Taxonomy design

Creating a taxonomy from scratch, if you haven’t done it before, can feel daunting. There are many ways to tackle this challenge, and here we are going to share practical advice that will help you build your taxonomy with confidence.

Here are a couple of things that you should know about taxonomies:

• Taxonomies are never finished. You can start with a simple draft and iterate as you learn how your team is using it to classify or search information.

• Taxonomies like to grow. If you are not careful you can end up with taxonomies that are very difficult to maintain. Make sure you are constantly simplifying and asking yourself, can we cut this first taxonomy draft by half? Would it still be helpful? Are they any overlapping concepts or patterns that we could simplify?
• Taxonomies are for the people who will be using them, not only for the creators. You need to research and validate whether or not you are going in the right direction. In the same way, you can’t build a product in isolation; you can not build a taxonomy without end-users feedback. 
• Taxonomies already exist in your company, especially in a business context; taxonomies come from the vocabulary people are already using. Don’t try to reinvent the wheel. Piggyback on the hidden taxonomies your companies use, for example, the information architecture of your product or website, the way the business is organized, the methodologies the teams are already using, and the customer journey map. The concepts your team uses to describe work or customer behaviors are a good starting point.

When it comes to classifying your data, whether your goal is to help people find documents easily or use a better tagging convention for data analysis, sharing a common vocabulary will set the right foundations for success, and taxonomies are designed to help you do just that.

At its simplest, a taxonomy is just a way of naming and classifying things according to their similarities and differences.

Taxonomy allows you to divide and group a large set of information into more manageable chunks. It helps you visualize how you want to organize your data.

Our recommendation is to create the first draft of your taxonomy and then ask yourself:

• Did you eliminate any potential synonyms? (fridge and refrigerator)
• Does any have more than one meaning? (Jaguar car or jaguar animal)
• Which have true relationships and need to have nuanced details?
• Is it the right level and not too specific or too broad?
• Is it understandable by everyone and not only understood by certain members?
• Did you start with the top use cases?
• Did you create documentation on governance and protocols?
• Are these aligned to your stakeholder needs and problem spaces?
• Did you prioritize and exclude edge cases?
• Does it or not fit in its correct structure?

As we mentioned before, testing your first draft is important, one way to test your taxonomy before you formalize it is to build a tree jack exercise and invite different types of stakeholders to it.