Filing Crash Course: Classification Schemes

A thoughtful filing system can save a lot of time and tears when it comes to retention and disposition. However, designing and implementing such a system is no mean feat. Seasoned records management officers (RMOs) and newly-minted records custodians alike often struggle with this task. “How,” you may ask yourself, “is it possible to store so many records with different retention requirements in a way that makes timely disposition actually achievable?”

The answer is to create a classification scheme, which is a fancy way of saying a plan for naming and organizing records. An effective classification scheme helps staff across your organization use consistent terminology, identify who is responsible for records, and locate information. Classification schemes are as unique as the entities that use them, so we can’t recommend a scheme that works perfectly for everyone. The good news is that creating a classification scheme is something any records manager can do! Here are some basic steps to getting started.

Determine the Scope

Before you get started, be sure you have a realistic idea of your goals and what you have to work with. Ask yourself, who will be using this classification scheme? What sorts of records do they create and manage? How will you get buy-in for planning and execution? What resources will you have for organization-wide training? To help you determine your scope, it’s a good idea to conduct a records inventory before moving on to creating your scheme. This initial stage is crucial to the ultimate efficacy of your new filing system. Take your time and enlist lots of help.

Choose a Filing Method

There are two main components to consider when designing a classification scheme. Every scheme, no matter the style, should outline consistent structure and usage of the filing system as well as standard naming conventions for folders and files. Bear in mind that your physical and digital records may require different filing systems, and that is ok.

Structure

There are three basic ways of organizing information, and all can be mixed and matched to address any filing situation.

  • Alphabetic – Good for records grouped by name (e.g., client name, county, building).
  • Subject – Good for records arranged according to the type of activity they reference (e.g., accounting, contracts, personnel, safety).
  • Numeric – Good for records arranged by year or number (e.g., purchase order, invoice, client number, birth date).
Image of a hierarchical file path including folders, with "Human Resources" being the first level; "Personnel" being the second level; "Johnson, Michelle" being the third level; "Annual Reviews" being the fourth level; and "2018 Review," "2019 Review," and "2020 Review" being the fifth level.

It is unlikely that just one of these filing systems will work for all your records. You will need to employ a combination of each filing type in order to effectively organize a wide array of records. We recommend using big buckets by grouping records into hierarchical retrieval units based on organizational functions and activities, which is an example of filing by subject.

In the example file path to the left, the function of Human Resources conducts the activity of personnel management, including annual reviews. Notice that this classification scheme includes all three types of filing systems – alphabetic for employee name, numeric for review year, and subject for the activities of personnel management and annual reviews. This type of hierarchical structure allows for a high level of interoperability between staff with varying degrees of expertise as well as intuitive discovery in cases when rarely accessed records need to be located or staff who are less familiar with departmental functions need to locate information.

Naming

Image of a list view of Word documents named "03.04.2012.docx," "03-04-12.docx," "04-03-2012.docx," "12.3.4.docx," and "040312.docx."

Now that we have an idea of how to structure files, it is just as important to understand naming conventions. Consider: Which of the files from the list on the right do you think contains meeting minutes from March 4, 2012?

Could be all, could be none! It’s hard to determine because any of these dates could possibly indicate March 4, 2012, but they could also be April 3, 2012, or December 3, 2004. This is because the files were not labeled consistently. To alleviate this problem, a classification scheme should establish controlled vocabularies, or naming rules. The types of rules you include will depend on the functions of your office, but below are common situations to consider.

