Minimum Viable Taxonomy Level 2: Principles and Best Practices for Organizing Research Knowledge

Minimum Viable Taxonomy Level 2: Principles and Best Practices for Organizing Research Knowledge

MVT Level 2 is a toolkit for building out a taxonomy that facilitates report categorisation and more complex activities such as data analysis and insights discoverability.

The MVT Level 2 process was led by Annette Boyer who authored the original toolkit on which this article is based.

This article was written by Will Edmiston, Rogério Lourenço, Karolina Kowalczyk, Janene Batten, and Susan Montgomery on behalf of the entire MVT Team.

Defining levels of research taxonomy

Taxonomies are essential for effectively managing research knowledge. We've separated our guidance on taxonomies into two levels:

MVT Level 1

In October 2022, the Minimum Viable Taxonomy (MVT) working group, as a part of The ResearchOps Community, released the MVT Level 1 which was a continuation of the work begun in the Research Repositories Program Team as described in the article Research Repositories: A ResearchOps Community Program of Work.

MVT Level 1 is intended to be the first level of organizing and indexing research documents and artifacts, and it supports research registers and research library knowledge management systems (KMS). You can use it to add standard, descriptive information across your research deliverables and other assets. MVT Level 1 provides a common starting point for the taxonomy of various types of research knowledge management tools, regardless of industry, organization type, or tool being used.

MVT Level 2

The Minimum Viable Taxonomy Level 2 (MVT Level 2), the focus of this article, is a process for creating and managing a taxonomy for your organization. If MVT Level 1 is a baseline of descriptive terms, MVT Level 2 is a toolkit for building out a taxonomy that facilitates report categorisation and more complex activities such as data analysis and insights discoverability.

Taxonomies are unique and dependent not only on the sector but also on specific organizational goals and processes. Although components of MVT Level 2 might be shared across organizations and sectors, we determined that providing a set of principles and best practices to help you move forward with a larger group of terms will be more beneficial than  developing a list of specific terms. Our goal is to equip you with the skills to create your own taxonomy in order to help you manage more detailed information such as user research insights and transcripts from interviews.

Getting started

Is there a need?

When embarking on a taxonomy project, one of the first things to do is a needs assessment. What problem is your taxonomy solving for? Assess where you are with your research program in the organization. In short, take stock.

3 key indicators that your organization would benefit from a more detailed taxonomy:

  1. Research insights cannot be found by co-workers who need them when they need them.
  2. Research is repeated unnecessarily because the team could not find the prior work.
  3. Frequent requests received for information from previous research.

Organizational Maturity

Keep in mind that organizational maturity will affect the build-out of the repository and its taxonomy. Do you have the resources and the infrastructure that can both support and benefit from MVT Level 2?

MVT Level 2 functional requirements include:

  1. Alignment - You must have an understanding of your organizational needs and strategy.
  2. Buy-in - You must be able to get buy-in from the teams (your researchers) most affected by implementation as well as key stakeholders.
  3. Budget - You will need a budget for tools, personnel, legal resources, and time.
  4. Data - You will need the data and the ability to create a list of material to be organized and described.

The why and how of MVT Level 2

MVT Level 2 connects your data and research (the why)

A well-thought-out taxonomy creates better searches, enables filtering capacities, and perhaps most importantly improves your access to the research within your organization.

The principal idea of the Minimum Viable Taxonomy project is to give you a set of tools, procedures, and best practices to organize and describe your research, artifacts, reports, templates, findings, and insights by connecting them through consistent categorization. Making your research  discoverable and actionable across teams and projects will also reduce research waste by making your information reusable and interconnected.

A well-developed taxonomy is not only about documentation or organization. It makes it possible to create opportunities for new discoveries. It can bring new meaningful connections and drive conversations outside of discrete projects and teams. When you organize things in a way that most people in your organization can understand, it can bring about serendipity, setting the stage for new patterns to emerge and take shape.

The tools, procedures, and best practices proposed in MVT Level 2 are designed to bridge the social and cognitive gaps brought on by time, personnel changes, new initiatives, and leadership changes. Information connections are not a given. They need to be built. Taxonomies are living structures, and require attention to be and stay relevant.

So, to keep them relevant, continually address the common gaps that organizations face when information silos exist. The first gap is the one that enables you to synchronize the information you have in the repository from past projects with the current one. The second one arises from a social and cognitive perspective. Conceiving, producing, and storing information is a fundamentally human activity. It demands soft skills as much as technical ones.

MVT Level 2 applications (the how)

We thought of a few scenarios in which a user experience taxonomy would be of use. There are many possibilities to organize information into hierarchies that help you.

