’ or ‘non-record’. Eliminate that requirement and the information worker would have much more time to do the important work he or she is paid to do.

____________________________________________ 

Categorizing a piece of recorded information as a ‘record’ or a ‘non-record’ is one of the many relics of a paper-based records management program that has no real value in a digital information environment. Discontinuing the practice would significantly improve any organization’s information lifecycle management capabilities and should be standard policy everywhere.