High-value Datasets Best

Practices Report

1
This study has been prepared as part of data.europa.eu. Data.europa.eu is an initiative of the
European Commission. The Publications Office of the European Union is responsible for contract
management of data.europa.eu.

For more information about this study, please contact the following:

European Commission
Directorate-General for Communications Networks, Content and Technology
Unit G.1 Data Policy and Innovation
Email: [email protected]

data.europa.eu
Email: [email protected]

Authors:
Martina Ramponi
Maddalena Moro
Jolanta Suárez

Edited by:
Beatriz Fernández Nebreda
Nataliya Rozbroj Jasinskaja

Last update: 12 June 2024

https://data.europa.eu/

DISCLAIMER
The information and views set out in this publication are those of the author(s) and do not necessarily
reflect the official opinion of the Commission. The Commission does not guarantee the accuracy of the
data included in this study. Neither the Commission nor any person acting on the Commission’s behalf
may be held responsible for the use that may be made of the information contained therein.

Luxembourg: Publications Office of the European Union, 2024

© European Union, 2024

The reuse policy of European Commission documents is implemented by Commission Decision

2011/833/EU of 12 December 2011 on the reuse of Commission documents (OJ L 330, 14.12.2011,
p. 39, ELI: http://data.europa.eu/eli/dec/2011/833/oj). Unless otherwise noted, the reuse of this
document is authorised under a Creative Commons Attribution 4.0 International (CC-BY 4.0) licence
(https://creativecommons.org/licences/by/4.0/). This means that reuse is allowed, provided
appropriate credit is given and any changes are indicated.

ISBN: 978-92-78-44203-3 DOI:10.2830/113131 OA-02-24-637-EN-N

2
Contents
Executive summary ........................................................................................................... 4
1. Introduction ............................................................................................................... 5
1.1. Context ......................................................................................................................... 5
1.2. Objective of the high-value datasets best practices study ............................................... 5
1.3. Method for data collection ............................................................................................ 5
2. Establishing the legislative framework ....................................................................... 6
2.1. Short overview of the varying legislative landscape in the EU ......................................... 6
2.2. Analysis of the regulation’s requirements ...................................................................... 6
3. High-value data – State of play at a glance ................................................................ 7
3.1. Czechia .......................................................................................................................... 9
3.2. Denmark ..................................................................................................................... 10
3.3. Estonia ........................................................................................................................ 11
3.4. Italy............................................................................................................................. 11
3.5. The Netherlands .......................................................................................................... 12
3.6. Romania ...................................................................................................................... 13
3.7. Finland ........................................................................................................................ 14
4. Empirical findings..................................................................................................... 14
4.1. Best practices .............................................................................................................. 14
Policy and legal framework .......................................................................................................................... 16
Governance and processes........................................................................................................................... 16
Technical aspects, metadata quality and new requirements ....................................................................... 18

5. Remaining challenges .............................................................................................. 20

5.1. High-value dataset implementation roadmap............................................................... 21
6. Conclusion ............................................................................................................... 23

3
Executive summary

This study delves into Commission Implementing Regulation (EU) 2023/138 , which complements the
open data directive (Directive (EU) 2019/1024) by setting specific requirements for high-value
datasets. These datasets provide significant social and economic benefits, enhancing public sector data
accessibility. As the regulation came into force on 9 June 2024, this report aims to highlight best
practices and strategies employed by EU Member States (MSs) in implementing these requirements.

A key component of the study’s methodology involved conducting interviews with representatives
from seven MSs (1): Czechia, Denmark, Estonia, Italy, the Netherlands, Romania and Finland. These
interviews offered insights into their current compliance status, challenges faced and strategies
employed in implementing the HVD requirements. The report highlights the best practices and
recommendations provided by the mentioned MSs, addressing legal, organisational and technical
aspects.

Additionally, the study identifies ongoing common challenges across MSs, including data protection
concerns and issues with licencing ambiguity. It also highlights the areas where support at the EU level
is necessary in areas related to data interoperability, timely access to information and clarification of
the overlapping requirements with other legislation, such as the Inspire directive, as well as
requirements that may sometimes be contradictory in other regulations, namely the general data
protection regulation.

The study also includes a suggested roadmap for MSs, outlining steps to successfully implement the
new HVD requirements. It is designed to guide MSs towards compliance with the HVD implementing
regulation, promoting a more transparent and data-driven EU landscape.

Through its practical and actionable insights, the report aims to facilitate the implementation of HVD
requirements across the EU, leading to improved public sector efficiency, increased economic
opportunities through data reuse for all kind of users and strengthened cross-border collaboration
within the EU.

The key findings are the following:

• the degree of adoption of the regulation varies among MSs;

• in specific domains, such as cadastral data, business registries and meteorological data, challenges
have been identified in making HVDs available free of charge;
• the main obstacles have been encountered in the use of the EU vocabularies, APIs and bulk
downloads;
• further guidance is needed with regard to the overlap between the requirements of the regulation
and the Inspire directive in relation to the quality of the datasets (description, documentation and
data catalogue vocabulary;
• cooperation is encouraged, among MSs to share effective practices and with the European
Commission for direction to overcome existing challenges.

(1) The MSs are listed according to EU protocol order.

4
 1. Introduction
1.1. Context
In an era defined by digital transformation, the European Union has taken significant steps to enhance
the potential of data in the public sector through the open data directive (Directive (EU) 2019/1024).
This directive marked a significant milestone in promoting transparency and innovation by facilitating
access to public sector data, including geographical, statistical, legal and publicly funded research data.
Within this directive, the concept of high-value datasets (HVDs) emerged, recognising that certain
datasets hold substantial social and economic benefits and that their use is particularly valuable in the
roll-out of improved digital services.

