Acordos de partilha de dados e auto-serviço regido por dados

Unlike many other assets, data has the essential feature that its value increases the more and the better it is shared. That is why sharing data is vital for an organisation to become data-driven. However, establishing the mechanisms to do so in a safe, governed, effective, and efficient manner is one of the greatest challenges of a data governance programme.

In this context, we come from a world where data was organised in silos guarded by pure IT teams (Systems, DBAs, etc.) and these were served in the form of reports, metrics or projects fully cooked and with long development times; this is something that, to this day, has become totally obsolete and is already breaking up in many organisations.

For years, we have been experiencing a technological torrent that brings new capabilities in data processing (Big Data, Cloud, IoT, Advanced Analytics, AI & ML, etc.). We have a greater data culture among people and organisations, there is more training in data, data literacy and data storytelling, and new skills are being acquired to consume data and generate insights in business areas thanks to more and better data analysts and scientists, etc. so it is clear that data is no longer something that can be guarded by pure IT teams and served in long development times, but something that must be put at the service of the business (people or algorithms) for rapid and efficient decision-making that generates a competitive advantage.

What are the benefits of sharing data within an organisation or between organisations?

• Maximise synergies between different actors to improve their daily operations. We can see this in the environment of a single organisation where different departments have to share information in order to be able, in their day-to-day work, to carry out their tasks, activities or make certain decisions. Another very clear case is for research and innovation, hence the fact that there are many Open Data projects worldwide focused on open data sharing (the latest and largest one in the European Union

• Furthermore, Compartir esos datos mejora la calidad de los datos clave para tomar mejores decisiones más rápidamente.. If data is made available to other consumers, producers will place greater emphasis on sharing quality data that makes it easier to make decisions more quickly.

• Reuse developments and work already done by increasing the efficiency of data projects. In this way, we will ensure that each data initiative that requires the development of processes or previous analytical work does not have to start from scratch and can reuse the work already done in common frameworks or collaboration models.

• Reducing operating costs and improving the productivity of professionals. Data sharing also leads to savings in operating costs because it prevents duplication of data for certain activities or overlapping manual data management tasks (search, understanding, request, cleaning, etc.), and this results in an improvement in the productivity of the organisation's professionals.

• In turn, it also aims to improve the time-to-market and time-to-value of new products and services of any kind. Today, a large part of the products or services that are launched into the market, both from the public administration and from private organisations, depend to a great extent on analyses made from historical data or from future predictions. We all know the very successful case of the Netflix recommendations algorithm (https://research.netflix.com/research-area/recommendations)

• In turn, it also aims to improve the time-to-market and time-to-value of new products and services of any kind. Today, a large part of the products or services that are launched into the market, both from the public administration and from private organisations, depend to a great extent on analyses made from historical data or from future predictions. We all know the very successful case of the Netflix Recommendation algorithm. 

• Provide transparency and confidence both in data and in the processes of capture, storage, transformation and consumption of data and information. If we share, it is clear that we have nothing to hide and this is something key especially for the Public Administration of the 21st century. The case of Estonia is one of the most talked about in this example: https://www.tallinn.ee/eng/Uudis-Regions-and-Cities-contribute-the-most-of-open-data-in-Estonia-Tallinn-is-the-top-publisher-overall

Once we are clear about the benefits of sharing data, this leads to a dilemma between two parties: on the one hand, we need to enhance the use and reuse of data and for that we need to democratise its access; and on the other hand, we have to have control over the use that is made of that data to ensure regulatory compliance and the safety of those data. In the end, there is always that turning point that makes us reflect on the different paths involved in sharing data.

«Aurea mediocritas: At the midpoint is virtue» – Aristotle

That is the middle ground that is achieved with the entry into play of Data Sharing Agreements (DSAs) to share data in an environment where we feel comfortable and there is a government. In this sense, we always have to seek the balance between giving value to the business and complying with the regulation; follow what the business needs, but at the same time not having great technological costs; enhance self-service, but having control of the information that is being made available to the rest; and then also pursuing innovative and disruptive solutions without forgetting that the traditional cannot disappear overnight.

The data is a strategic asset, but has different perspectives.

Whenever we discuss data sharing, what we mean is making information available by producers for use by consumers, and in that process there are roles that have to ensure that everything works correctly. Ultimately, each person or type of role will have their own profile, their own visions and needs; and thanks to DSAs we will be able to cover all of them.

If we look at a data consumer's point of view, they need to have access to knowledge of the data because of its context and meaning; a guarantee of the quality of the data used for decision-making; compliance with SLAs by producers; and regulatory compliance from the point of view of use.

From the point of view of a producer, this seeks to control the production processes and the availability of their data, prepare and certify the data that will be shared, know which processes and which people are using the data and for what, and comply with data regulation and regulations.

From a cross-cutting management and monitoring perspective, the objectives will be: maximising the ROI of data initiatives, increasing efficiency in process productivity, automating and reducing operational costs and risks, achieving a homogeneous view of information consumption, and ensuring regulatory compliance.

But what exactly is a Data Sharing Agreement?

A DSA is essentially a mechanism for sharing data between producers and consumers in an agile and personalised manner, facilitating regulatory and regulatory compliance in the use of data. In addition, thanks to DSAs, we can standardise the way in which data and information access are requested, granted, and managed, gaining in efficiency and productivity.

On the other hand, DSAs are represented as logical assets that group different technical data assets or Datasets (tables, files, views, documents, events, etc.) so they offer stakeholders a new experience of sharing information by facilitating all the technical complexity behind data sharing by containing all the metadata information of the technical assets they encompass. Thus, DSAs bring the data closer to the business consumer, abstracting them from all the underlying technical concepts.

Finally, DSAs also allow for the definition, implementation and operation of Data Contracts in order to provide the necessary flexibility to implement different data access policies for each specific case of information sharing, since sharing public data from three years ago is not the same as sharing citizens' addresses or confidential information about an organisation's strategy. There are times when a Data Contract can serve multiple purposes and multiple consumers, but there are other times when specific Data Contracts are needed for very specific cases.

But what exactly is a Data Sharing Agreement?

Incorporating innovative elements in data governance such as DSAs is not an easy task, but if you are genuinely interested and want to do it, the best thing is to have the right people, the right processes and the necessary technology. If you want to know more about how to operate DSAs, different use cases and success stories, take a look at this. document, this webinar or contact us.