Open a Common Reporting (COREP) own funds template, `C 01.00`, and the first thing that strikes an accountant trained on International Financial Reporting Standards (IFRS) is what is missing. There is no line item called "Own funds" sitting in a friendly, human-named hierarchy the way "Cash and cash equivalents" or "Property, plant and equipment" would sit in an IFRS statement of financial position. Instead there is row 010, column 010, and a code. The template looks anonymous, and a reader's first instinct is to treat that as a gap in the tooling. It is not a gap. It is a different architecture, and understanding it is the difference between reading a supervisory return and merely staring at it.
IFRS and United States Generally Accepted Accounting Principles (US-GAAP) taxonomies start from a presentation tree: a hierarchy of human-named concepts, expressed in XBRL (the eXtensible Business Reporting Language), that an accountant can walk top to bottom and recognise. The Data Point Model starts somewhere else entirely. It is a relational metamodel, closer in spirit to a database schema than a concept tree. The template you are looking at is not the source of the data model; it is a rendering of it. The row and column codes are the visible surface of a structure that lives one layer beneath, in what the DPM calls the dictionary.
The dictionary's core concepts are metrics, dimensions, domains, and domain members. A metric is the primary item, the "what is measured" of a fact: it carries a data type (monetary, decimal, string, percent, boolean), a period type (instant or duration), and a balance where one applies. A dimension qualifies a metric, narrowing it along an axis such as currency, maturity, or accounting portfolio. Each dimension draws its values from a domain, and domains come in two flavours. Explicit domains enumerate their members directly in the taxonomy, so "accounting portfolio" might offer members such as held for trading, amortised cost, or fair value through other comprehensive income (FVOCI), all fixed and known in advance. Typed domains do the opposite: they accept a filer-supplied value, such as a counterparty's Legal Entity Identifier (LEI) or a contract identifier, values that cannot be enumerated because the universe of counterparties is not closed.
A data point, the unit of meaning in a DPM return, is one metric plus a complete vector of dimension members. That vector is doing real work: it is the reason a single Financial Reporting (FINREP) fact routinely carries far more dimensional context than a comparable IFRS fact does. Where an IFRS taxonomy encodes meaning by giving each line item its own named concept, the DPM encodes the same meaning by combining a smaller set of metrics with a rich, explicit dimensional context. Once you see a cell as "metric plus dimension vector" rather than "a named line item," the anonymous-looking codes stop being an obstacle and start being a coordinate system.
That coordinate system is layered, and the layering explains where a template like `C 01.00` actually comes from. The dictionary sits at the bottom, one shared vocabulary of metrics, dimensions, domains, and members. Above it sits the framework, such as COREP, FINREP, or Solvency II. Above the framework sits a specific taxonomy release, an EBA reporting framework version or an EIOPA Solvency II taxonomy version. Within a release sit modules, which are the taxonomy entry points a filer actually loads, and within a module sit the tables themselves. Crucially, the legal templates and the filed taxonomy are tied together by that dictionary: the published Implementing Technical Standards (ITS) templates and their business logs are the modelling input, the dictionary captures every cell as metrics and dimensions, and the XBRL taxonomy, together with the annotated templates, is generated from the dictionary. The legal template and the filing vocabulary are not separately authored artefacts that happen to agree; one model sits between them, which is exactly why the codes on the template and the tags in the filing line up.
The dictionary itself is not static, and the two authorities have moved on different clocks. DPM Standard 2.0, the "DPM Refit," is a joint project of the European Banking Authority (EBA) and the European Insurance and Occupational Pensions Authority (EIOPA), but it upgraded the syntactic layer of the metamodel, how the dictionary's objects are formally expressed, while a parallel project separately enhanced the semantic glossary that defines what each object means. On the EBA side, EBA reporting framework 4.0, released December 2024, was the first framework built on DPM 2.0, and framework 4.2 concluded the transition. The tooling changed with it: DPM Studio became the development environment, the published DPM 2.0 database superseded the legacy DPM database, and DPM-XL is the formal language the EBA uses to express validation rules under DPM 2.0. The enhanced glossary also reorganises the familiar domain and member vocabulary into categories, items, and properties.
The migration reached its endpoint on the filing side too: for EBA returns, xBRL-CSV (the comma-separated-values serialisation of XBRL) became mandatory for reference dates from 31 March 2026, and xBRL-XML (the Extensible Markup Language serialisation) is no longer accepted for those dates. EIOPA has not followed the same timeline. Solvency II filings still run on DPM 1.0 in xBRL-XML, the 2.10.0 taxonomy that applies from the first quarter of 2027 keeps that classical architecture, and EIOPA has announced no CSV cutover date. A reader moving between a COREP return and a Solvency II return today is moving between two different DPM generations, not two dialects of the same one.
The Data Point Model (DPM) is a relational metamodel, not a concept tree: a dictionary of metrics, dimensions, domains, and domain members. The legal supervisory templates (published in European Union Implementing Technical Standards, ITS) are the modelling input; the dictionary captures them, and the machine-readable taxonomy in XBRL (the eXtensible Business Reporting Language), plus annotated templates, is generated from it.
A metric is the primary item: the 'what is measured' of a fact. Metrics carry data type (monetary, decimal, string, percent, boolean), period type (instant or duration), and balance where applicable.
Dimensions qualify the metric. Each dimension draws its values from a domain: explicit domains enumerate members in the taxonomy (accounting portfolio: held for trading, amortised cost, fair value through other comprehensive income (FVOCI)); typed domains accept filer-supplied values (a counterparty's Legal Entity Identifier (LEI), a contract identifier).
A data point is one metric plus a complete vector of dimension members. Templates are views over the data-point space; the dimensional context does the work that human-named concept hierarchies do in International Financial Reporting Standards (IFRS) or United States Generally Accepted Accounting Principles (US-GAAP) taxonomies.
The layering runs dictionary → framework (Common Reporting (COREP), Financial Reporting (FINREP), Solvency II) → taxonomy release (a European Banking Authority (EBA) reporting framework 4.x release, or a European Insurance and Occupational Pensions Authority (EIOPA) Solvency II taxonomy) → module (entry point) → tables.
DPM Standard 2.0 (the 'DPM Refit'), run jointly by the EBA and EIOPA, upgraded the syntactic layer of the metamodel; a parallel project enhanced the semantic glossary, which reorganises the legacy domain/member vocabulary into categories, items, and properties. EBA reporting framework 4.0 (December 2024) was the first DPM 2.0 release and framework 4.2 concluded the transition; DPM Studio became the development environment (the published DPM 2.0 database superseding the legacy DPM database), and DPM-XL is the EBA's formal validation-rules language.
ITS templates + business logs (legal inputs)
|
| modelled into
v
DPM dictionary (one shared vocabulary)
metrics dimensions domains -> members
|
| taxonomy generated from the dictionary
v
framework (COREP, FINREP, Solvency II, ...)
|
taxonomy release (EBA reporting framework 4.x,
| EIOPA Solvency II taxonomy)
v
module = entry point (e.g. COREP Own Funds, consolidated)
|
tables (C 01.00, F 01.01, S.02.01, ...)
|
v
cell = metric + dimension members = the data point3 questions. Druk op 1 tot 4 om te kiezen, Enter om te bevestigen.
Door deze quiz af te ronden, markeer je deze sectie als voltooid.
Tip: kies met 1 tot en met 4 en bevestig met Enter.