Skip to content
Grunddata
Get started

Reference data management: What it is and where it lives

  • Reference data

Grunddata team

Reference data is the data you use to describe and classify everything else. Currency codes. Country codes. Industry classifications. Legal entity identifiers. Most of it is defined by someone outside your organisation, and all of it is used everywhere inside it.

What reference data is

Section titled “What reference data is”

Reference data is the set of allowed values that other data points to. A loan record says the currency is SEK. The list of valid currencies is reference data. A counterparty record carries an industry code. The classification that code comes from is reference data.

Some common examples in finance:

  • Currencies. ISO 4217 sets the three-letter and three-digit codes for nearly 300 currencies. The list is amended over time by a maintenance agency, SIX Financial Information, on behalf of the Swiss standards body.
  • Industry codes. NACE is the European classification of economic activities. Eurostat used NACE Rev. 2 for European statistics from 2008 to 2024. NACE Rev. 2.1 applies from 2025 onwards.
  • Legal entity identifiers. The LEI is a 20-character code under ISO 17442. Each LEI connects to reference information about the entity, including who owns it.
  • Internal code lists. Product types, risk classes, customer segments and similar lists your own organisation defines.

How it differs from master data and transactional data

Section titled “How it differs from master data and transactional data”

It helps to think in three layers.

Transactional data records what happened. Loans, trades, payments, postings.

Master data describes the core entities the business is about. In a bank that usually means parties, products, internal organisation and geography.

Reference data classifies both. It changes less often, but when it changes, everything that uses it is affected.

Between these layers sits a fourth kind of data that fits none of them: mappings, hierarchies, limits and parameters the business adds on top. We cover that in our guide Master data, and what Grunddata is not.

Why reference data is harder than it looks

Section titled “Why reference data is harder than it looks”

It comes in versions. When NACE moved from Rev. 2 to Rev. 2.1, every system that used industry codes had to decide which version applied to which period. A report for 2024 and a report for 2025 may need different code lists, and both may need to be available at the same time.

It gets copied. Each system loads its own copy of the list. Over time the copies drift apart, and two reports that should agree do not.

It has no obvious owner. Everyone uses it, so nobody owns it. When it needs updating, it is unclear who should do it.

It changes rarely. Because changes are rare, there is seldom a process for them. When a change finally comes, nobody remembers how the last one was handled.

Where it lives today

Section titled “Where it lives today”

In most organisations we talk to, reference data lives in one of four places. A spreadsheet on SharePoint. A table that IT updates on request. A hardcoded list inside an ETL job. Or a separate copy inside each source system. Often all four at once.

What reference data management means in practice

Section titled “What reference data management means in practice”

It does not have to mean a large platform. It means being able to answer a few questions with confidence:

  • Where is the list we treat as correct?
  • Who owns it and is allowed to change it?
  • What changed, when, and who changed it?
  • Do all systems and reports read the same version?
  • What did the list look like when last quarter’s report was produced?

If those answers depend on asking a specific person, the reference data is managed by memory, not by a system.

Where Grunddata fits

Section titled “Where Grunddata fits”

Grunddata gives reference data a home next to your existing systems and the data warehouse. It does not replace them, and it is not an MDM tool.

The team that owns a list maintains it in Excel, Google Sheets or the web app. Every change is validated against the rules defined for the table, and recorded with author, timestamp and message. Systems and reports read the same data through API and OData, so a change made once reaches everything that depends on it. The full history shows which values applied at any point in time.

Where to start

Section titled “Where to start”

List the reference data your most important reports depend on. For each list, find out where it lives today and how many copies exist. Start with the one that has the most copies. That is usually where reports disagree.

Get in touch if you want to walk through where your reference data lives today.

Sources

Section titled “Sources”