Podcast: data governance

This episode focuses on data governance and why it is important to have standards, policies and processes in place to ensure that data is correct, reliable, secure and useful. For this purpose, we analyze the Model Ordinance on Data Governance of the Spanish Federation of Municipalities and Provinces, known as the FEMP, and its application in a public body such as the City Council of Zaragoza. This will be done by the following guests:

• Roberto Magro Pedroviejo, Coordinator of the Open Data Working Group of the Network of Local Entities for Transparency and Citizen Participation of the Spanish Federation of Municipalities and Provinces and civil servant of the Alcobendas City Council.

• María Jesús Fernández Ruiz, Head of the Technical Office of Transparency and Open Government of Zaragoza City Council.

Listen to the full podcast (only available in Spanish)

Summary of the interview

1. What is data governance?

Roberto Magro Pedroviejo: We, in the field of Public Administrations, define data governance as an organisational and technical mechanism that comprehensively addresses issues related to the use of data in our organisation. It covers the entire data lifecycle, i.e. from creation to archiving or even, if necessary, purging and destruction. Its purpose is that data is of quality and available to all those who need it: sometimes it will be only the organisation itself internally, but many other times it will be the general public, re-users, the university environment, etc. Data governance must facilitate the right of access to data. In short, data governance makes it possible to respond to the objective of managing our administration effectively and efficiently and achieving greater interoperability between all administrations.

2. Why is this concept important for a municipality?

María Jesús Fernández Ruiz: Because we have found that, within organisations, both public and private, data collection and management is often carried out without following homogeneous criteria, standards or appropriate techniques. This translates into a difficult and costly situation, which is exacerbated when we try to develop a data space or develop data-related services. Therefore, we need an umbrella that obliges us to manage data, as Roberto has said, effectively and efficiently, following homogeneous standards and criteria, which facilitates interoperability.

3. To meet this challenge, it is necessary to establish a set of guidelines to help local administrations set up a legal framework. For this reason, the FEMP Model Ordinance on Data Governance has been created. What was the process of developing this reference document like?

Roberto Magro Pedroviejo: Within the Open Data Network Group that was created back in 2017, one of the people we have counted on and who has contributed a lot of ideas has been María Jesús, from Zaragoza City Council. We were leaving COVID, just in March 2021, and I remember perfectly the meeting we had in a room lent to us by the Madrid City Council in the Cibeles Palace. María Jesús was in Zaragoza and joined the meeting by videoconference. On that day, seeing what things and what work we could tackle within this multidisciplinary group, María Jesús proposed creating a model ordinance. The FEMP and the Network already had experience in creating model ordinances to try to improve, and above all help, municipalities and local entities or councils to create regulations.

We started working as a multidisciplinary team, led by José Félix Muñoz Soro, from the University of Zaragoza, who is the person who has coordinated the regulatory text that we have published. And a few months later, in January 2022 to be precise, we held a meeting. We met in person at the Zaragoza City Council and there we began to establish the basis for the model ordinance, what type of articles it should have, what type of structure it should have, etc. And we got together a multidisciplinary team, as we said, which included experts in data governance and jurists from the University of Zaragoza, staff from the Polytechnic University of Madrid, colleagues from the Polytechnic University of Valencia, professionals from the local public sphere and journalists who are experts in open data.

The first draft was published in May/June 2022. In addition, it was made available for public consultation through Zaragoza City Council's Citizen Participation platform. We contacted around 100 national experts and received around 30 contributions of improvements, most of which were included, and which allowed us to have the final text by the end of last year, which was passed to the legal department of the FEMP to validate it. The regulations were published in February 2024 and are now available on the Network's website for free download.

I would like to take this opportunity to thank the excellent work done by all the people involved in the team who, from their respective points of view, have worked selflessly to create this knowledge and share it with all the Spanish public administrations.

4. What are the expected benefits of the ordinance?

María Jesús Fernández Ruiz: For me, one of the main objectives of the ordinance, and I think it is a great instrument, is that it takes the whole life cycle of the data. It covers from the moment the data is generated, how the data is managed, how the data is provided, how the documentation associated with the data must be stored, how the historical data must be stored, etc. The most important thing is that it establishes criteria for managing the data while respecting its entire life cycle.

The ordinance also establishes some principles, which are not many, but which are very important and which set the tone, which speak, for example, of effective data governance and describe the importance of establishing processes when generating the data, managing the data, providing the data, etc.

Another very important principle, which has been mentioned by Roberto, is the ethical treatment of data. In other words, the importance of collecting data traceability, of seeing where the data is moving and of respecting the rights of natural and legal persons.

Another very important principle that generates a lot of noise in the institutions is that data must be managed from the design phase, the management of data by default. Often, when we start working on data with openness criteria, we are already in the middle or near the end of the data lifecycle. We have to design data management from the beginning, from the source. This saves us a lot of resources, both human and financial.

Another important issue for us and one that we advocate within the ordinance is that administration has to be data-oriented. It has to be an administration that is going to design its policies based on evidence. An administration that will consider data as a strategic asset and will therefore provide the necessary resources.

And another issue, which we often discuss with Roberto, is the importance of data culture. When we work on and publish data, data that is interoperable, that is easy to reuse, that is understood, etc., we cannot stop there, but we must talk about the data culture, which is also included in the ordinance. It is important that we disseminate what is data, what is quality data, how to access data, how to use data. In other words, every time we publish a dataset, we must consider actions related to data culture.

