European Data Portal

How to build an Open Data Strategy

How to build an Open Data Strategy

Contents

 

Before starting to publish any Open Data, it is important to have a clear strategy in place that defines the key goals and sets the ambition. This chapter will address these key ingredients for a successful Open Data initiative as well as addressing barriers that one might face along the way and how these can be best tackled.

/en/file/howtobuild1png-0howtobuild1.png

 How to build an Open Data Strategy

How to build an Open Data Strategy

 

Setting the ambition

Before taking action, you need to define what you want to achieve. This is often called “setting the ambition.” Defining your ambition implies answering a series of questions such as: where do you want to stand? And by when? Will all data be available by default? Is all data stored centrally? In order to set a clear ambition, follow the steps as visualised below. Start by defining the As-Is situation, before going into the definition of the To-Be Situation and define measurable goals. 

/en/file/howtobuild2pnghowtobuild2.png

 The steps for setting the ambition

The steps for setting the ambition

 

Defining the As-Is Situation

In order to set the right ambition you first need to define a clear picture of the current situation. This we call the ‘As-Is situation’. To get a clear picture of the As-Is situation in your organisation, the following 4 steps will guide you through the assessment of your current situation: 

1. Gather a representative group 
Identify key representatives of different sectors, units and departments of your organisation that can help you create the right As-Is situation.

2. Identify what the units do <
Which units collect data? Which ones use data? Which ones produce data? What type of data do they gather? What format? A visualisation of the situation might help to create an overview.

3. Is data centrally organised? 
The eventual effort required highly depends on two factors: the organisational data structure and the willingness of people. Throughout the organisation, data is generated by multiple units and is maintained at the same level. As a result, data is stored in numerous places with multiple responsible people dedicated to its storage and quality. On the other hand, data can also be centrally stored by a single data storage and maintenance unit, including a team delivering the service of data storage, maintenance and delivery. When your data is generated by multiple units and they all have their own policies and ways of working and people, you can imagine that implementing a new way of working becomes more challenging than when data is centrally managed.

4. Is data currently published? 

If so, very good! This can be a good starting point. Find out what the current policies are around this process and try to identify best practices.

If no data is published yet, there is no need to worry. You can contribute to launching the process, which has its benefits as well and enables avoiding former restrictions, processes or wrongly manipulated data that might blur the view on the root of your data.

One of the main questions where many Open Data initiatives end is: if we were to publish data, what data should we open up? 

This question frequently leads to taking a step back and ends up in a discussion about whether to publish data or not. Sometimes it ends up in the decision not to publish data unless citizens request for it. Try to avoid such discussions and check whether you can identify some quick wins within your organisation that are part of your end-goal.

In summary, start by gathering a representative group, identify what the units do, check whether data is centrally organised and currently published.

/en/file/howtobuild3pnghowtobuild3.png

 Steps to define As-Is situation

Steps to define As-Is situation

 

Recommendation: Visualize the As-Is situation. This will help you when you create an overview
This is a good example of how a data stream can be visualised

Define the To-Be situation and define measurable goals

The next step is to detail your ambition. Therefore, you need to think of the To-Be situation: What do you want to achieve? Where will your organisation stand in 2 years’ time? Or perhaps in 5 years’ time? Discuss this with your group of representatives and create a clear picture of that To-Be situation. Again, we recommend you to create a visualisation of this situation. 

Compare and define the goals 
Put both visualisations next to each other and compare the As-Is and To-Be situation: What are the differences? What needs to be done? By means of defining clear and measurable goals, your organisation will be enabled to work towards those end goals and measure whether you have achieved them or not. While creating the goals, think of the primary reasons for publishing the data. These could be, for example, reaching the goal of becoming transparent or stimulating the economy. Make sure to be precise. Goals should be described in terms of scope, timing, deliverables and quantities among others. 

Define measurable goals to demonstrate your success over time!

Do you want to know more about the current Open Data situation in the EU Member States? Read the Open Data Maturity in Europe report:

 

Creating your strategy

The next step will be about defining how you will achieve these goals. The As-Is and To-Be situations are visualised and the goals have been defined. You are now entering the phase of creating a strategy. While creating the strategy, take into account the questions you foresee and which actions you are going to take. 

