Good Practices for Document Control: Categorizing, Classifying & Tagging
Document Control - 10 February 2014
Continuing our blog series on Good Practices for Document Control in the Life Sciences, this post will focus on stage 1 “Categorizing, Classifying and Tagging”.
Mastering this stage involves having the right skillset and processes in place to consistently produce documents with not only the right metadata (attributes and/or tags), but more importantly, the right assignment of workflows and templates, as well as to facilitate document retrieval through a variety of ways (Reports, Views, Search, etc.). It may also take into account archiving requirements (if any).
Definition: In a document management system (DMS), when referring to a Standard Operating Procedure (SOP), for instance, the word Document (or Content) encompasses the ‘soft copy’ file (Word, etc.), its rendering in PDF (if any, and the e-signature manifestations), as well as its meta-data and workflow information.
Since we promised to focus on people and processes in our SECC (* see below, and ‘SOLABS Electronic Content Control’) model rather than on the document life cycle, we will first review the basic concepts involved, and then move on to people and processes.
Categorizing involves creating a system for documents to be created with the appropriate metadata. You generally start by selecting the correct document type. This will provide you with the predefined attributes for the metadata, workflows, etc. We should note here that, as with most things, the need to categorize generally increases as volume increases. No one needs an inventory management system to categorize bicycles owned by a family of four! Shoes… well maybe…
Classifying involves the configuration done on a per-document basis to store documents. By defining a set of rules and system security parameters, you are able to control document access appropriately.
Tagging (for us anyhow) (for us, anyhow) involves assigning tags to a document in a non-rigid way. Tags serve the same purpose as metadata, but can generally be added by almost anyone accessing documents, even those users with limited privileges.
Now back to people and processes….
Initial Configuration & Periodic Review
The listing of required document types, along with their corresponding attributes, is generally produced at the time of implementation. At this time, the folder or library structure should be established to ensure that access rights can be handled easily as the volume of documents increase. You may also want to define standard document workflows per document type i.e. the people (use Job Titles, not names) generally responsible for reviewing, approving and retiring documents. Finally, the PDF rendering settings should be defined per document type, i.e. details on headers and footers, watermarks, and system-generated cover pages. You may also want to identify which documents require training.
As a reference, here at SOLABS, we conduct one-day configuration workshops to define the configuration for our Document and Training off-the-shelf sections. So fear not: this is nothing to be afraid of.
Once the initial configuration is in place, we recommend an annual review of the Document Types and their corresponding attributes (SECC-1). Review of the folder/library structure and access rights will be discussed in a future article.
We also recommend implementing an approval mechanism before adding new document types, where the system owner or business analysts are involved (SECC-2).
What Metadata or Attributes Do You Really Need?
Generally you should define a common set of attributes that are used from one document type to the next i.e.:
• Title, description, author, owner, effective date, control number, version, etc.
And add to this a series of non-mandatory attributes specific to your organization e.g.:
• Department (Owner), Manufacturing Site, Product Name, etc.
A few ideas that will help you keep your number of document types to the minimum:
• The meta-data provides additional information on documents, not the document type, i.e. do not create SOP Baltimore and SOP Mumbai assuming you have 2 sites. You should have only one SOP as document type with the attribute “Site”.
• Type is not created with the assignment of security in mind. For instance, this is a “blog”. This is not a “Blog to publish for only these selected individuals”.
Key Responsibilities to Maintain Classification and Categorization in Good Shape
We often refer to the post office as a good comparison for the processing of documents. Hubs are created to route packages and letters (documents) by a series of trained individuals (document coordinators). For the rest of us (authors/contributors), the only thing we need to provide are a few pieces of information, such as delivery address (attributes).
Keeping the post-office metaphor in mind we believe:
• Authors should be responsible for providing the document and its meta-data to Document coordinators;
• Document coordinators should verify the meta-data, and assign the right reviewers and approvers. They should classify the document correctly to ensure security is maintained;
• In terms of business processes, authors/reviewers/approvers should interact in similar ways with the system and have the same level of knowledge with regards to managing electronic documents.
• General users need only to have easy access to Approved and Effective documents. This can be achieved in different ways e.g. using Tablets. We will come back to this in a future article.
A few ideas that will help you with the classification and categorization in the long run:
• Review Cycles available in DMS or EQMS systems can be used to send document to Authors for the initial creation or for review of an existing document. This leaves a Task open in the system, and it’s easy to know who is responsible;
• The folder is dead… Create less, not more. Folders or libraries are mostly useful for managing access rights. That being said, we very much like the ‘tree view’ component [French: arborescence], which can be used to represent documents outside of the traditional folders/files hierarchy.
We have not spoken much about Tagging, since it is a public means of assigning attributes. The way we see it is there are certain fields open to more people to help in the categorization and retrieval of documents. Tags are ‘soft’ attributes that more people can assign at any stage of the document Life Cycle which serve the same purpose as document attributes (meta-data).
And finally, for those who would like to read further on the theme and subject matter of this article, you will find more information under the following search queries:
• Information architecture document management
• Content management and information architecture
• Enterprise content management best practices
Founder & CEO, SOLABS
* SECC defines items we feel should be activities mastered as part of our SECC discipline. We will review these and provide them in a checklist format once we have completed our exercise.