Skip to main content

Attributes: Types, LoVs and data behaviour

Attributes define the structure and meaning of your product data. Each attribute tells SKULaunch what kind of information you expect (for example Brand, Length, Flavour) and how to validate, normalise and enrich that information.

S
Written by SKULaunch Support
Updated over 4 months ago

Attribute fundamentals

An attribute describes a single piece of product information.
Examples:

  • Brand

  • Product Name

  • Life Stage

  • Voltage

  • Pack Quantity

  • Net Weight (g)

  • Wood Species

Attributes can apply across multiple families or only a single family, depending on your schema design.

Every attribute has three important parts:

  1. Type (controls the format of the data)

  2. LoVs (optional lists of allowed values)

  3. Behaviour (how SKULaunch stores, validates and enriches the data)

Attribute types

SKULaunch supports a wide range of attribute types.
Each type has its own behaviour, rules and impact on AI enrichment and data validation.

Below is a description of each type.

1. Text

Free-form text for short strings.

Examples:
Brand, Colour Name, Model Name

Behaviour:

  • Accepts any string

  • No validation beyond basic formatting

  • AI extracts explicit text values


2. Textarea

Long-form text used for descriptions or multi-sentence content.

Examples:
Long Description, Ingredients, Care Instructions

Behaviour:

  • Supports paragraphs

  • Used for content generation

  • Not validated structurally


3. Identifier

Machine-focused string intended to uniquely identify a product.

Examples:
MPN, SKU code, GTIN (string)

Behaviour:

  • Not normalised

  • AI only fills when explicitly detected


4. Number

Numeric values without units.

Examples:
Pack Quantity, Protein Percentage, Wattage (if stored without units)

Behaviour:

  • Only accepts numeric values

  • AI normalises numbers

  • No unit interpretation


5. Dimension

Numbers with units.

Examples:
Length, Width, Height, Volume, Weight

Behaviour:

  • Supports unit detection and normalisation

  • AI can convert units

  • Ensures consistent measurement data


6. Boolean

Yes/No fields.

Examples:
Is Organic, Waterproof, Battery Included

Behaviour:

  • Accepts true or false

  • AI classifies explicit or strongly implied values


7. Date / Time / Timestamp

Structured date and time fields.

Examples:
Release Date, Expiry Date, Manufacturing Date

Behaviour:

  • Validates against date formats

  • AI extracts explicit dates only


8. Simpleselect

One value from a predefined List of Values (LoV).

Examples:
Life Stage, Food Format, Country of Origin

Behaviour:

  • Must match one LoV value

  • AI maps extracted text to the closest allowed value


9. Multiselect

Multiple values from a predefined LoV.

Examples:
Suitable For, Features, Compatible Models

Behaviour:

  • Allows multiple selections

  • AI may output several matches


10. Range

A numeric range.

Examples:
Voltage Range, Temperature Range, Age Range (minimum/maximum)

Behaviour:

  • Stores a lower and upper bound

  • Validates numeric structure

  • AI can extract ranges when explicitly stated


11. Price

Structured price information.

Examples:
MSRP, List Price, Wholesale Price

Behaviour:

  • Stores amount and currency

  • Validates number format

  • AI only extracts explicit prices


12. Rating

A scored rating.

Examples:
Energy Rating, Safety Rating, Efficiency Rating

Behaviour:

  • Stores a numeric score with optional scale

  • AI extracts explicit ratings only


13. Image

An image asset.

Examples:
Main Image, Additional Image

Behaviour:

  • Stores a single uploaded image

  • Allows validation and replacement

  • Used by AI Image Extractor to pull text from packaging


14. Video

Video asset storage.

Examples:
Demo Video, Installation Video

Behaviour:

  • Stored as uploaded video or URL

  • AI does not interpret video content


15. File

Any uploaded document.

Examples:
Datasheet PDF, User Manual, Warranty Document

Behaviour:

  • Stores the file in SKULaunch

  • Can be used by the AI File Extractor


16. URL

Single web link.

Examples:
Manufacturer Page, Product Link

Behaviour:

  • Must be a valid URL

  • AI populates only if explicitly present


17. Multiurls

Multiple web links.

Examples:
Download Links, Safety Certifications, Additional Resources

Behaviour:

  • Accepts multiple URLs

  • Validates formatting


18. Geolocation

Latitude/longitude coordinates.

Examples:
Manufacturing Location, Origin Coordinates

Behaviour:

  • Stores structured coordinate pairs

  • AI does not extract geolocation data


19. Multivalue

Multiple free-form text entries.

Examples:
Alternate Names, Synonyms, Internal Tags

Behaviour:

  • Allows an unstructured list of values

  • Unlike multiselect, values are not restricted by a LoV


20. Table

A structured multi-row, multi-column dataset.

Examples:
Technical Specifications Table, Size Chart, Nutritional Breakdown

Behaviour:

  • Stores tabular data

  • Rows and columns are defined per attribute

  • AI does not auto-populate tables in current versions

Did this answer your question?