On 21 December 2022, the European Commission introduced Commission Implementing Regulation

(EU) 2023/138, outlining further conditions for the publication and reuse of HVDs. This regulation aims
to harmonise the reuse of HVDs across EU Member States (MSs), defining conditions for their
publication and outlining concrete datasets falling under the six categories of HVDs: geospatial, earth
observation and environment, meteorological, statistics, companies and company ownership, and
mobility.

The availability and accessibility of HVDs play an important role in enabling effective decision-making,
fostering research and driving innovation in today’s interconnected world. An initial overview of the
progress MSs have made in implementing the requirements of this new legislation was presented in
the 2023 Open Data Maturity Report.

1.2. Objective of the high-value datasets best practices study

The objective of this study is to assist MSs in the process of implementing the HVDs by presenting a
collection of effective best practices that can be adopted to accelerate compliance. It also aims to
outline the common challenges that need to be addressed at both national and EU levels to facilitate
HVD implementation across the EU. Finally, this study also offers a practical roadmap informed by the
successful practices in the leading MSs to guide and assist other MSs with compliance.

1.3. Method for data collection

The methodology employed in this study followed a three-step approach. Each step aimed at gathering
comprehensive insights into the challenges, compliance needs and successful strategies related to HVD
implementation across MSs.

Step 1: Analysis

We began with a content analysis of the implementing regulation on HVDs, involving textual
interpretation and desktop research of existing analyses and reviews. The objective of this analysis was
to identify key themes, patterns and trends that would inform the development of questionnaires on
the status of HVD implementation in countries with potential for growth and in the top-performing
MSs.

Step 2: Qualitative interviews and best practices research

Next, we conducted qualitative 1-hour interviews with data representatives from selected MSs, as
mentioned above: Czechia, Denmark, Estonia, Italy, the Netherlands, Romania and Finland.

5
The interview questionnaire focused on the implementation of the requirements outlined in the
regulation, exploring challenges faced by all participating countries, successful strategies employed to
address them (with a special focus on the top-performing MSs) and concluding with a summary
overview of each MS.

Step 3: Actionable recommendations development

Finally, we developed actionable recommendations by synthesising best practices observed in the

leading MSs. These recommendations were translated into practical guidelines that can be adapted to
different national contexts by other MSs, facilitating the effective implementation of the regulation.
Furthermore, the report includes MSs’ requests for the Commission related to unresolved issues,
hence opening the dialogue for further cooperation and assistance, along with links to resources the
Commission has made available to address some of the challenges.

2. Establishing the legislative framework

2.1. Short overview of the varying legislative landscape in the EU
The enforcement of the regulation is influenced by previous EU legislation, notably the open data
directive and Inspire directive. These directives have laid the groundwork for HVD implementation by
establishing frameworks for the reuse of public sector information and promoting the sharing of
geospatial data for environmental policies and initiatives.

Previous legislation paved the way for shaping the framework for HVDs: the public sector information
directives established principles for the reuse of public sector information, aligning with the objectives
of the HVD regulation to enhance the accessibility and reuse of valuable datasets. Additionally, the
general data protection regulation (GDPR) set standards for data protection and privacy, ensuring that
HVDs adhere to stringent legal requirements to safeguard individuals’ rights and personal data.

2.2. Analysis of the regulation’s requirements

To begin with, it is worth noting that unlike the open data directive and Inspire, which require national
transposition, the HVD implementing regulation is directly applicable in MSs. Building upon the open
data directive, this regulation introduces significant changes to public data management practices
encompassing various aspects, including guidance on access and charges, technical formats and
standards, and arrangements for data reuse.

This regulation also introduces a specific list of HVDs, categorised into six thematic groups regulated
in the Annex, which provides the specific description and requirements the HVDs must adhere to. This
list includes geospatial data, earth observation and environment, meteorological, statistics, companies
and company ownership, and mobility. By delineating these thematic groups, the implementing act
provides clarity on the scope and coverage of HVDs, facilitating their identification and publication by
MSs.

HVDs are required to be downloadable in bulk – where relevant – and through application
programming interfaces (APIs), while also providing extensive documentation for their metadata
(Article 3). In terms of access and charges, the regulation offers more detailed guidance compared to
the open data directive. While the directive established the general rule that HVDs should be available
free of charge, the regulation provides further clarification on any permissible exceptions to this

6
rule (2), ensuring transparency and consistency in data access policies. Moreover, with a few
exceptions, HVDs are characterised by specific technical and legal requirements, including the open
data licence, the availability of public documentation and the ensuring of machine readability
(Article 4).

Additionally, the regulation clarifies any specific conditions for the reuse of HVDs, ensuring that
relevant stakeholders are aware of any restrictions or limitations associated with the reuse of these
datasets. By providing clarity on reusability arrangements, the regulation promotes transparency and
encourages the maximum utilisation of valuable data resources.

3. High-value data – State of play at a glance

In this section, we present the state of the implementation of HVD requirements in the MSs that were
interviewed for this report, as of April 2024. The MSs in this study are listed according to EU protocol
order. The summary of the state of play of the participating MSs is presented in Figure 1: High-level
overview of the implementation of the regulation in the interviewed MSs.

(2) Recital 6 of the HVD implementing regulation: ‘Pursuant to Directive (EU) 2019/1024, the requirement to make high-value datasets
available free of charge shall not apply to libraries, including university libraries, museums and archives. Member States may exempt
individual public sector bodies, upon their request and in line with the criteria set out in the Directive from the requirement to make
high-value datasets available free of charge for a period not exceeding two years from the date of the entry into force of this
Implementing Regulation.’

7
8
 Figure 1: High-level overview of the implementation of the regulation in the interviewed MSs.

