Faceted Classification

Facets are independent, mutually exclusive attributes of content that together describe an entire domain, think of an outdoor gear catalog described simultaneously by activity, brand, price range, and season, where each of those is a separate facet a shopper can filter by independently. Facet analysis is an eight-step process: collect a representative content sample, briefly describe each item, compile a list of indexable concepts and preferred terms, decompose any compound concepts into simpler parts, create and refine the facets themselves, create and refine the ordered term lists within each facet, organize everything into a coherent classification scheme, and finally revise the content to conform to it.

Faceted classification earns its complexity in specific situations: new or emerging domains without an established vocabulary, interdisciplinary fields where different experts bring different perspectives, and large or complex information spaces where a single hierarchy, even a polyhierarchy, can't flexibly express items that genuinely belong along several independent dimensions at once.

A strict hierarchy asks "where does this belong?" as if there's one right answer. Faceted classification asks "what are all the true things about this?" and lets someone find it along whichever one they happen to be thinking about.

Exercise

The scenario: An online retailer selling outdoor equipment wants to improve findability across a catalog of four thousand products. A taxonomy already exists, organized by product type, tents, boots, backpacks, and so on. Users are consistently failing to find products when they approach the catalog by use case, season, skill level, or brand rather than product type.

Given that users are searching along multiple independent dimensions the current taxonomy doesn't support, what should the retailer do?
Try It With Your Data: Faceted Classification

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

Designs facet dimensions and values so the same content can be found along multiple meaningful paths.

  • Content needs multiple access paths (by topic, type, difficulty, language)
  • Users need to combine filter criteria
  • Large content sets requiring filtering UI
  • Building documentation with multiple dimensions (e.g., API docs by language + topic + content type)
Brief description of the content
Name and values, e.g. "Difficulty (values: Beginner | Intermediate | Advanced)"
Name and values
Name and values
Name and values
Paste the list of content to classify
Describe the filtering scenario
Technical limitations
Assembled prompt