In-place records management versus in-place records management..?

A new research paper has clarified the three types of records management model, and they are (to paraphrase):

  • centralised
  • in-place traditional
  • and in-place modern (what we call the Data Castle). 

Rival records management models in an era of partial automation” by Lappin, J., Jackson, T., Matthews, G. et al., published in Archival Science in January, Expounds on the history and heritage of these three ‘rival’ models. Much is familiar in the discussion of centralised versus in-place, but what is perhaps novel is the differentiation of two types of in-place management. 

“Two rival records management models emerged during the 1990s. Duranti’s model involved moving records out of business applications into a repository which has a structure/schema optimised for recordkeeping. Bearman’s model involved intervening in business applications to ensure that their functionality and structure/schema are optimised for record keeping.”

As we have seen, the centralised model has more and more fallen by the wayside for most records of business. It cannot handle structured records, effective collaboration, or support multi-team business processes effectively. It has a high user impact, and tends to have high cost. The ‘traditional’ in-place model emerged as an alternative, but required those more collaborative, process-driven line-of-business systems to be customised to support as much compliance as the centralised model. The cascading risks of system customisation made this infeasible for the most part, and the rise of cloud solutions effectively obsolesced it before it ever really made a mark. 

“A third records management model has therefore emerged, a model in which records are managed in place within business applications even where those applications have a sub-optimal structure/schema.”

As a result, the new model, arrived at by default rather than design, has been to just allow records to be managed in their line-of-business systems, even where those systems are not compliant. Lappin et al. note that in this model it is easy to automatically apply a bulk classification to a whole system, site, library or mailbox – but not very accurate. Conversely, it’s accurate for records managers to manually assign classifications, but not at all efficient. As such, there is an imbalance:

“The imbalance will disappear if and when the capability is acquired to automatically and reliably assign electronic correspondence to a structure/schema that is designed with recordkeeping in mind.”

We now have this capability. With ‘modern’ in-place records management like Castlepoint, we can get efficiencies of automated classification, and all the efficiencies of the line-of-business systems for users. 

There has been an emergence since Castlepoint was launched of more of this type of model, with companies looking for the balance that allows user freedom with full compliance. Most of these systems do still require manual classification in the form of rules engines and file plans, which does not quite achieve the balance – but the ‘modern’ in-place ‘Data Castle’ model is here to stay, and  we will hopefully see more and more of it.