Atlas Term Sets and Taxonomy

What is a Taxonomy and how does it work?

Wikipedia defines Taxonomy as:

"...the practice and science of categorization or classification
...in which things are organized into groups or types."

Originally used in life sciences, particularly biology, the first official taxonomy was created by Carolus Linnaeus in 1758, who drew up rules for assigning names to plants and animals. Aristotle had attempted to classify organisms into defined groups almost exactly 2000 years before this- but that's enough history for one day!

During this article we will look at and explain the Atlas Taxonomy and it's accompanying term sets, what they are, what they do, and the best way to leverage these for your organisation and wider knowledge management practices.

Today, we are defining Taxonomy as:

"The organisation of search and navigation systems that will help people complete tasks,
find what they need and understand what they’ve found.​"

The Atlas Taxonomy will group terms, also known as labels or tags, into a hierarchy based on 5 core hierarchical 'term sets'. This Taxonomy, when applied to content, will help drive the quality and sustainability of Atlas, as well as best practices when it comes to information management, which will all help reinforce a positive user experience on a day-to-day basis and increase ROI by slicing through confusion and poor information governance which inevitably increases over time.

Metadata

"Metadata is a set of data that describes and gives information about other data."

Think of a clothes shop through an online website - I'm sure we have all purchased an article of clothing online - but the example can be defined to any website selling products. In this clothing website, you find many types of clothing, but, in this example, you only want to find a specific pair of jeans. 

The company and the website will often use metadata to narrow down your search in real life by allowing you to refine your search results using a series of pre-configured options or "tags".

In this example, you want to find a pair of size 12, bootcut jeans. We can break down this additional information in the following way:

Category = Jeans
Type = Bootcut
Size = 12

This will give us the result we set out to find.

This example can be applied to shoes - you need trainers, in white, size 8, from one manufacturer or brand only. You take the world of clothes, and you refine what you need against some already known existing categories and options (e.g. color or size).

In the same way, we can break down metadata for knowledge into categories when looking for items in your digital landscape. Let's say you want to find the most recent word document template from the Marketing Department:

Department = Marketing
Document format = word document
Document type = template
Time = within the last 3 months

Sort by = most recent first

Many out-of-the-box solutions do not provide a full range of refiners for metadata as part of their functionality. This is where the power of Atlas term sets come in, but before we delve into the detail, let's look at some other pre-requisites for a full knowledge transfer.

Microsoft Term store

The term store acts as the vehicle and platform which houses the Atlas Taxonomy, and as a SharePoint out-of-the-box tool, can be leveraged for the purpose of creating metadata. In our example of the clothes shop, the terms Category, Type and Size would be stored here. The 5 core Atlas term sets we are about to detail are deployed to, created in, and managed from within the termstore, and will be used to apply metadata to your workspaces and content in the form of tags.

Tagging

​Using the terms within the termstore, you can tag content with appropriate metadata to ensure additional information and context is provided to the document by those tags, so that the content can be searched for and refined; not only by the system through pre-configured search pages and directories but also in user-centric searches.

Each workspace during its setup (in Atlas ConneX) will have some default tags included, which will be applied to any content uploaded to that area. Of course, the user can alter these tags or add additional ones if they need to provide additional context to their specific content.

This is one of the core benefits and unique features of Atlas.

​These terms and concepts appear frequently in articles about Managed Metadata. You can learn more about managed metadata here Introduction to managed metadata - SharePoint in Microsoft 365.

Atlas's 5 core Taxonomy Term sets

There are 5 taxonomy classes or term sets in Atlas. We'll look at each one in turn below.

• Departments
  • Defined as the departments, sub-departments and teams, etc in your organisation. These should already be known and be well defined in terms of established organisation hierarchy. Some common examples may look like the below:
    • HR
      • Recruitment and staffing
      • Compensation and benefits
      • Training and development
    • Finance
      • Accounts Payable
      • Accounts Receivable
      • Book keeping
      • Audit 
    • IT
      • Infrastructure
      • Service Desk
      • R&D
      • QA
    • Operations
      • PMO 
      • Project Delivery Groups
      • Project Assurance

• Location
  • Defined as where you work or where your work takes place, such as as physical office locations AND projects, operations and/or customer geographical locations. Common examples may look similar to the below:  

• • • Office Locations
      • Europe 
        • Spain
          • Barcelona Office
          • Madrid Office
        • UK
          • London Office
          • Manchester Office
      • Africa
        • Egypt
          • Cairo Office
    • Project Sites
      • UK Sites
        • Birmingham
          • Birmingham New Street
          • Birmingham Cinema Complex
        • Wigan
      • Spain Sites
        • Alicante

• • Optional - Legal jurisdictions. These are incorporated as considerations which can impact someone’s work, project or document type. There are predefined ISO accredited legal jurisdictions available to include all of or cherry pick from. 
  • Some larger clients may also list all continents and countries of the world or at least the ones which touch their current marks. We say this as we are yet to engage a client who needs 'Antarctica' as a location term, but with scientific research being undertaken there it isn't out of the question when logistics are involved. Some clients prefer to include it, however others prefer to keep their list as concise as is needed.

