Compliance first: We never take HIPAA/PHI data or personal customer data. Everything is de-identified before it leaves your control.
Home / Guidelines / Open data policy guideline 15
Open data policy guideline 15

Unique Identifiers: Open Data Policy Guideline 15

Unique Identifiers: Open Data Policy Guideline 15

Guideline 15 asks governments to use public, non-proprietary unique identifiers for entities and places across data sets, so records can be linked accurately rather than matched by inconsistent names.

Why identifiers matter

A vendor might appear as "ABC Corp" in one data set, "A.B.C. Corporation" in another and under a parent company's name in a third. Without a shared identifier, linking those records is guesswork. Unique identifiers, such as a consistent ID for each business, property, facility or contract, make it possible to join data sets accurately.

The guideline stresses that identifiers should be public and non-proprietary, so anyone can use them without paying a third party.

How policies typically phrase it

  • A requirement that published data sets include unique identifiers for the entities they reference.
  • A charge to data stewards to develop identifiers for the entities they handle most.
  • Alternatively, a lead office responsible for a citywide identifier scheme.

How often it was adopted

This is the least adopted guideline in the collection. Annotators tagged it in a single policy: Fort Collins's 2017 resolution, which calls for publishing processes that ensure data sets contain unique identifiers. Most municipal open data policies are silent on the issue, leaving identifier choices to technical standards or individual departments. That absence is one reason cross-agency analysis of local data often requires heavy cleanup.

Common pitfalls

  • Name-based matching, which breaks on spelling variants.
  • Internal IDs that change between systems or are never published.
  • Proprietary identifiers that require a license to use.

Approaches to identifier schemes

The guideline describes two broad approaches, each with trade-offs:

  • Steward-led identifiers: each data owner assigns stable IDs for the entities it manages, such as parcels, facilities or permits. This is quick to start but can produce several competing schemes.
  • A central scheme: one lead office defines identifiers for common entities across the government. This takes coordination but makes joins between departments far simpler.

In both cases, published data sets should carry the identifier as a field, the scheme should be documented, and identifiers should never be reused for a different entity. Where widely used public identifiers already exist, such as parcel numbers or standard geographic codes, reusing them avoids reinventing the wheel and connects local data to state and federal sources.

Why this matters for your business data

Consistent keys are just as important inside a company. Operational data that uses stable identifiers for locations, products and transactions can be joined, deduplicated and analyzed far more easily, which makes it more valuable to buyers. Identifiers tied to people, however, should be removed or replaced before any data is shared.

See de-identified data for how personal identifiers are handled and how to value my data for why linkable data is worth more. We never take HIPAA/PHI data or personal customer data. Everything is de-identified before it leaves your control.

Our compliance promise: We never take HIPAA/PHI data or personal customer data. Everything is de-identified before it leaves your control.

Frequently asked questions

What is a unique identifier in open data?

A stable code assigned to a specific entity, such as a property parcel, business, facility or contract, that stays the same across data sets and over time so records can be linked reliably.

Why should identifiers be non-proprietary?

If an identifier is owned by a private company, users may need a license to use it, which undermines open access. Public identifiers can be used freely by anyone.

Do unique identifiers create privacy risk?

They can when applied to individuals. Identifiers for businesses, places and contracts aid analysis; identifiers for private persons require careful protection or removal.

Policies in the index that reflect guideline 15

PlaceAdoptedInstrumentGuidelines
Seattle, WAFeb 1, 2016Administrative policy (CTO policy)14678101112
Fort Collins, COFeb 7, 2017Council resolution12345678

Sources

  1. Sunlight Foundation, Open Data Policy Guidelines (historical reference)
  2. Sunlight Foundation blog, "A little math could make identifiers a whole lot better" (2014)

Updated December 31, 2024.