3.1. Czechia
As per the 2023 Open Data Maturity Report, Czechia is on track for the timely adoption of the
regulation. The quick turnaround for the implementation was made possible by an early delineating
of the relevant stakeholders (data providers and data owners), with clear responsibilities in each of
the regulation categories. This early engagement with the main stakeholders set the stage for
streamlined data handling and resulted in successful integration of the requirements for HVD
management into Czechia’s national legislation, strengthening the policy framework and expediting its
journey towards compliance. This measure has proven beneficial in facilitating the reporting processes
to the government, where data providers are required to demonstrate compliance with both national
and EU open data directives.

Metadata standards applicable to HVDs have been published and distributed among the Czech data
providers, enabling appropriate and timely labelling of data. HVDs in Czechia were published on the
national portal and open data catalogue in May 2024 and will be fully accessible to the public by June
2024. The portal will offer an extended set of search and filter features that are in line with EU
standards and even go beyond what is required by the regulation to promote the accessibility and
reuse of HVDs.

9
Czechia also highlighted its successful collaboration with Slovakia as a contributing factor to its
compliance with the regulation. Both countries use the same contractor for their national data
catalogues, which has facilitated easier and more efficient resolution of technical issues in both
countries.

While Czechia has made significant progress, the journey has not been without challenges. Gaps in
understanding the detailed content of various datasets, especially those under Inspire categories that
are managed by different government authorities, have complicated efforts to enable interoperability
at the metadata level. In addition, compliance within the meteorological category poses the biggest
challenge due to the cost of handling large amounts of data and a drastic change in the national
agency’s funding model needing to move from a paid to an open data-sharing model. Furthermore,
Czechia underscores the importance of ensuring that EU legislators consistently provide timely and
clear communication. This is essential for national coordinators and data providers to adequately
prepare for the implementation of the regulation. Finally, there is an ongoing need to enhance clarity
regarding reporting formats and requirements, ensuring that all providers have a comprehensive
understanding of their compliance obligations.

3.2. Denmark
In Denmark, the process of implementing the regulation has been rather smooth, thanks to its long
tradition of digital data management and transparency in the public sector enabled by robust digital
infrastructure. A holistic approach to understanding the big picture and focusing on the value-adding
aspects first helped Denmark to assess existing solutions, map them against the regulation, identify
gaps and in doing so create conditions for its successful implementation. This involved a considerable
degree of coordination with the government authorities to address both technical challenges and
their operational needs. Denmark also coordinated with and explored best practices from several
Nordic neighbours and Ireland but opted for their own solution for data catalogues rather than
replicating existing models in partner countries.

Denmark’s strategy has focused on effective coordination and management of data across different
governmental departments, ensuring that each entity knows its specific role and responsibilities
concerning HVDs. While broader communications have occurred in the form of a reference group, the
main interaction with data providers is bilateral. This works well because the number of authorities
is reasonably small in Denmark, and a more targeted collaboration ensures a more efficient resolution
of issues that differ based on each authority’s unique conditions and technical solutions.

Another integral part of Denmark’s strategy concerns direct responsibility of data authorities in
marking their data as high-value. This method was chosen over a centralised approach to ensure that
data publishers take responsibility for the metadata of their datasets. By making individual authorities
responsible for labelling their data as high-value at the source, Denmark also ensures that the
metadata management remains accurate and up to date on the national portal. Consequently, this
minimises discrepancies between the national portal and the source that could potentially confuse
users, reduces the burden of metadata maintenance at the central level and establishes a sustainable
practice that ensures the portal’s long-term value and usefulness.

Finally, Denmark’s approach to handling HVDs has been a phased process. The focus in the beginning
was on understanding the implications of the legislation and identifying which datasets are impacted
by it. Currently, the effort is shifting towards ensuring the metadata is accurately attributed, grouping
authorities based on how they deliver metadata to the national portal and addressing discrepancies

10
on a case-by-case basis. Since the start, Denmark’s focus has been on data management aspects of
HVDs, including documentation and accessibility of datasets, rather than promotion of their use
through improved search or filtering functionalities on the national portal’s front-end interface.
Similarly, monitoring reuse is not among Denmark’s top priorities.

In terms of challenges, similarly to other countries, Denmark has encountered difficulties in

transitioning from the Inspire requirements to implementing the HVD regulation due to a lack of a
common EU-wide solution. Aligning the new regulatory requirements with the pragmatic, demand-
driven national approach has also presented challenges. As a standard prevailing practice, Denmark
only develops new interfaces and features (such as APIs for bulk downloads) when they are necessary
to meet the needs of the community.

3.3. Estonia
Estonia is currently the front runner in the implementation of the HVD regulation in Europe, as per the
2023 Open Data Maturity Report. Its successful adoption has been made possible by Estonia’s
proactive and timely measures capitalising on both the country’s advanced digital data management
capabilities and its digital infrastructure. Adoption of the Data Catalogue vocabulary Application
Profile (DCAT-AP) as a common standard for metadata in Estonia, with controlled vocabularies being
adopted as early as 2018, is a great example of the strategic measures taken to achieve compliance.
This has led to one of Estonia’s biggest achievements – the rapid transition of the business registry
from a paid to a fully open data service in 2020. The last category to be made regulation-compliant
was meteorological data at the end of 2023, for which aspects such as granularity required
enhancement to meet the new standards.

The legal and policy framework in Estonia is designed to facilitate the publication and accessibility
of HVDs, although the portal itself is still pending features that label HVDs as such to promote their
reuse. The framework allows for the introduction of new categories of HVDs beyond the categories
currently defined by the regulation, such as language data resources.

Estonia’s methodical approach involves regular and detailed discussions within its open data working
group, which now includes nearly 600 members from various government sectors. This group is
instrumental in ensuring that EU updates are efficiently communicated and that discussions about data
category inclusion or expansion are timely and relevant. Estonia has previously also collaborated with
the Netherlands on personal data handling, an important topic to address when opening Estonia’s
business registry to the public.