Applications of Minimum Viable Taxonomy Level 2: 1. Search Optimization: More accurate and relevant search results, 2. User Navigation: Easier to find what they're looking for, 3. Content Categorization: Simpler to retrieve specific info, 4. Data Analysis: Easier to derive insights and patterns from data, 5. Content Management: Improved structuring for publishing and distribution, 6. Metadata Management: Better retrieval and preservation of assets, 7. Product Development: Improved alignment with user needs, 8. Compliance and Regulation: Better adherence to industry standards and regulations, 9. UX Design: More user-friendly product structures, 10. Personalization: Improved categorization of user preferences and behaviors.

Here are some examples:

    • Data Analysis: Taxonomies can be used in data analytics to categorize and classify data, making it easier to derive insights and patterns from the data.
    • Content Management: Taxonomies are valuable for content management systems, aiding in the organization and structuring of content for publishing and distribution.
    • Metadata Management: In digital libraries and archives, taxonomies help manage metadata for digital assets, aiding in retrieval and preservation.
    • Product Development: They can support product development by categorizing features, functionality, and user requirements, ensuring the product aligns with user needs.
    • Compliance and Regulation: Taxonomies can assist in ensuring that a digital product adheres to industry standards and regulations by categorizing and tracking relevant information.
    • User Experience Design: Taxonomies can guide the design of user interfaces, helping to create a logical and user-friendly structure for the product.
    • Search Optimization: Taxonomies can improve the search functionality of a digital product by enabling more accurate and relevant search results.
    • User Navigation: Taxonomies can be used to organize and label the content or features within a digital product, making it easier for users to navigate and find what they're looking for.
    • Content Categorization: A taxonomy can help categorize and tag content, making it simpler to search and retrieve specific information or media within the product. Example of reducing waste/ recycling/ scaling with what you have
    • Personalization: They can assist in creating personalized user experiences by categorizing user preferences and behaviors, allowing for tailored content recommendations.

Best practices for creating your MVT Level 2

These recommendations originally came from Annette Boyer’s Toolkit, and we’ve retained much of the language and spirit of her work. They are intended to assist in building a taxonomy after or in conjunction with the terms laid out in MVT Level 1. The following recommendations are not linear and are not a prescribed path but guidance to draw on as needed.

Best practices for creating your Minimum Viable Taxonomy Level 2: Arrange listening sessions, Listen continuously, Collaborate actively, Empathize with your users, Align objectives and customize, Build in governance, Carefully select your tools, Be intentional and methodical, Persevere and trust yourself, Create a community of practice, Track your progress.

Arrange listening sessions: Conduct stakeholder interviews before and after your rollout

Actively collect input and feedback from your users including listening to the various key audiences.

Framing your questions is important. Considerations might include if there is already a research repository and some kind of taxonomy in place. Consider an audit to discover terms already in use. Another is to be aware of your audience and their familiarity with research. Here, we conceived of 4 possible scenarios with questions to ask.

The first scenario is where a repository is already in place and you're interviewing stakeholders who are not researchers.

Questions to ask when there is an existing repository and your meeting with non-researchers:

  • What are some examples of how you use the repository?
  • Via screen sharing, show me how you look for what you need in the repository.
  • What works about it fulfilling your needs?
  • How else might it help you with your work?

Watch for:

  • Unexpected usage of your repository (using search instead of filters, for example)
  • Terms repository users want to search or filter on that you did not expect.
  • Terms that have changed in your organization when an area of your work is renamed.
  • Examples of success stories that you can use to promote your work and inspire further usage. Record the session for this purpose and leverage the transcript as quotes in promoting your repository.

The second scenario is when there is an existing repository and you're interviewing the researchers who use it

  • Remember that research teams are juggling a lot of competing priorities.
  • Create an easy low-effort way for researchers to contribute to your taxonomy, even as simple as suggesting an update or change in an online discussion channel.

The third scenario is when the repository is new and you are meeting with researchers

These are examples of questions you may ask your colleague during workshops or one on one:

  • What is your ideal day as a researcher?
  • What does utopia look like in your research work as you share your insights and look for ways to have an impact? What could make it easier for you to impact your work?
  • Could you walk me through the last time you were looking for previous research?
  • How do you determine if past research can be repurposed for future work?

Watch for:

  • Differing definitions and assumptions of the role and function of a repository

And the fourth scenario is when you have a new repository and are interviewing with non-researchers

Conduct a workshop or 1:1 interviews to ask:

  • In a perfect world, how would you find past research reports or insights for your work?
  • What has worked in the past?
  • What has not worked in the past?
  • How do you know if a topic you are interested in has already been covered by past research?
  • How do you determine if past research can be repurposed for future work?
  • What systems do they already use to organize and capture their work? Do they have a filing system? A wiki? A template library? Workflow documentation? What works well and not so well about any current tools or systems?