There should be a policy and a direction in place once the Open Data initiative is implemented. If not already done so when capturing the As-Is and To-Be situation, create a list of people within your organisation you will have to contact and involve. Create a scope, timeframe and planning, and start executing your project. Make sure your plan includes the topics as shown in the figure below. 

As a summary, here are the nine topics every Open Data policy should cover:

/en/file/howtobuild4pnghowtobuild4.png

 the 9 key topics every policy should cover

the 9 key topics every policy should cover

 

The UK, one of the leading countries in Europe with regard to Open Data publication, created a strategy with an action list. The various departments of the British government executed this strategy.
Look at the strategy, policy and action list at the following page:

https://www.gov.uk/government/publications/2010-to-2015-government-policy-government-transparency-and-accountability

Nine Dutch municipalities collaborated with their Open Data initiatives. With a collective program, they created commitment and achieved their goals faster and more effectively. Look at their website for the program and the success stories (Dutch):

http://www.noord-holland.nl/web/Projecten.htm

Publishing data: ‘Open by default’ or ‘Closed unless’? 
Concerning the discussion of what data to publish, organisations often end up choosing to close their data unless there is a clear purpose. This is a missed opportunity. Why? Because it is impossible to identify all opportunities different types of data create from the back of one’s head; the opportunities are endless. You might classify a data set as useless whereas a solutions architect has valuable ideas for the use of that set. Furthermore, assessing the usability of data case by case is a timely and unnecessary effort. 

/en/file/howtobuild5pnghowtobuild5.png

Be Open

Therefore, we advise you to choose the ‘Open by default’ as a standard for your organisation, when it is legally feasible of course. However, if the assessment of the initial list of data shows that a very low percentage of data can be published, it might be wise to choose a case by case assessment.

 

Drafting an Open Data Policy

Your Open Data policy is one of the most effective documents to detail your ambition and the way you intend to realise it. It will support your implementation and set the standard for the field. It will create the transition, increase the transparency of your organisation and ensure the best use of your data! The translation of your Open Data strategy into a solid policy is of great importance to ensure its successful implementation. First, an overview of the policy is given. Afterwards, the content of the policy is further detailed. 

The Open Data Policy: an overview

What is part of the policy? What data will you publish? Under what conditions? When? How often? And why? For what expected impact? What benefits? The policy will answer all the questions for the people within and outside your organisation. Take enough time to design your policy and make sure the decision makers within your organisation endorse it. 

There are several policies available to look at. In line with your plan, make sure that the policy lands on the desk of those who are responsible for releasing data. You can expect many questions from stakeholders concerning the practicalities, such as budget, technical and practical aspects and legal boundaries. Therefore, make sure sufficient consultation and involvement takes place.

Your policy should cover topics such as definitions and expected benefits, the scope of the policy and expected outcome, legal aspects, etc. You may also consider defining data types and data quality and mandating an organisation, department or unit as responsible for the implementation of the Open Data policy.

/en/file/howtobuild6pnghowtobuild6.png

 Topics to include in your policy

 

The content of your policy

Definitions and benefits, scope and goals, legal aspects, data types and quality, point of contact are all important aspects to include in your data policy. These are all described below:

 

/en/file/howtobuild7pnghowtobuild7.png

Defenition and benefits

 

In the policy, it is important to indicate what data you want to publish, including a clear definition of Open Data. For example, part of the definition of Open Data could be that it is not only free to use, but also free of charge. Include definitions of the terms you use to clarify the scope. It is helpful to emphasize the benefits by including several examples that make explicit ‘what’s in it for them’. You can use examples from this Goldbook. Try to translate these benefits to specific benefits that fit the vision of your organisation. For example, having data available not only for externals, but also internally for your colleagues, does that increase efficiency within your organisation? Then apply this argument as a key driver for convincing colleagues about your Open Data initiative.

 

/en/file/howtobuild8pnghowtobuild8.png

Scope and goals

 

The scope is important to bring focus to the Open Data initiative. For example, it is possible to prioritize the release of data from specific departments first. Include what is part of the ambition, and what is not, by making your goals explicit and incorporating those in your policy. Describe the purpose, for instance, to improve transparency of your organisation and support the growth of the economy. 

