Overwhelmed by the sheer number of documents you have to deal with, it could help to have a clear and simple guide to help you decide what to keep, and what to heap. There is a clear and principled way to make the decision: The Rule of Record, Reference, Replica, and The Rest. This concept can make the life of a knowledge manager a whole lot easier.
First the quick guide and then some more explanation (and, at the end, the one key pitfall to avoid and the one major difficulty and how to deal with it):
| Record | Reference | Replica | The Rest |
|---|---|---|---|
| Original or official document. | Knowledge content document | “Convenience copy” – a duplicate or personal copy of a Record or Reference | Other ordinary documents without the special status of a Record or Reference |
| Referred to for the true facts of a matter such as a business agreement or decision, for example a contract. | Referred to for the most complete, up-to-date information and guidance on a subject; and as a source of reusable knowledge or content | Used as a temporary or personal use; not the official source | No official value; created for current use or as part of a project, process or communication |
| Retain unaltered for the time period required by regulations or your organizational guidelines – then dispose of. | Maintain for as long as useful; update regularly and continuously; dispose when obsolete. | Use as needed; discard after use; always refer to the original for ‘official’ or up-to-date information | Discard after use, which is at the “end of the matter”; prevent clutter by not retaining unnecessarily |
1. Record: The Official Document
Records hold official information that has to be correct and has to be retained as, well, a reliable record. Every department, from finance to HR, creates records. Most processes and projects create records. Client records, employee records, contracts and agreements, so on. Most departments know what their records are. It’s a record if you need to be able to refer to it for the true information about that client or employee matter, or that agreement, later on. But here’s the thing: not every record needs to be kept forever. It’s sensible to keep records for as long as there might be a need to refer back to the actual facts of the matter – but after that they just become more clutter. Like keeping a receipt for as long as the guarantee is still valid – but no longer. By identifying records, you can focus on what truly needs to be managed and retained, and for how long – and you can delete the rest if no longer required as a record.
The common errors here are over-classifying too many items as records; failure to be able to preserve records and find them when needed; retaining records long past their useful life. Classifying and actively managing records makes them more effective as the true touchstone of the facts of a matter when needed.
2. Reference: Your Knowledge Content
References are your go-to documents for knowledge. Unlike records, they should evolve and improve over time. They don’t need to be kept forever—just as long as they’re useful and current. Sometimes, old knowledge content documents might need to stick around to show historical knowledge, but generally, they’re dynamic and meant to be updated periodically and continuously.
For a knowledge manager like me, I find that one of the main motivations for getting involved in broader document and record management is precisely to make sure that the knowledge content is highlighted, and other items are managed down to make sure the knowledge stands out.
3. Replica: Handy Duplicates
Replicas are just that—convenient duplicates. These are copies made for ease of use, like taking a copy of a record on a trip. However, they’re not the official document itself, and they shouldn’t be treated as such. When you need the most up-to-date info, always check the current record or reference. Replicas are temporary and can be disposed of when no longer needed.
For example a project team has a reasonable need for having a copy of the client contract – but it’s their copy for their information and not the formal record itself, which is probably held by the Legal department, and tit doesn’t need to be kept by the project after use. The project team might also take copies of relevant knowledge references from related, previous projects so that knowledge can be reused and built on. But, again, these are copies and not the main knowledge content documents. New knowledge gained in the project should be added back into the shared knowledge, but the copies they took of previous knowledge documents don’t need to be kept after the project.
Common errors include retaining the copies of records and references taken with the project documentation forever, which adds to clutter and volume; and mistaking these replicas for the actual items – both during the project and afterwards.
Project deliverables (when they’re documents) are an interesting case that help make the point. They likely need to be retained both as Records and References. As Records, because they’re a critically important part of the contract and communication. And, as a Record, a deliverable should be retained for as long as all the project work is retained (as long as you think the facts of the matter might need to be referred to) unchanged as a reliable Record. As a Reference, the deliverable may need to be sanitised to preserve client and project confidentiality, but, in that form, may well constitute a useful knowledge Reference to be retained for as long as it is useful to refer to and reuse elements of.
4. The Rest: The Regular Stuff
Anything that doesn’t fall into the above categories is just “The Rest”. These documents don’t have the status of Records or References and should be discarded once they’re no longer useful. The same goes for Replicas which are not the Record nor Reference itself and should not be kept after their useful life is ended or the purpose for which they were taken is ended. These practices help keep your document pile manageable, ensuring important knowledge stands out.
Why This Classification Rocks for Knowledge Managers
- Spotlight on Knowledge Content: This system highlights the importance of references, or ‘knowledge content’.
- Clear Distinction: It separates the roles of Document and Information Management (IM) from Knowledge Management (KM). They often overlap, but this helps clarify their distinct roles. The pattern really belongs to IM, but makes clear what is the content that is of interest to KM.
- Declutter with Confidence: Knowing that ordinary documents can be disposed of prevents your valuable knowledge from getting lost in a sea of irrelevant files. It helps manage storage volumes and costs, which prevents drag on search, and, now the risk of AI processing old content.
The one key pitfall, the one major difficulty
What is the one key pitfall in this process? The key pitfall is the key confusion that I’ve consistently met and it’s this. It’s the far greater anxiety that people have about deleting or disposing of documents vs their concerns about retaining everything for ever. “What if I delete it and then find out I need it in years to come?” I often hear, along with “storage is cheap these days – why not keep everything in case we need it?”.
Well, what if you don’t need it in future? What if storage isn’t cheap? (it isn’t, btw).
We know that the psychology of loss is that it’s valued double the value of gain, and so this information anxiety is understandable. Deciding if an item or a whole collection of items constitute Records, References, Replicas or The Rest will help make a principled and reliable decision that addresses the “what if” fears at least to a reasonable level. And there’s one other thing to appreciate …
Keeping everything forever is being a hoarder living with all their trash. It really may not be easy to find that thing you unexpectedly find you need 15 years later if you kept 15 years of trash.
What is the one major confusion in this area? It’s confusing this with archives and backups. This isn’t about archives and backups. Archives are for taking a whole body of content out of the normal flow for historical research. Also, people use the term ‘archive’ to mean a separate or offline store of content prior to eventual deletion: It’s a reasonable and effective strategy for dealing with information anxiety: a six month archiving of content prior to deletion can ease that anxiety and prove or disprove whether a need for that content did or did not arise. But backups are full copies for restoring what was accidentally or incorrectly deleted or lost through some disaster. Backups also need to be managed, but they’re a somewhat different matter. Your own personal copy that you regard as a ‘backup’ is really a Replica.
What is the one major difficulty in this area? Nobody wants to spend time classifying or deleting content. That’s the problem. But if you don’t do something you end up a hoarder anyway by default with all the problems for retrieval of knowledge content and records that that presents. The answers are classification of content by design, management of their specific Records as part of the role of each function or department, knowledge management for the References/knowledge content, and automation of retention and deletion as far as possible. But start with those four: [1] classify content by design (for instance, have a place for each type and put the types in those places from the start and metatag content appropriately); [2] records management in each function; [3] knowledge management for the knowledge content; [4] automated retention and deletion ([1] will help with that because you’ll know what you have and where).
AI4KM - Robert Taylor 