Watch for:

  • Differing definitions of the role and function of a repository

Listen continuously: Maintain an ongoing, open line for feedback and communication

The taxonomy lead should develop their practice of noting possible new terms in your organization that should be accounted for in the taxonomy. During project meetings, all hands, and 1:1 discussions, you may hear valuable terms that can be used to expand your taxonomy.

Be alert for terminology being updated across your organization and reflect that in your taxonomy. Like today’s software development, your taxonomy will experience frequent small iterations and improvements. It will evolve. The more you can be attuned to these small additions, the more accurate and precise your taxonomy will be.

Watch for:

  • new terms
  • synonyms
  • terms that might need disambiguation

Collaborate actively: Encourage participation to ensure the process is productive

Create a process for your team to nominate items for your taxonomy and periodically revisit the board to reconsider your taxonomy. Use a digital whiteboard tool to collect input on your taxonomy. Using a whiteboard tool, you can help everyone keep the full taxonomy in mind as they work with it.

Empathize with your users: Respect their time, effort, and work

Your internal customers are often overworked and very busy. Keep their mindsets and needs at the forefront of your planning.

At its core, a taxonomy is an efficiency tool. Do not over-engineer your taxonomy. Taxonomies that are too complicated or too hard to remember will not be adopted and will eventually fail. A good-faith effort to create a highly detailed taxonomy might be more than what your team or project actually requires. Keep your team and organizational needs at the forefront of your taxonomy project.

Unnecessary complexities create administrative problems as well. If managing your taxonomy starts to feel more like wrangling an unwieldy list of words and how they’re being applied then it is time to reel it in, assess and reestablish a sustainable, manageable scope.

When in doubt, focus on the simple and yet elegant goal of helping people find research more easily. Rather than taking a maximal approach, a more appropriate goal might be to develop the minimum level of taxonomy required to support your internal teams’ busy work lives and then stop there.

Align objectives and customize: Keep the needs of your organization foremost in your mind

Make sure that your listening and collaboration efforts are in line with your broader organizational goals and objectives. Specific organizations and organizational types are going to have different needs. There's no one-size-fits-all way to build your taxonomy. There's a need to be patient and trust yourself and the process. This applies not only to the taxonomy but to your entire repository rollout.

The researchers using your taxonomy might need encouragement, reassurance, and reminders about how the taxonomy works. Use gentle reminders and updates with your researchers, and when people make mistakes in their taxonomy application, help guide and redirect them. The researchers and teams using the taxonomy and research repository are part of your community of practice. Part of your role in getting buy-in for the taxonomy means creating space for your researchers to make mistakes as they begin to use it. Reassure them that there’s no need for guilt about lack of knowledge, just as there’s no need for guilt from the customers for whom we build our products and services.

Keep your corporate/organizational culture and key audiences in mind. Also, account for the sector you work in: private, education, government, healthcare, etc. What needs are relevant to your type of organization?

Build in governance: Establish a framework for governance to oversee the process and ensure that it aligns with your objectives

Set principles and standard practices for your taxonomy to increase future-proofing.

Establish a single editor. Suppose more than one person is making the final decision to update the taxonomy. In that case, you may end up with terms that must be fully considered in their impact across the entire organization.

The single editor/taxonomy lead should establish a small team of trusted colleagues to become advisors and experts to occasionally convene, collaborate, and talk through ideas and brainstorm. This group can also advocate for the taxonomy across the organization.

Document your approach to building out the controlled vocabulary and synonyms.  Include tips for eliminating ambiguity and defining relationships between terms. At a minimum, a simple one-sheet overview would help anyone grasp the essential details of your taxonomy and enable them to pick up its maintenance and development in your absence.

Collaborate with your team to create clear definitions for vocabulary terms in your taxonomy. Ask others to review for understanding and reduce ambiguity.

Your taxonomy is dynamic and will continue to evolve. Taxonomy terms and their relationships and meaning will evolve with your research, your team, and your organization. Taxonomies are not a set-it-and-forget-it type of tool. Create a system for updating your taxonomy on an ongoing basis.

Carefully select your tools: Consider how the features of your repository tool will support your intended level of taxonomy

Evaluate your chosen tools carefully because they will give shape and form to your taxonomy. The tools you use will influence how you manage your taxonomy. Features to look out for are automated rules, interoperability, permissions, and control features. Ensure that your taxonomy tool, if separate, is compatible with your repository tool.

Be intentional and methodical: Approach the process with purpose and a systematic methodology

Take your time to listen closely to your internal customers and apply your overall institutional knowledge and experience to translate what they share with you into themes that you can act upon. Apply your research skills for the best results. It is tempting to jump into action when someone important in your organization gives you feedback. Still, there is value in seeing the entire landscape of input before acting.