 

/en/file/howtobuild9pnghowtobuild9.png

Legal aspects

 

The implementation of Open Data has to be in line with current legislation. Publish datasets under an open licence. These legal implications are of major importance for any stakeholder dealing with Open Data. It is your responsibility as a publisher to be up to date with the most recent legislative rules applicable in your country. Your policy should address the following legislative topics.
 

Licensing 

The data is not freely re-usable if you do not attach a licence to it. Thus, rule number one with regard to legally opening up your data set is to attach an appropriate licence to it. There are many different types of licences you can apply, for instance one of the Creative Commons (CC) licences. A Public Domain Dedication is part of the CC licences and important to attach as it indicates that the public domain is the author of the data. More information is available on their website: 
https://creativecommons.org/licences/publicdomain/
The Open Data Institute created a comprehensive publishers guide to Open Data licensing: 
https://theodi.org/guides/publishers-guide-open-data-licensing 
Furthermore, the European Commission has published an introduction to data and metadata licensing: 
d2.1.2_training_module_1.4_introduction_to_metadata_management_en_edp.pdf

 

Law: The Public Sector Information Directive

Every country has its own specific legal regulations when it comes to data. Therefore, all EEA countries should have transposed the 2013/37/EC PSI Directive into the legislation of their country. The status of the PSI Directive transposition per country can be accessed here: 

https://ec.europa.eu/digital-agenda/en/implementation-public-sector-information-directive-member-states 

 

Intellectual Property

Intellectual property broadly includes everything created by the human mind. Open Data is free of intellectual property, as it is free to download, manipulate and re-use for any purpose, by anyone. It is possible to limit the re-use of Open Data by setting some legal boundaries to protect the provider, by adding restrictions of re-use in the licence. 

 

Privacy

Data is sensitive and one must be cautious not to act in violation of privacy regulations. Make sure data is cleaned thoroughly and the appropriate level of anonymization is applied to prevent the identification of a specific individual through your data. There are several legislative matters at EU level that determine how to handle data protection and how that applies to PSI. Re-use is only allowed in full compliance with the personal data protection rules. Regulation (EC) No 45/2001 of the European Parliament and of the Council of 18 December 2000 on the protection of individuals with regard to the processing of personal data by the Community institutions and bodies and on the free movement of such data applies (European Union, 2015). The most comprehensible guideline is published by the ePSI-Platform, accessible via the following link: 

http://www.europeandataportal.eu/sites/default/files/data-protection-in-re-use-psi.pdf
More official documents are available that apply to data protection of individuals with regard to the processing of personal data and on the free movement of such data. Be aware of these legislations and regularly keep yourself updated.
http://eur-lex.europa.eu/legal-content/en/TXT/?uri=CELEX:31995L0046 

 

Liability

Another important legal aspect is the liability of your organisation. Make sure all regulations are followed to prevent liability issues. The data that you publish should be reliable, which means with limited or no errors and anonymised.

 

Commercial law

Less likely to be a concern, but certainly a current debate: commercial law. Keep in mind that some datasets might intrude on the competitiveness of companies who have their business model built around particular data, which is now accessible by anyone.

 

/en/file/howtobuild10pnghowtobuild10.png

Data types and quality

 

Data is available in numerous types, formats and quality. Releasing relevant and quality data requires effort, but serves multiple purposes: 

  • It is easy to re-use data in Open Data formats
  • Relevant data is used more frequently
  • High quality (complete, well-documented) datasets enable re-users to identify its value for their purposes quickly

 
It is important to describe data types, topics and quality in your policy. It will serve as a standard for most of the datasets published by the public bodies and will increase the overall quality of your Open Data initiative and thus increase its effect. 

 

/en/file/howtobuild11pnghowtobuild11.png

Point of contact

 

As a reference, it is useful to mention who is the responsible Point of Contact within your organisation. He or she will act as a single point of contact for queries related to Open Data within the organisation.

 

Overcoming barriers in publishing Open Data

Despite the wide variety of benefits, governments might still not acknowledge the potential of Open Data and seek for excuses not to publish the data as Open Data. This section offers a summary of frequently used excuses and an answer to the misconception. 

