Summary Benchmarks

The purpose of these criteria is to suggest an “industry standard” for absolute minimum requirements and a “gold standard” for the quality of a descriptive metadata record.

Usage:

• Each criterion is intended to be “system agnostic” but some may not apply to every situation (e.g., local field requirements)

• Criteria are binary – i.e., the set being evaluated must meet all points or it does not meet the benchmarking standard for that level

• These benchmarks focus solely on the quality of metadata entry, not the quality of information (i.e., available information is all entered correctly, although we might wish that additional information is known about an item to improve the record)

• This framework is intended to be scalable (it is written in the context of 1 record, but could apply across a collection, resource type, or an entire system)

• Minimal criteria apply in all cases; suggested criteria do not rise to the level of “absolute minimum” but are suggested as priorities for “better-than-minimal” based on our research and experience; ideal criteria tend to be more subjective and may not apply in every situation

Benchmarks

This is the summary version, listing only the individual benchmarks, organized according to the various aspects a metadata manager might review or verify in a record. See the Expanded Benchmarks for more details and examples related to each benchmark. The summary is also downloadable as a PDF file.

Content relevance

These criteria are about the relationship between the record and the item.

Minimal	Suggested	Ideal
The record is specific/scoped correctly (e.g., records for single manuscripts are not described as whole collections of materials or vice versa)	The record describes the item that it is attached to	All metadata values align with expectations for the material type (e.g., language use for text vs. images; theses/dissertations have creator or author values) When applicable, relationships between items and parent collections are clearly represented All information not specific to the item has been removed (e.g., duplicated records propagated across a collection have been individualized; information not appropriate for the item type is removed) Administrative/acquisition information is appropriately noted and/or hidden from the public, if applicable

Existence of values

These criteria are related to when fields should have values or not (regardless of value formatting or correctness).

Minimal	Suggested	Ideal
Every record has a title	All locally-required fields have values All conditionally-required fields have values Fields that require multiple parts or qualifiers have all parts Records have some type of subject value (e.g., subject, keyword, genre, etc.) A rights statement is present (standardized statements preferred)	Relevant recommended/optional fields have values All relevant information about the item is included Non-required qualifiers or field parts are added to provide enhanced information or functionality “Null” values are used consistently, according to local guidelines (e.g., N/A, unknown, leave blank, etc.) Fields/subfields that cannot be repeated occur only once

Content of values

These criteria are focused on the field values specifically – e.g., string values, formatting, etc.

Minimal	Suggested	Ideal
Value content matches the field type (e.g., “date” fields do not contain text strings) No values exceed applicable system character limits There is no text encoding that “breaks” records (i.e., records are not prevented from being displayed)	Stray character encoding has been removed (e.g., MARC subfields that may have been pasted in) All “placeholder” values have been replaced/removed and are not present in the publicly accessible record (e.g., {{{name}}} or YYYY-MM or notes intended to be removed by editors)	All values are appropriate lengths for their fields All values that ought to align with standards conform to applicable vocabularies or rules (e.g., names, dates, locations, subjects, etc.) All values are spelled correctly Text fields use appropriate punctuation, grammar, abbreviations, etc.

Community/audience

These criteria describe how metadata records reflect the needs of users.

Minimal	Suggested	Ideal
	Extremely problematic/offensive terms have been removed or handled appropriately	Reading level & language use is appropriate for all (relevant) communities or audiences (e.g., collections intended for students match their reading level; collections intended for scientific research use technical terminology) Vocabulary usage aligns with the needs of the audience and material type (e.g., technical vocabularies are used for scientific materials, but not for collections intended for amateurs) Values connected to interface functionality are correct (e.g., field values used for browsing or filtering search results) Record language has been evaluated/updated to align with best practices related to reparative metadata, inclusive language, etc.