While progress in achieving compliance with the regulation is remarkable, Estonia’s biggest challenge
was accessing Creative Commons (CC) licensing terms in Estonian. As there was no translation available
at the time, Estonia provided the translation themselves, but getting it validated by the Commission
has been a challenge. Monitoring reuse and identifying data (re)users have also posed obstacles, which
Estonia has been actively working on overcoming since last year. Good progress has been made in
monitoring business registry data use in research.

3.4. Italy
In Italy, most HVDs are already hosted on the national open data portal but are not yet specifically
labelled as such. Efforts are under way to make some datasets that were previously behind a paywall
freely available from June 2024, with the objective to align Italy’s practices with other MSs. Efforts to
synchronise metadata formats across portals are pending updates from the Inspire directive’s working

11
group and necessary technical upgrades to the Comprehensive Knowledge Archive Network system to
accommodate HVD metadata standards.

Italy’s progress in the implementation of the regulation is challenged by the distribution of

responsibilities over the different categories of datasets spread out across many public
administrations. To coordinate the efforts, Italy has established a working group that published a
guiding document in December 2023 to streamline the approach towards the implementation of the
regulation across the country. The second version of this document is currently under revision. The
working group’s efforts rest on Italy’s information and communications technology plan and the
transposition act of the open data directive. They form the core of the legal and policy framework in
Italy and outline the need for specific technical guidelines for HVDs.

Another significant challenge for Italy has been privacy protection in the category of company data,
hampering the sharing of existing data, such as company addresses and accounting data. The main
concern is that making company data widely available as open data could potentially distort market
competition in Europe if not all countries publish analogous data. Efforts are under way to tackle these
issues from both legal and technical perspectives in the revised version of the guiding document (3).

Furthermore, Italy faces the issue of limited resources for the implementation of the regulation. While
there is a general willingness to advance towards compliance, there is also resistance at the regional
level fuelled by the scarcity of financial and human resources in places outside of the large
metropolitan areas. In addition, technical challenges such as bulk downloads of large volumes of data
pose issues for the optimal performance of the platform operations. This has prompted Italy to
consider partial dataset downloads to mitigate negative impacts on the portal performance. Finally,
the varying degrees of technical readiness across ministries, each responsible for datasets under the
different categories of the regulation, add complexity to compliance at the national level.

Lastly, Italy also faces challenges with cadastral data, which is fee-based until the end of 2024, as per
the current contract with the data owner. Similar to company data, opening cadastral data to the
public from 2025 onwards will require a change in the current business model. Italy is exploring
alternative revenue streams to reduce the impact on its current administrative and business practices.

3.5. The Netherlands

The Dutch approach towards implementing the regulation began by engaging with Inspire data
providers to assess the additional efforts required from them to comply with the regulation, such as
adapting APIs and conducting impact analyses. Most data providers selected compliance strategies
that would have the least impact on their operations. This preliminary work included knowledge
sharing and the development of guidelines to assist data providers in understanding and implementing
HVD requirements.

However, the Dutch national strategy does not currently include specific measures to incentivise the
publication of and access to HVDs beyond a general digital agenda that promotes public data as open
data. This is partly because the Netherlands is still catching up with the broader open data legislation,
and HVD-specific policies are relatively new. Nevertheless, efforts are under way to align HVD

(3) Recital (8) of the HVD implementing regulation: ‘Member States should make use of appropriate methods and techniques (such as
generalisation, aggregation, suppression, anonymisation, differential privacy or randomisation), thus making as much data as possible
available for re-use.’

12
implementation with existing digital agendas and sector-specific programmes and to update the
governance structure.

Several additional challenges are impacting the Netherlands’ progress in becoming compliant with the
regulation. First, there is a lack of consistency in the technical set-up of APIs and bulk downloads across
different data holders. Many organisations are not in favour of putting resources into the
implementation of the DCAT-AP 2.0 standard while the release of the DCAT-AP 3.0 standard is
forthcoming. Also, not all data is available via APIs, and not all organisations have the technical and
financial capacity to support this infrastructure. Furthermore, the HVD regulation might present a
potential conflict with the GDPR (4) and prioritising one can lead to non-compliance with the other. As
mentioned above, following the HVD regulation, the focus is on using appropriate methods and
techniques to protect personal data. However, in terms of business impacts, there is a concern and
hesitation about making HVDs freely available, especially for data providers like the cadastre, which
previously relied heavily on fees.

While HVDs are published through the national open data portal, these datasets are not currently
identified as such on the portal and are not marked according to the DCAT-AP guidelines for HVDs. As
a result, they do not yet align with legislative definitions.

3.6. Romania
In Romania, the implementation of the regulation is still in its early stages and faces several challenges.
This can largely be attributed to Romania’s emerging open data culture and its initial steps toward
developing a national open data strategy. Efforts to map the data providers and identify responsible
institutions have commenced but are challenged by recent restructuring at various governmental
levels, leading to unclear responsibilities across different agencies.

Despite these difficulties, Romania has established working groups and crafted a guiding document
for data providers to align with the HVD regulation on legal, technical and semantic aspects. Inventory
models to streamline the approach across agencies have also been shared with the data owners as a
first step in the process, although actual data publication plans are still under development.

Not all government data is aligned with the technical requirements of the regulation, though some
datasets from the cadastre and trade registry are being considered for harvesting on the national
portal. The granularity of these datasets, however, is not yet in line with the legislation, and the API is
unavailable.

Efforts to categorise datasets based on the six groups outlined in the regulation are ongoing to advance
compliance with the regulation. In the future, to improve the accessibility of HVDs, Romania has plans
to enhance the search functionality on the national portal, which requires a major upgrade. This,
unfortunately, is impacted by the funding constraints. An annual reporting system is also being
established to monitor and measure data reuse, incorporating standardised documents for institutions
to follow.