• Subject
  • This term set represents the  Intellectual Property (IP) of your organisation, often containing terms to define Knowledge Topics, SME Areas, disciplines or niche topics used by your organisation.​ Some examples may include more generic or sector-related business subjects and usually the sectors or technologies themselves.
  • Subjects are usually not bound by time constraints, and will have no defined start or end point
  • They can be quite high-level, for example:

• • • Sectors
      • Residential
      • Commercial
    • Consulting
    • Risk
    • Ethics
    • Legal & Compliance
    • Geopolitics
•  
  • Or they can go very in-depth - using the Legal example above, but presenting many more tags for granular tagging.

• • • Legal
      • Acquisition
      • CoSec
      • Dispute
        • Contractual
        • Intellectual Property
      • Litigation
        • Appeal
        • Costs
        • Disclosure
        • Settlement

• Entities
  • Entities could be external organisations, such as customers, partners and suppliers, but they could also be internal entities, such as groups, sister or parent companies or even head offices.​ These may include, but are not limited to:
    • Customer names
      • Client A
      • Client B
    • Sub-contractors
    • Suppliers
      • Building Supplies
        • Supplier A
        • Supplier B
      • Office Supplies
        • Supplier A
          • Supplier B
    • Partners

• Activities
  • Internal or external activities (jobs, projects, tasks,...) are usually defined by a time span or time sensitivity, such as projects or taskforces. You can define activity by type, and also insert project codes for ease of search against your internal guidelines.
    • Activity Type
      • Bid
      • Matter
      • Lease Agreements
    • Engagement
      • Projects
        • PR001 - Project A
        • PR002 - Project B
    • Internal Projects
    • Supplier Engagement 
    • Social activities
      • Webinars
      • Blogs

How these term sets and tags are used within search

When the taxonomy is applied to Atlas, users can see and filter against used terms from within each term set. As an example, if we look at Department from within an Atlas search page, a drop down list of departments allows the user to refine the content being brought back to them.

Best Practice for Atlas Taxonomy

• Keep the Taxonomy sets simple to start with and use fewer terms.
• Terms can always be moved merged or renamed if needed. Taxonomies remain live and are very much a moving target, so do not big yourself down with trying to perfect a fixed Taxonomy - it isn't going to happen. Terms will always change and be introduced. For example, your organisation may adapt and enter new market or adopt new technologies, products, processes or working practices, creating an influx of new terms.
  • COVID 2019 and flexible working is always a good example we refer back to, where macro-environment impacts changed everyone's taxonomies. WFH became a new term and was tagged to new policies and guidance.
• Apply your “corporate taxonomy” within the “Atlas core term sets”, and make it relatable to your business, using terminology an end-user will understand.
• Consider what sets of terms may need to be common across the whole business, and where you will allow areas of the business some flexibility.
• ​Within the core term sets, create logical hierarchies – but try to keep it to 3-4 levels below the global term.
• Do not try and receive input and sign-off from every business stakeholders as they will likely have no context to the task and will put up resistance or unnecessary effort and detail. Move forward with the main structure and concept, and add terms needed as you go.
• Only two or three individuals should have admin access to the Atlas terms in the termstore, and the sets and structure should be carefully managed with proper governance to reduce mishandling. After a few years of wildly inputting terms the structure can be lost, duplications identified and the overall quality of information management reduces.

Consistent Challenges

• We often find the trickiest area of defining your Taxonomy within Atlas comes in defining which terms go into Subject and which go into Activity
  • For example, take the term 'survey' - this could be a document type (i.e. survey results, or survey templates), or it could be a subject covered, or, it could even be an activity when undertaking building/construction surveys. Often there is no wrong answer here, but the guidance on time span is a key factor here, otherwise it comes down to which makes the most sense for your use, however, we do advise not to put survey in all 3 unless specified as the best option for you.
    • Training is the same, and can be training material (information type), training exercises (activity) or training as a wider subject (subject/IP).
• Unfortunately we are unable to change what the core Atlas term sets are called, for example we cannot rename 'departments' to 'divisions' or 'teams'
• Terms can always be moved merged or renamed if needed, but must be centrally managed. If a user needs a new term, we usually advise they submit a ticket to IT, or even better - they raise a 'feedback' item within Atlas for a new term and it's sent to your Atlas platform owner to insert, manage, and generally be aware of.

If you have any queries or would like additional support, please contact your closest ClearPeople or Atlas representative.