Taxonomies

A taxonomy is a top-down hierarchical classification scheme of preferred terms, dividing general concepts into progressively more specific ones through parent-child relationships. Three types of logical relationship can connect terms within a taxonomy: broader/narrower, the most common: "a narrower term is a kind of the broader term," like "oak is a tree"; instance-of, where a general category has specific, often proper-named instances, like "Zoom is an instance of an online-meeting application"; and whole/part, like "a page is part of a website."

Most taxonomies place each term in exactly one location, a strict hierarchy, but a polyhierarchy allows a term to appear in more than one place when a concept genuinely belongs in multiple categories, like a piano being both a percussion instrument and a stringed instrument. Polyhierarchy improves findability but is harder to maintain, so it's used sparingly rather than by default. Interestingly, the categories people find easiest to work with aren't at the very top or bottom of a taxonomy. Mid-level categories, concrete enough to picture but general enough to be widely understood, are the most cognitively natural starting point for most people.

"Animal" is too abstract to picture. "Himalayan cat" requires knowledge most people don't have. "Cat" sits in the middle, and that middle is where most people's natural thinking actually lives.

Exercise

The scenario: A media company's content taxonomy is being reviewed. Three structural problems have been identified in it.

1. The taxonomy places "Podcasts" as a narrower term under "Articles," even though a podcast is a completely different kind of content object, not a subtype of an article.
2. "The New York Times" appears in the taxonomy as a narrower term under the broader category "News Sources," treated the same way as a broader/narrower content category.
3. An article about a celebrity chef's restaurant opening has been placed under both "Food" and "Business," with no consistent rule for when this dual placement is allowed elsewhere in the taxonomy.
Try It With Your Data: Hierarchical Taxonomy Generator

Fill in your own details below; the prompt updates as you type. When it's ready, copy it into Claude or whatever AI tool you use.

What this prompt is made of (RICE breakdown)
R

Role

Casts the AI as: You're an information architect specializing in software documentation taxonomies.

I

Instructions

Spells out exactly what to analyze and return.

C

Context

Target audience, Current problem and Business goal tell the AI what your specific situation is, not a generic one.

E

Expected format

Output as markdown indented list with: top-level categories, one level of children, and a short scope note under each top-level label.

Whole number, e.g. 3
e.g. "User task (what they want to accomplish)"
e.g. "Feature area"
e.g. "Content type"
Character limit, e.g. 25
Any other requirements
e.g. "200 pages of developer documentation"
One per line, e.g. "Authentication (40 pages)"
Brief description
What's wrong today
What success looks like
Assembled prompt

        
        
      
Try It With Your Data: Exploratory Taxonomy Pattern

Fill in your own details below; the prompt updates as you type. When it's ready, copy it into Claude or whatever AI tool you use.

What this prompt is made of (RICE breakdown)
R

Role

Casts the AI as: You're an information architect exploring organizational approaches.

I

Instructions

Names 3 specific outputs to produce, so the response comes back structured rather than a general summary.

C

Context

Content, Audience, Content areas and Goals tell the AI what your specific situation is, not a generic one.

E

Expected format

For each taxonomy option: the organizing principle, a sample top level, strengths, risks, and when you would choose it over the others.

Whole number, e.g. 3
Name and one-line description, e.g. "Task-based: organized by what users want to do"
Name and one-line description
Name and one-line description
e.g. "API documentation with 200 pages"
Brief description
Comma-separated list
What you're trying to achieve
Assembled prompt

        
        
      
Try It With Your Data: Diátaxis Integration Pattern

Fill in your own details below; the prompt updates as you type. When it's ready, copy it into Claude or whatever AI tool you use.

What this prompt is made of (RICE breakdown)
R

Role

Casts the AI as: You're an information architect who specializes in the Diátaxis framework.

I

Instructions

Asks for a taxonomy that integrates all four Diátaxis types, using the structural approach you choose (Diátaxis-first, feature-first, or hybrid).

C

Context

Content and Features/Topics tell the AI what your specific situation is, not a generic one.

E

Expected format

Output as indented list with: Diátaxis types as top level (or facets), topic categories beneath, and a note on where hybrid pages should live.

e.g. "3-level taxonomy for our developer docs"
e.g. "Option A: Diátaxis-first (content types at top level)", "Option B: Feature-first with Diátaxis at lower levels", or "Option C: Hybrid (user journey + Diátaxis)"
What organizes the top level
What organizes the second level
Usually the Diátaxis types themselves
e.g. "200 total: 40 tutorials, 60 how-tos, 80 reference, 20 explanation"
One per line, e.g. "Authentication (40 pages)"
Assembled prompt

        
        
      
Try It With Your Data: Taxonomy Validator

Fill in your own details below; the prompt updates as you type. When it's ready, copy it into Claude or whatever AI tool you use.

What this prompt is made of (RICE breakdown)
R

Role

Casts the AI as: You are an information architect validating taxonomy balance and quality.

I

Instructions

Names 4 specific outputs to produce, so the response comes back structured rather than a general summary.

C

Context

Total pages, Target audience and Content domain tell the AI what your specific situation is, not a generic one.

E

Expected format

Provide: a pass/fail per check, the specific nodes that fail, and a concrete rewrite for each failing label or relationship.

Paste your taxonomy structure with page counts per category
Whole number
Brief description
e.g. "developer documentation"
Assembled prompt

        
        
      
Try It With Your Data: Balance Validator

Fill in your own details below; the prompt updates as you type. When it's ready, copy it into Claude or whatever AI tool you use.

What this prompt does

Use this prompt when: Creating primary taxonomy from scratch.

  • After generating or refining a taxonomy
  • Before finalizing navigation structure
  • When taxonomy feels "off" but you can't pinpoint why
  • As part of systematic taxonomy validation (5-Check system)
Paste your taxonomy structure with page counts per category
Whole number
Brief description
Assembled prompt

        
        
      
Try It With Your Data: Taxonomy Refinement Pattern

Fill in your own details below; the prompt updates as you type. When it's ready, copy it into Claude or whatever AI tool you use.

What this prompt is made of (RICE breakdown)
R

Role

Casts the AI as: You're an information architect refining an existing taxonomy.

I

Instructions

Names 3 specific outputs to produce, so the response comes back structured rather than a general summary.

C

Context

Total content, Audience and Business goal tell the AI what your specific situation is, not a generic one.

E

Expected format

Output:

Paste the taxonomy as it stands today
One problem per line
One requirement or constraint per line
Whole number
Brief description
What success looks like
Assembled prompt