In addition, resource limitations pose significant issues, in particular those related to technical skills.
The lack of infrastructure and skills impedes the ability to make datasets available in the desired data
quality and through APIs or bulk downloads. Furthermore, institutions such as the cadastre, which has

(4) Recital (8) of the HVD implementing regulation: ‘Member States should make use of appropriate methods and techniques (such as
generalisation, aggregation, suppression, anonymisation, differential privacy or randomisation), thus making as much data as possible
available for re-use.’

13
historically funded as much as 90 % of its operations through data sales, are now forced to look for
alternative means of financing, creating internal resistance to the change and resulting in slow
progress. Government departments are helping to look for solutions collaboratively, with some data
categories like meteorological data moving towards free access following internal impact assessments
and exploration of alternative funding models.

3.7. Finland
Similarly to Denmark, the implementation of the HVD regulation in Finland has capitalised on the
country’s strong foundation in open data practices. Finland’s overall compliance with HVD
requirements has been relatively straightforward because most of their datasets were already
available and free of charge. The primary challenge was aligning the Inspire directive with the HVD
regulation, especially in managing metadata that needed to meet both Inspire and HVD requirements.

Finland’s solution to the metadata challenge was to incorporate HVD specifications within the Inspire
descriptions to ensure compliance without duplicating efforts. This approach reflects Finland’s
broader strategy of utilising existing frameworks and adapting them to new regulations, leveraging a
cultural and institutional predisposition towards data openness and transparency. This transparency is
also made possible thanks to the country being relatively small both in size and population, which
facilitates easier management and dissemination of public data.

Regarding the legal and policy framework, Finland has taken significant steps in recent years to
enhance the accessibility and use of government data. In 2022, the government took a crucial decision
to open government data further, backed by legal frameworks and an information management act
that reinforces transparency at the constitutional level. Two active working groups, one for data
holders and an inter-ministerial coordinating group for legal matters, ensure that practical, technical
and legal aspects of HVD implementation are continuously addressed and monitored.

While other countries mentioned resource scarcity, budgetary constraints have not hindered Finland’s
HVD strategy, thanks to the pre-existing financial allocation and the general government budget for
these needs. However, challenges persist with cadastral data in some municipalities that require
updates to meet the granularity standards of the regulation. Local authorities have until 2029 to
comply, a timeline that reflects the degree of difficulty in upgrading certain datasets to the required
digital standard.

As for the datasets themselves, apart from company data, which only became available as from 9 June
2024, most adjustments have been minimal. In Finland, a DCAT-AP format is used to ensure
interoperability between HVD categories, and a comprehensive catalogue of HVDs is maintained by
the coordinating working group. Finland does not actively monitor the reuse of open data, adhering to
the policy that the public can use the data as they see fit.

4. Empirical findings
4.1. Best practices
In this section, we present examples of best practices gleaned from MSs, showcasing effective
strategies and lessons learned. These examples highlight what has worked well for those countries and
provide a valuable context for understanding the successful approaches taken by various MSs
regarding the main challenges. A high-level overview of the best practices is provided in Figure 2: Best
practices.

14
Figure 2: Best practices.

15
Policy and legal framework
Encourage a governmental culture that prioritises transparency in data management.

This best practice underscores the fundamental relationship between open data and cultural values
ingrained within governmental systems. Across several MSs, such as Finland and Denmark,
transparency towards citizens is not merely a contemporary trend but a deeply rooted tradition
reflected in constitutional frameworks. By recognising this cultural foundation, governments can
better understand the intrinsic importance of open data initiatives, which not only align with
constitutional principles but also serve as a commitment to fostering transparency and accountability
in governance. Therefore, efforts to promote open data should be framed within the broader context
of cultural values and historical precedents, reinforcing the notion that transparency is not just a policy
but a core societal value.

Embrace a wider perspective and adopt a comprehensive approach.

As a result of the ongoing process of implementing the open data directive and the Inspire directive,
MSs face challenges in simultaneously complying with the HVD implementing regulation. Some MSs,
such as Denmark and Finland, had pre-existing legislation in place, necessitating only minor
adjustments to align with the national transposition of the open data directive. Estonia incorporated
HVD provisions into its existing open data legislation, specifically the Public Information Act. As a result,
the implementation of previous regulations is nearing completion, facilitating the transition to meeting
the new requirements.

For MSs encountering challenges in this process, a strategic approach based on a wider perspective of
the regulation, such as focusing on implementation of standards like DCAT-AP in the public sector and
publishing the data, is advised. This approach prioritises sustainable solutions over quick fixes, ensuring
long-term effectiveness and compliance.

Practice prioritisation and focus on achievable goals.

Prioritising and focusing on achievable goals are crucial, especially when faced with numerous tasks
and limited resources. By identifying what is most important and feasible, organisations can streamline
their efforts and maximise efficiency. For example, when implementing HVDs, it is prudent to prioritise
categories that overlap with existing initiatives, such as Inspire. As Czechia and Denmark pointed out,
since three of the HVD regulation categories also fall under Inspire, concentrating efforts on these
areas allows for progress in the implementation of both pieces of legislation simultaneously, optimising
resources and expediting progress. This ensures that valuable resources are allocated effectively,
leading to a more efficient and timely achievement of compliance objectives.

Governance and processes

Engage in strategic partnerships and foster collaboration with key stakeholders at the national level.

Establishing collaborative networks and partnerships with key stakeholders at the national level is
crucial for effective implementation of the HVD regulation. This involves coordinating efforts among
various ministries and agencies responsible for different categories of HVDs, especially in MSs with
decentralised governance structures.

To address challenges related to the diverse pace of progress across HVD categories, creating
interdisciplinary working groups comprising relevant ministries, agencies and stakeholders is

16
recommended. These working groups can build upon existing structures established for previous
directives, ensuring continuity and leveraging prior collaboration experiences.