  • Dates – Consider whether your files are best sorted by year (YYYY-MM-DD), month (MM-DD-YYYY), or day (DD-MM-YYYY). Is a single date format appropriate for every file, or are different rules required for different organizational structures?
  • Numbers – Determine which numbers are significant, for instance leading or trailing zeros (e.g., consider if your team would see “12,” “012,” and “120” as different numbers). If you use sectioned numbers, should they be separated with a space, dot, or dash?
  • Names – Your scheme should dictate whether files be sorted by last or first name and whether the full first name be used, or only the initial.
  • Acronyms – Many offices use acronyms to identify departments and programs. For instance, the TSLAC records management assistance team is commonly known as RMA. Ask yourself if the use of acronyms in filing is appropriate for your organization.
  • Abbreviations and Codes – Similarly, it is also common for offices to refer to functions and activities using abbreviations or codes, for instance commonly used shortenings like “ACC” for accounting, or home-grown reference codes such as LDZConf04 (Lorenzo de Zavala Building, Conference Room, 4th Floor). Use your classification scheme to communicate whether these abbreviations should be used in filing.

These are just a few common considerations, but keep in mind that your controlled vocabularies do not need to specify preferred usage for every instance above. These are jumping off points for identifying how your office refers to its functions. Once identified, codify those naming conventions in your scheme so that all records custodians can use the filing system consistently.

Retention-Conscious Filing

Once you are comfortable with the concept of structure and naming conventions within classification schemes, you can design your scheme in such a way that actually makes dispensation easier! One useful trick is to include the record number and/or retention period in the folder name of each individual record type. This method has the least impact when it comes to disposition, as retention periods mostly factor into the naming rather than the structure of the classification scheme. In the example below, a records custodian would need to look at every individual folder in order to determine which records are ready for disposition.

Image of a hierarchical file path including folders, with "Training Materials" being the first level and "GEN.018 Class Books -AC+1," "GEN.018 Classes -AC+1," "GEN.018 General Supp Materials -AC+1," "GEN.079 Course Development," and "GEN.080 Presentations MIS -5yrs" being the second.

Another method is to combine multiple related record types with equal retention periods into one bucket. For instance, Local Schedule GR, Part 3: Personnel and Payroll Records contains several records with disposition dates of “Date of separation + 5 years.” These could all be bucketed together in one folder for ease of disposition. In the example below, the records custodian only has to reference one master folder to determine whether these four different records types are ready for disposition.

Image of a hierarchical file path including folders, with "Personnel and Payroll" being the first level; "2020 Date of separation + 5yrs" and "2021 Date of separation + 5yrs" the second; and "GR1050-03 Awards and Commendations," "GR1050-14c Employment Applications," "GR1050-19 Fingerprint Cards," and "GR1050-28a Training and Educational Achievement" the third.

A final bucketing method is to combine several types of records in one folder and simply apply the longest inclusive retention period to the entire bucket. This method is the most convenient in terms of up-front effort, but can also lead to errors and unnecessary retention of records that have long been eligible for disposition. Remember that if a record is retained by a government and is subject to an open records request, litigation, or audit, that record must still be produced, even if it is eligible for destruction.

Implementation

Developing a classification scheme may seem like a daunting task, but the upfront investment can drastically improve your records management efforts. Ultimately, purposeful implementation is key. Whatever form your classification scheme takes, the most important steps are to document your use rules, disseminate them to the appropriate staff, arrange training if necessary, and be sure that all parties receive any future updates. Be consistent!

If you have implemented a classification scheme at your local government or state agency, we would love to hear about it! What was successful about your scheme, and what would you have done differently? Do you have tips and tricks to offer other records managers? Comment on this post to share your experiences.

Like it? 1

One thought on “Filing Crash Course: Classification Schemes

  1. As government employees with access to file rooms (and in the case of TSLAC – the stacks), we often dance like no one is watching. But when it comes to filing schemes, it’s incumbent upon us to file like EVERYONE is watching. Atavistic and individualized filing schemes are the bane of change management. Training new staff with, “this goes here – but we don’t know why – it’s just what it has always been” is embarrassing. Never too late to rework the wheel if you’ve got a bad set up. You’ll also likely discover some forgotten folders filled with dispo ready files waiting for a merciful axing.

    Good stuff, MJ.

Leave a Reply

Your email address will not be published.