Records and Information Management: Statistical Analysis

The only groups within our industry who seem to be realizing profit from records and information management best practices are vendors and professional organizations – not the companies who buy the products or send their subject matter experts to conference(s).


I’ve served many companies in multiple industries over the years. The general reticence to enhance records management project services into a more executive program seems to be entirely due to poor communication skills. Let me give you an example. Today, we calculate the success of an electronic records management software by qualification, declaration, and classification rates:

  • Qualification Rate: Are records owners correctly, appropriately, and consistently identifying those objects that are corporate records and should, therefore, be placed in the records repository? Expressed as the percentage of total objects per user, known to be business records and therefore stored in the repository. Seeking an accuracy of 90% or better.
  • Declaration Rate: Of the qualified objects (“qualified records”) that have been identified, are they in fact being declared? Expressed as the percentage of the known qualified records that are physically stored. Seeking an accuracy of 90% or better.
  • Classification Accuracy: Did the records that are stored have the correct retention rule applied? This is absolutely critical, as final records disposition and destruction cannot be performed unless the accuracy meets a specified threshold acceptable to records management. Expressed as the percentage of the total records that are known to have the proper retention rule applied. Seeking a n accuracy of 98% or better.

Do you see? A very process-based approach – which is why our records and information management best practices are process-based.

No, we need to improve our statistical analysis (and express the results correctly) to impress senior leadership, who will then consider the information management function in a more positive light, which will then increase our overall standing in the organization, therefore we can improve best practices, and provide the return on investment our companies seek.

For example, I’ve been working on an idea for a while – representing the queues for a file share cleanup project. This is the typical set of internal and external criteria that drive the “do we maintain the object, do we delete the object, or do we transfer the object?” decision-making. Each object on a file share will fall under one of the combinations:

This is the baseline graph only – the geography of the ellipses may expand or contract based on results of the study (and your results may vary).

But I’m exploring the idea that the above is easier for senior leadership to ingest. Thoughts?