For instance, Denmark has a reference group for the open data directive and for the Inspire directive,
where they also address the HVDs. Finland has also established two working groups to address the
implementation of HVDs. The first group, which is a coordinating group of data holders, was active
until 9 June 2024 and focuses on practical and technical questions. This group shares best practices
and resolves practical issues. As legal experts, they also provide guidance on understanding the HVD
regulation from a legal perspective. The second group is an inter-ministerial coordinating group, a legal
working group ensuring there are no conflicts or overlaps between the HVD regulation and the national
legislation. Establishing such a network is crucial, as certain MSs currently lack a comprehensive
overview of available datasets. Such a network can enable countries to conduct thorough stakeholder
analysis, facilitating a comprehensive data inventory and clarifying which agency is responsible for each
dataset.

Coordinate with other Member States.

Engaging in regular exchanges with other MSs, facilitated by events like the 2023 data.europa.eu
workshop for data providers in Copenhagen, Denmark, enables the sharing of ideas and solutions to
common challenges. The ability to collectively address issues not only fosters innovation but also
addresses challenges such as the shortage of IT skills, as observed in countries like Czechia, Italy and
Romania. Workshops like this provide a platform for the MSs with similar issues or needs to connect
with each other and continue bilateral engagements beyond the event. The collaboration between
Czechia and Slovakia is an inspiring success story of effective bilateral cooperation.

Promote sustainability through individual agency responsibility for its respective datasets.

Ensuring the sustainability of the national data portals means making sure that the metadata is
maintained with the resources available. In order achieve that, some MSs, like Czechia and Denmark,
entrust each agency with the responsibility of maintaining and managing their own datasets. In this
way, the national portal only plays a monitoring role, ensuring that data publishers mark and maintain
the data as agreed.

This approach is recommended for all lead authorities that seek a robust and sustainable model for
maintaining metadata over time. It is crucial because data portals require regular maintenance to
prevent them from becoming outdated quickly and to keep their relevance and value to businesses
and society. Individual agencies’ responsibility aids in inventory management, as data providers are
accountable for updating the inventory when new datasets are published or existing ones updated on
the portal.

Promote awareness among stakeholders, data holders and users through knowledge sharing.

Facilitating bilateral communication with data publishers is a crucial aspect of ensuring effective
implementation of guidelines, particularly when certain categories may not align perfectly with
established standards. For instance, Estonia provided awareness training sessions for stakeholders on
topics such as open data and privacy.

Similarly, Czechia held two webinars dedicated to HVDs. The first was in cooperation with the Ministry
of Environment, focusing primarily on the relationship between Inspire and HVDs. Following the
publication of the DCAT-AP HVD standard, a second webinar for all data providers and users took place

17
to disseminate the information received from the Directorate-General for Communications Networks,
Content and Technology.

By providing a platform for interactive communication and knowledge-sharing, these webinars serve
as valuable tools for ensuring that stakeholders are well informed and equipped to meet their
obligations effectively. Moreover, such engagement initiatives foster a sense of community and shared
responsibility among stakeholders, contributing to the overall success of data governance efforts.

Use open data to address challenges and find solutions.

Consider using crowdsourcing platforms to involve citizens in solving real-world challenges such as
enhancing data quality. Initiatives like Estonia’s Kaggle competition invite participation from those
with diverse skill sets, offering an opportunity for individuals, regardless of their IT proficiency, to
contribute innovative ideas and problem-solving approaches. In addition, leveraging such platforms
can help bridge the gap in IT skills, enabling countries like Romania and Czechia to tap into a broader
pool of talent and expertise.

Establish a robust framework for data governance.

Develop a comprehensive data governance framework by first assessing available resources, including
technical expertise, data management tools and key stakeholder inputs. This assessment process
enables a clear understanding of the rules, processes and responsibilities necessary for effective data
governance implementation. For example, Estonia mapped out existing roles and identified specific
actions needed across various domains, from technology to data management. This thorough mapping
ensured that all aspects of data governance were addressed, facilitating a more cohesive and efficient
implementation.

Technical aspects, metadata quality and new requirements

Prioritise effective data management.

Develop a comprehensive understanding of the specific requirements for HVDs. This involves
identifying existing datasets to determining their compliance with the standards outlined for HVDs. For
example in Denmark, a reference group created an inventory of datasets. The Agency for Digital
Government initiated this process, which was then accepted and approved. They meticulously
described each dataset, including which authority was responsible and where to harvest it, ensuring
no gaps were left. Similarly, Finland uses a spreadsheet containing all necessary elements for a
catalogue of HVDs, managed by the coordinating working group. By identifying areas where datasets
fall short of meeting these requirements, organisations can establish a roadmap to address gaps and
ensure full compliance over time. This process provides a systematic basis for improving data quality
and accessibility and enhances the overall value of the HVDs for stakeholders and end users.

Enhance the quality of metadata directly at the data source before publishing it on the portal.

As previously mentioned, managing metadata at the source before publication on the portal offers
several benefits. Firstly, it places the responsibility for maintenance on data providers rather than on
the portal itself, as Denmark has successfully demonstrated. Additionally, this approach mitigates the
risk of confusion among users caused by discrepancies in metadata between the national portal and
the data source.

18
To enhance the quality of metadata, using the DCAT-AP high-value dataset guidelines, the controlled
vocabularies for the six HVD categories and the controlled vocabulary for the subcategories, which is
still ongoing, is recommended. This is also in alignment with the Inspire legislation, as the work on
GeoDCAT allows Inspire metadata to be mapped to DCAT-AP. Merely relying on the Inspire team to
denote metadata may not suffice, as noted by Italy, indicating the need for additional measures
beyond the DCAT-AP document.

Top-performing MSs are integrating DCAT-AP and Inspire requirements into the metadata practices
adapted to their national context. For instance, Czechia has developed specifications for local open
data catalogues to ensure compatibility with the National Catalogue of Open Data. These specifications
adhere to the DCAT-AP 3.0 standard and encompass both SPARQL endpoints and file-based record
storage. Similarly, Denmark provides guidance within the DCAT framework, outlining essential fields
to prioritise for effective metadata description.