Persevere and trust yourself: Stay committed and have confidence in your abilities and the process

As we have mentioned, there is no one route toward a completed taxonomy because it will morph over time and never truly be complete. This also means that the way forward  may be undefined and ambiguous which can lead to feelings of uncertainty. Avoid imposter syndrome by building your community of practice, trusting yourself, and staying the course. A sustainable taxonomy is not fixed but evolves and improves over time.

Create a community of practice: Be specific and intentional

When embarking on a taxonomy project you are creating a community of practice where you and your teams will learn together. A taxonomy that is implemented well is an ongoing process and not a final product. It changes and evolves. But it also needs definition and control.

Managing productive conversations around your taxonomy requires a light touch. You will want to remain firm in your definitions and confident in your choices and process while also creating a space where your researchers and other users feel comfortable experimenting, practicing, learning, and adopting.

Set the tone and move toward success by agreeing on some basic terminology early on. Determine what you are calling your list of words. Are they: fields, properties, elements, facets, or something else? Being mindful and making deliberate choices in your terminology provides some constraints that will facilitate understanding and help your conversations move forward.

Track your progress: Monitor and assess the success and impact of your efforts

Own your contributions. At the pace that most of us work, it can be tempting to keep moving forward without reflecting on or recording your wins. For efficiency tools like a repository with a supporting taxonomy, it’s incredibly helpful to track your successes for a few reasons. Some benefits of recording your achievements and sharing them out are:

  • To provide the evidence needed to continue funding the role of the repository and your place in it year over year - adopt an OKR measurement (objectives and key results).
  • Helps decision-makers understand the importance of your taxonomy work and enables you to continue with your taxonomy work instead of being redirected to other work.
  • Enables you to reflect on how to make the next set of taxonomy improvements.
  • Serves as a reminder of your progress on those days when you need a reminder of your accomplishments.
  • Adds to your list of achievements for your annual review preparation. You are your own best advocate.


A summary list of our recommendations for developing your MVT Level 2 are:

  • Arrange listening sessions: Conduct stakeholder interviews before and after your rollout.
  • Listen continuously: Maintain an ongoing, open line for feedback and communication.
  • Collaborate actively: Encourage participation to ensure the process is productive.
  • Empathize with your users: Respect their time, effort, and work.
  • Align objectives and customize: Keep the needs of your organization foremost in your mind.
  • Build in governance: Establish a framework for governance to oversee the process and ensure that it aligns with your objectives.
  • Carefully select your tools: Consider how the features of your chosen repository tool will support your intended level of taxonomy.
  • Be intentional and methodical: Approach the process with purpose and a systematic methodology.
  • Persevere and trust yourself: Stay committed and have confidence in your abilities and the process.
  • Create a community of practice: Be specific and intentional.
  • Track your progress: Monitor and assess the success and impact of your efforts.

Reach out on the ResearchOps Community #taxonomy Slack channel if you’d like to be involved in future aspects of the taxonomy project.


How we created this MVT Level 2 guidance

The Team

The MVT Level 2 team consisted largely of ResearchOps people and Librarians. Our group did not include Taxonomists, but rather people who have Research expertise who work with taxonomies and structured language and information. We met weekly to discuss taxonomies and the MVT Level 2 project.

MVT Level 2 participants include the following: Rogério Lourenço, Will Edmiston, Mark McElhaw, Annette Boyer, Janene Batten, Karolina Kowalczyk, Susan Montgomery, India Anderson. Other contributors include George Jensen, Ian Hamilton, Emily DiLeo, Constanza Reca, Fatima Kamali, and Lourenço Rodrigues.

Biases, constraints, and topics covered

We wrote MVT Level 2 to be broadly applicable building blocks to a robust taxonomy to support Research teams organizing their research by building research libraries and repositories. MVT Level 2 is intended to be tool agnostic. It has not been designed for a specific user research software tool available on the market.

Further, the toolkit considers how taxonomy and Personally Identifiable Information  (PII) intersect in different regions of the world where research is being done, the ethics of taxonomies, and how multilingualism can shape how a taxonomy develops. Especially within the global ReseachOps Community, each of these areas of interest and inquiry are worthy of their own study and development. While MVT Level 2 attempts to account for them, it does not cover these areas in detail.

Our process and research

MVT Level 2 development hinged upon weekly discussions. What was an MVT Level 2? What would our deliverables look like? Are we putting together a list of terms or something different? A playbook? A board game? What does MVT Level 2 look like?

Through our discussions, we created term outlines, categories, miro boards, taxonomy sandboxes, and visuals. Members from previous group iterations joined us to share their findings from previous research and to hand off Level 1 to Level 2. Other members of the community who were interested in working on taxonomies joined our group and shared the taxonomies they built for their organizations. We surveyed the Research Ops Community to learn more about people’s experience working with taxonomies and where they could use guidance and support in those areas.