The Guide was prepared by the United Nations Office on Drugs and Crime (UNODC) together with the United Nations Statistics Division, with support from the Data for Now initiative, to help countries operationalise the UNODC Statistical Framework to Measure Corruption.

Developed in 2023, the Statistical Framework comprises 145 indicators covering corruption and related offences, preventive measures, and broader enabling conditions relevant to accountability and integrity. These include indicators on the prevalence of bribery, prosecution and conviction for corruption offences, asset recovery, transparency in public procurement, conflict-of-interest declarations, audits, and access to information.

While the Statistical Framework identifies what can be measured, the new Guide explains how to organise this work at the national level and sets out a sequence of seven steps for developing national corruption-related statistics:

1. Coordination. This step involves identifying the public institutions and other actors relevant to corruption measurement, establishing an inter-institutional task team, defining national priorities and an implementation strategy, and addressing data access and data-sharing arrangements. The Guide envisages shared leadership by the national statistical office and the anti-corruption body: the former provides statistical standards and quality assurance, while the latter ensures links to anti-corruption policy and contributes knowledge of the relevant institutional environment.

2. Data collection. Available data sources are identified for the selected priority areas and indicators. The Guide distinguishes three main groups: administrative data, corruption surveys, and complementary sources, including information produced by non-governmental organisations and research institutions. Sources are considered in terms of data availability, coverage, stability over time, accessibility for statistical use, and legal and technical constraints.

3. Data validation. The information collected is assessed for completeness, internal consistency, documentation quality, and suitability for producing the relevant indicator. A source may be considered ready for immediate use, usable with caveats, or not yet ready; in some cases, an alternative source or indicator may be needed. The authors emphasise that missing or incomplete data do not necessarily make a source unusable, provided that the limitations are understood and documented.

4. Data integration and statistical production. Validated data are transformed into coherent statistical indicators and arrangements are made for their regular production. Integration does not necessarily require a single information system: during an initial cycle, it may be sufficient to harmonise definitions and reference periods, align classifications, or coordinate the production of indicators from separate validated sources. More advanced data integration can be developed as the necessary legal, organisational, and technical conditions become available.

5. Analysis and interpretation. Indicators are analysed with regard to what they actually measure, the characteristics of the underlying sources, denominators and units of observation, changes over time, and differences in legal and institutional settings. The Guide cautions against conclusions that go beyond what the available evidence can support. For example, a change in the number of detected corruption offences may reflect not only changes in the prevalence of corruption, but also changes in detection, recording, or enforcement practices.

6. Dissemination. Statistics should be presented in ways that make them useful to public institutions and other stakeholders, while taking into account the sensitivity of certain indicators and clearly communicating their methodology and limitations. The Guide provides for different formats for different audiences and recommends establishing clear procedures for publication and documentation.

7. Monitoring and sustainability. Following the first cycle, countries are encouraged to review how the measurement system is functioning, revise, add, or retire indicators where necessary, improve data sources, and gradually broaden the scope of measurement. Inter-institutional coordination should evolve from arrangements for an individual measurement cycle into sustained collaboration, while data sharing and statistical production should become regular processes.

Countries are not expected to produce all 145 indicators at once. The authors recommend starting with a limited set that reflects national priorities and available capacities and then gradually expanding the measurement system.

The Guide stresses that no single source of information can provide a complete picture of corruption. Administrative data on detected offences, investigations, and prosecutions, for example, depend not only on the prevalence of corruption but also on the activities of law enforcement and oversight institutions. Corruption measurement should therefore draw on a combination of sources and indicators. The Guide is not intended to produce international corruption rankings or country scores.

A draft of the Guide was presented in March 2026 and opened for global consultation. Its development was shaped primarily by practical experience in applying the Statistical Framework in Colombia, the Dominican Republic, and Kenya, while also drawing on exchanges and lessons from other settings, including Brazil and Mexico. In the three pilot countries, UNODC and the United Nations Statistics Division worked with national counterparts on identifying priority indicators, assigning responsibilities for their production, and using different data sources. In Colombia, for example, 125 of the 145 indicators were identified as potentially measurable, of which 49 were selected as priorities; by April 2026, the first 15 indicators had been produced. In September, the United Nations presented the Guide as a new practical tool for strengthening national corruption-related statistics.

On 23 September, the U4 Anti-Corruption Resource Centre also published a piece on operationalising the Statistical Framework. In it, experts involved in work with the pilot countries and in developing the Guide discuss the experiences of Colombia, the Dominican Republic, and Kenya and the key lessons emerging from piloting the approach.