Strengthen accessibility of high-value datasets through external funding or strategic planning.

The regulation mandates that all HVDs must be accessible free of charge; some Member States
diversify funding sources by seeking financial support from external avenues such as fundraising
initiatives and participation in EU projects, as exemplified by Czechia. In Denmark, the implementation
of a basic data program, coupled with strategic planning outlined in the Free Data initiative, has
facilitated the transition towards compliance with the regulation, particularly in sectors like
meteorology.

Adapt the business model gradually to provide free data.

Monetisation strategies play a crucial role in the adoption of the regulation, particularly concerning
company data and cadastral data, where fees are often mandated. A good example is Italy, where the
current contractual obligations to keep cadastral data behind a paywall are in force until the end of
2024, although historically this has been a common business model across most MSs.

Denmark’s experience transforming their meteorological data business model through the initiative
called Free Data, one of the main initiatives in Denmark’s digital growth strategy, is a noteworthy
success story. This transition for the Danish Meteorological Institute was well planned, divided into six
phases between 2020 and 2023 to ensure a gradual and effective shift. In the first phase,
meteorological observation data for measurements like temperature, rainfall and wind from the
Danish Meteorological Institute’s network of weather stations in Denmark and Greenland was made
public. Later in 2020, data on water levels and lightning patterns followed, and since 2021, processed
data in the form of climate data, radar data and forecast data has been released free of charge.

Improve the implementation of API and bulk downloads at the source.

The implementation of APIs and the facilitation of bulk downloads present significant challenges across
various MSs (5). Czechia and Romania, for instance, struggle due to a scarcity of resources and
expertise, resulting in a restricted availability of data through APIs.

Estonia delegates data responsibility to individual agencies, and if they do not want a specific API,
there is a native functionality for APIs on the national portal. This solves the issue of implementing APIs

(5) Digital government APIs: The road to value-added open API-driven services (APIS4DGOV). The Commission’s Directorate-General for
Communications Networks, Content and Technology together with the Joint Research Centre (JRC) launched this study with the
purpose of gaining further understanding of the current use of APIs in digital government and their added value for public services.

19
on data holders’ source portals, often associated with resource- and technical expertise-related
challenges.

5. Remaining challenges
Despite significant progress in adopting the regulation, several challenges persist across MSs.

Data protection laws

The GDPR sets stringent requirements for the processing and protection of personal data within the
EU. While the HVD regulation aims to promote the availability and reuse of public sector data, MSs
must navigate a complex landscape where HVD requirements intersect with GDPR provisions. This
intersection poses challenges, particularly in sectors where sensitive or personal data may be involved.
For example, ensuring compliance with the GDPR while facilitating the reuse of health or demographic
datasets presents a significant challenge for MSs. The HVD regulation proposes solving such a conflict
as provided in Recital 8 (6); however, resolving these sector-specific challenges requires careful
consideration and possibly tailored solutions to balance data accessibility with privacy protection.

For instance, Estonia has recognised the issue of commercially sensitive data and stressed the
importance of conducting a cost–opportunity assessment to address these concerns effectively.

In Italy, a working group involving all public administrations was established to define guiding
documents for implementing the regulation with reference to specific data categories and datasets (7).
One issue they encountered with opening company data, namely company addresses, is that it poses
a risk to the infringement of individual privacy. Italy is now working to address these barriers from both
technical and legal perspectives.

Meanwhile, in the Netherlands, there are ongoing issues with private data protection. Some categories
such as health data or company data remain behind paywalls to make sure that the most sensitive data
remains publicly inaccessible and to safeguard entities’ privacy.

Licence ambiguity

Another persistent challenge is the ambiguity surrounding the types of licences allowed alongside CC
licences for HVDs. CC licences are widely used for open data sharing, but the specific conditions and
restrictions they impose may vary. Even though CC licences, namely CC 4.0, are not mandatory, this
generates confusion as MSs require clarity on whether additional or alternative licences can be used
in conjunction with CC licences for HVDs (8). This ambiguity complicates data-sharing practices and

(6) Recital (8) of the HVD implementing regulation: ‘Member States should make use of appropriate methods and techniques (such as
generalisation, aggregation, suppression, anonymisation, differential privacy or randomisation), thus making as much data as possible
available for re-use.’
(7) See Chapter 3, paragraph 4.
(8) Recital (12) of the HVD implementing regulation: ‘It is the objective of Directive (EU) 2019/1024 to promote the use of standard public
licences available online for re-using public sector information. The Commission’s Guidelines on recommended standard licences,
datasets and charging for the re-use of documents (https://eur-lex.europa.eu/eli/reg_impl/2023/138/oj#ntr5-
L_2023019EN.01004301-E0005) identify Creative Commons (‘CC’) licences as an example of recommended standard public licences.
CC licences are developed by a non-profit organisation and have become a leading licensing solution for public sector information,
research results and cultural domain material across the world. It is therefore necessary to refer in this Implementing Regulation to
the most recent version of the CC licence suite, namely CC 4.0. A licence equivalent to the CC licence suite may include additional
arrangements, such as the obligation on the re-user to include updates provided by the data holder and to specify when the data were
last updated, as long as they do not restrict the possibilities for re-using the data.’

20
introduces uncertainty for data users and providers alike. Therefore, the recommendation for MSs is
to use CC licences for as many HVDs as possible (9).

Denmark is among the MSs impacted by the licencing issues. Although there is a common
recommendation in the public sector to use the CC BY licence, a unified decision has not yet been
reached. This lack of consensus complicates the establishment of clear and straightforward
requirements for data sharing. Estonia has also faced substantial challenges in implementing the CC
licences due to the absence of EU-approved Estonian translations. In order to support the MSs with
this challenge, the European Commission has compiled the EU vocabularies, which includes
multilingual controlled vocabularies for licences and HVD categories.