The most frequent barriers recorded are 

  • Data is not interesting, 
  • The purpose or benefit for the organisation is unknown, 
  • There will be too many user requests for data, 
  • Users will draw superficial conclusions from the data, 
  • Data is not sufficiently accurate to be shared, 
  • It will cost too much to transform the data to a standard format, 
  • There is a risk to get a negative reputation,
  • Publishing low quality data could harm the image of the public sector organisation.

The picture below offers a remedy to each barrier: 

/en/file/howtobuild12pnghowtobuild12.png

 Summary of barriers for publishing Open Data

Summary of barriers for publishing Open Data

 

Ensuring organisational alignment as a Key Success Factor
Ensuring organisational alignment for Open Data initiatives and implementing it in a sustainable way is essential to the success of an Open Data initiative. The Open Data Institute has performed a thorough analysis of change in organisations that publish their data as Open Data. They performed an extensive literature review with regard to change management and tested their findings through interviews with seven countries (Broad, E., Smith, F., Duhaney, D., Carolan, L., 2015). Their findings are summarised in a comprehensive article that elaborates on managing change in general, and how to successfully apply the basic principles to an Open Data initiative. 

The 12 main principles of Open Data and managing changes:

/en/file/howtobuild13pnghowtobuild13.png

 Summary of basic Open Data principles

Summary of basic Open Data principles

 

Critical success factors

The success of an Open Data initiative depends on three dimensions, namely: the quality of Open Data publication (e.g. accuracy, completeness, timeliness, and consistency), the use of Open Data, and documenting the emerging impacts and benefits. 

So far, little evidence exists regarding measurable factors that influence the success or failure of Open Data initiatives. The most critical success factors seem to be around addressing legislation, regulation, and licences. These aspects are very dependent on the local context. In addition, a number of success factors (e.g. sustainability of publication process, user feedback) appear to be applicable more universally.

An initiative is any activity that aims at improving the publication and/or use of Open Data, including initiatives on different levels (e.g. international, national, local) and by different stakeholders (e.g. civil servants, citizens, universities). Some factors are only critical for the success of data publication, while others are only critical for data use. It is therefore best to determine success factors regarding data publishing on the one hand and data re-use on the other. 

Besides, the context of the initiative may also determine the criticality of a success factor. For example, multilingualism is critical when the Open Data initiative is organised globally and involves datasets from different countries, while this is less important for initiatives on a local or national level. Furthermore, whether the data is re-used for commercial or non-commercial purposes may influence the criticality of success factors for Open Data use. If there is a specific need to make Open Data use profitable, it is likely to lead to different success factors than in the case of not-for-profit data use.

Critical success factors for publishing
The most critical category of factors for the disclosure of Open Data seems to be legislation, regulation, and licences. 
Furthermore, the sustainability of the Open Data initiative is a category of factors that is most critical for Open Data publication. Essential factors concern identifying the need for data, ensuring the continuity of data supply (including timely and automatic updates of data), and being transparent towards Open Data users about the conditions under which data publication takes place. Factors regarding accessibility, interoperability and standards that were critical for Open Data publication success are multilingual metadata and data, the use of standards (for data, metadata, licences, URIs and exchange protocols), the integration of metadata 

schemas and federated controlled vocabularies, the provision of various types of metadata in line with metadata standards, and the supply of APIs for Open Data provision in the form of service needs.

Finally, within the category Open Data platforms, tools and services, the critical factors are the presence of one central portal, which combines data from many different governmental organisations, the integration of frameworks for assessing data quality and usability of data and platform, providing continuous feedback to developers and administrations and the development of a clear user interface. Stewardship and the development of a management plan seem critical success factors. Therefore, it is important to start with a clear Open Data strategy.

 

Critical success factors for re-use
Legislation, regulation, and licences seem also critical for the re-use of Open Data. For example, the provision of information on the meanings and implications of licences, and on privacy legislation and how Open Data can be used in compliance with this legislation, are critical. Furthermore, success stories are critical, especially the provision of readily available examples of Open Data use (e.g. applications) to non-experts. Success stories attract a large user base. In addition, all factors related to feedback and sustainability are critical for the use of Open Data. The provision of mechanisms for governmental agencies to know how their data is re-used is important. Furthermore, to know what can be learned from the re-use of their data and how the publication of their data can be improved. Open Data users should know precisely the methodology of how the data was produced described in a scientific manner. 

Contents