5. Zaragoza City Council has been a pioneer in the application of this ordinance. What has this implementation process been like and what challenges are you facing?

María Jesús Fernández Ruiz: This challenge has been very interesting and has also helped us to improve. It was very fast at the beginning and already in June we were going to present the ordinance to the city government. There is a process where the different parties make private votes on the ordinance and say "this point I like", "this point seems more interesting", "this one should be modified", etc. Our surprise is that we have had more than 50 private votes on the ordinance, after having gone through the public consultation process and having appeared in all the media, which was also enriching, and we have had to respond to these votes. The truth is that it has helped us to improve and, at the moment, we are waiting for it to go to government.

When they tell me how do you feel, María Jesús? The answer is well, we are making progress, because thanks to this ordinance, which is pending approval by the Zaragoza City Council government, we have already issued a series of contracts. One that is extremely important for us: to draw up an inventory of data and information sources in our institution, which we believe is the basic instrument for managing data, knowing what data we have, where they originate, what traceability they have, etc. Therefore, we have not stopped. Thanks to this framework that has not yet been approved, we have been able to make progress on the basis of contracts or something that is basic in an institution: the definition of the professionals who have to participate in data management.

6. You mentioned the need to develop an inventory of datasets and information sources, what kind of datasets are we talking about and what descriptive information should be included for each?

Roberto Magro Pedroviejo: There is a core, let's say a central core, with a series of datasets that we recommend in the ordinance itself, referring to other work done in the open data group, which is to recommend 80 datasets that we could publish in Spanish public administrations. The focus is also on high-value datasets, those that can most benefit municipal management or can benefit by providing social and economic value to the general public and to the business community and reusers. Any administration that wants to start working on the issue of datasets and wonders where to start publishing or managing data has to focus, in my view, on three key areas in a city:

• The personal data, i.e. our beloved census: who are the people living in our city, their ages, gender, postal addresses, etc.
• The urban and territorial data, that is, where these people live, what the territorial delimitation of the municipality is, etc. Everything that has to do with these sets of data related to streets, roads, even sewerage, public roads or lighting, needs to be inventoried, to know where these data are and to have them, as we have already said, updated, structured, accessible, etc.
• And finally, everything that has to do with how the city is managed, of course, with the tax and budget area.

That is: the personal sphere, the territorial sphere and the taxation sphere. That is what we recommend to start with. And in the end, this inventory of datasets describes what they are, where they are, how they are and will be the first basis on which to start building data governance.

María Jesús Fernández Ruiz: Another issue that is also very fundamental, which is included in the ordinance, is to define the master datasets. Just a little anecdote. When creating a spatial data space, the street map, the base cartography and the portal holder are basic. When we got together to work, a technical commission was set up and we considered these to be master datasets for Zaragoza City Council. The quality of the data is determined by a concept in the ordinance, which is respecting the sovereignty of the data: whoever creates the data is the sovereign of the data and is responsible for the quality of the data. Sovereignty must be respected and that determines quality.

We then discovered that, in Zaragoza City Council, we had five different portal identifiers. To improve this situation, we define a descriptive unique identifier which we declare as master data. In this way, all municipal entities will use the same identifier, the same street map, the same cartography, etc. and this will make all services related to the city interoperable.

7. What additional improvements do you think could be included in future revisions of the ordinance?

Roberto Magro Pedroviejo: The ordinance itself, being a regulatory instrument, is adapted to current Spanish and European regulations. In other words, we will have to be very vigilant -we are already - to everything that is being published on artificial intelligence, data spaces and open data. The ordinance will have to be adapted because it is a regulatory framework to comply with current legislation, but if that regulatory framework changes, we will make the appropriate modifications for compliance.

I would also like to highlight two things. There have been more town councils and a university, specifically the Town Council of San Feliu de Llobregat and the University of La Laguna, interested in the ordinance. We have received more requests to know a little more about the ordinance, but the bravest have been the Zaragoza City Council, who were the ones who proposed it and are the ones who are suffering the process of publication and final approval. From this experience that Zaragoza City Council itself is gaining, we will surely all learn, about how to tackle it in each of the administrations, because we copy each other and we can go faster. I believe that, little by little, once Zaragoza publishes the ordinance, other city councils and other institutions will join in. Firstly, because it helps to organise the inside of the house. Now that we are in a process of digital transformation that is not fast, but rather a long process, this type of ordinance will help us, above all, to organise the data we have in the administration. Data and the management of data governance will help us to improve public management within the organisation itself, but above all in terms of the services provided to citizens.

And the last thing I wanted to emphasise, which is also very important, is that, if the data is not of high quality, is not updated and is not metadata-driven, we will do little or nothing in the administration from the point of view of artificial intelligence, because artificial intelligence will be based on the data we have and if it is not correct or updated, the results and predictions that AI can make will be of no use to us in the public administration.

María Jesús Fernández Ruiz: What Roberto has just said about artificial intelligence and quality data is very important. And I would like to add two things that we are learning in implementing this ordinance. Firstly, the need to define processes, i.e. efficient data management has to be based on processes. And another thing that I think we should talk about, and we will talk about within the FEMP, is the importance of defining the roles of the different professionals involved in data management. We are talking about data manager, data provider, technology provider, etc. If I had the ordinance now, I would talk about that definition of the roles that have to be involved in efficient data management. That is, processes and professionals.

Interview clips