5.1. High-value dataset implementation roadmap

In moving forward, joint efforts are required to address these remaining challenges and advance the
implementation of the HVD framework across MSs. Collaboration between MSs and the Commission
will be essential in identifying and implementing effective solutions. As an additional aid, a suggested
high-level roadmap is provided in Figure 3: Suggested HVD implementing regulation’s compliance
roadmap. and Figure 4: Suggested HVD implementing regulation's compliance roadmap explained. to
outline specific steps to reach full compliance and successful HVD implementation.

(9) ‘Reuse policy: A study on available reuse implementing instruments and licensing considerations’. The Commission’s reuse policy
establishes that instruments like open licences can aid in implementing the reuse decision. It evaluates four main instruments for this
purpose: the current reuse notice used by the Commission, the CC licencing suite, the open data commons licencing suite and a
bespoke licence specifically created and maintained by the Commission.

21
Figure 3: Suggested HVD implementing regulation’s compliance roadmap.

22
 Figure 4: Suggested HVD implementing regulation's compliance roadmap explained.

6. Conclusion
As per the 2023 Open Data Maturity Report, MSs have made significant progress in implementing the
regulation. Through interviews with the selected MSs, this study highlights the progress and challenges
and proposes the next steps in implementing the HVD framework across MSs. To fully realise the
benefits of HVDs, continued collaboration with and support from the Commission will continue to be
essential.

23
Several challenges persist for all the MSs. First, there is a need for consistent maintenance and
availability of EU standards and vocabularies across all languages. Standardised terminology and
specifications play a crucial role in promoting interoperability and facilitating data exchange across
borders. Ensuring that EU standards are consistently updated and accessible in national languages
would mitigate barriers to cross-border collaboration and data sharing. As part of the effort to enhance
data interoperability and standardisation across the EU, the following guidelines and tools are at MSs’
disposition:

• DCAT-AP high-value dataset guidelines. The DCAT-AP HVD guidelines provide instructions for
using DCAT-AP to catalogue datasets under the regulation. These guidelines ensure that
datasets meet the minimum metadata requirements for mandatory reporting, though they do
not guarantee full compliance with all regulatory aspects. The DCAT-AP descriptions aid in
assessment and improve metadata quality across European data portals, benefiting citizens
and businesses.
• DCAT-AP 3.0. DCAT-AP is a DCAT profile for sharing information about catalogues containing
datasets and data services descriptions in Europe, under maintenance by the Semantic
Interoperability Community action, Interoperable Europe. It provides a minimal common basis
within Europe to share datasets and data services across borders and domains.
• Mapping between Inspire and DCAT-AP – GeoDCAT. GeoDCAT-AP is an extension of the DCAT
application profile for data portals in Europe (DCAT-AP) for describing geospatial datasets,
dataset series, and services. Its basic use is to make spatial datasets, dataset series, and
services searchable on general data portals, thereby making geospatial information more
findable across borders and sectors. For this purpose, GeoDCAT-AP provides an RDF
vocabulary and the corresponding RDF syntax binding for the union of metadata elements of
the core profile of ISO 19115:2003 and those defined in the framework of the Inspire directive
of the EU.
• EU vocabularies. It includes multilingual controlled vocabularies for licences and HVD
categories.

MSs also stress that it is very important to receive timely provision of information. In this sense, the
Commission is working on standardising the reporting process for high-value datasets, and this
involves the use of SPARQL queries for data retrieval and reporting purposes. Information about this
reporting process, including the relevant SPARQL queries, will be made available on data.europa.eu.
Also, an official portal for European data, data.europa.eu, is ready to host and display the high-value
datasets metadata from national catalogues. These datasets will be displayed on data.europa.eu in a
dedicated sub catalogue titled "High-value datasets," allowing users to filter by HVD categories and
other metadata across countries. To further assist the MSs, the first webinar with the MSs already took
place.

Another significant obstacle for several MSs in complying with the regulation is their budgetary
constraints. Dedicated policy implementation funding would improve MSs’ ability to meet regulatory
obligations, invest in necessary infrastructure and resources, and reduce resistance to change at
national, regional, and local levels.

Finally, MSs seek clarification on requirements that overlap between the INSPIRE directive and the
HVD regulation. More specifically, e.g. Finland is uncertain about how to express metadata that
complies with both pieces of legislation. To address the issue, Finland suggests including a notation in
the metadata description to indicate compliance with both INSPIRE and the HVD regulation.

To answer to this request, the European Commission has compiled the following resources:

24
 • Mapping between Inspire and DCAT-AP: GeoDCAT. GeoDCAT-AP is an extension of the DCAT
application profile for data portals in Europe (DCAT-AP) for describing geospatial datasets,
dataset series, and services. Its basic use is to make spatial datasets, dataset series, and
services searchable on general data portals, thereby making geospatial information more
findable across borders and sectors. For this purpose, GeoDCAT-AP provides an RDF
vocabulary and the corresponding RDF syntax binding for the union of metadata elements of
the core profile of ISO 19115:2003 and those defined in the framework of the Inspire directive
of the EU.
• List of specific priority datasets under Inspire that are all within the scope of the HVDs. The list
contains 89 datasets and was drafted in collaboration with the MSs in the Inspire expert group.
• Inspire Geoportal. It showcases the progress made by each MS on both Inspire priority
datasets and HVDs within the scope of Inspire.

Moving forward, collaboration between MSs and the Commission will be crucial in addressing these
challenges and advancing the implementation of the HVD framework. The next opportunity for
collaboration will be the HVD Best Practices in Europe webinar scheduled on 14 June 2024, aimed at
explaining the findings of this study in more detail.

25
 26

ISBN: 978-92-78-44203-3