Applying human-centred design principles to data governance
A democratic approach to policy-making calls for reliable and timely information to ensure effective decision making. The Government of Indonesia in light of this has introduced the Satu Data Initiative to improve the quality of data governance, not just for policy-making but also to increase transparency with open data for citizens. Over the past few months, Pulse Lab Jakarta in partnership with the Executive Office of the President of Indonesia (KSP) have been applying human centred design to model a data governance framework at the local government level. In this post, we discuss how five primary components of human centred design were integrated to help develop a tailored framework.
Data governance has many facets to it, generally ranging from how data is collected to how it is used. It encompasses the individuals, processes and technological infrastructure needed for data management. For national data governance, for instance, a kind of governing framework is needed to outline procedures and to provide support for data entities at different levels of government. Naturally, the availability and usability of data that such a framework enables is something that interests us given our analytics focus as a data innovation lab. And in a similar way, the President’s national strategic plan needs useful data to monitor and evaluate short- and long-term goals. Thus, our mutual love for data was the common denominator.
After attending a public consultation for the Draft Presidential Regulation on One Data Indonesia last year, it was clear from the very frank discussions in that forum that achieving the Satu Data initiative’s aim of better data governance will require many government agencies to change their existing behaviours. We joined forces with the President’s Office to apply the principles of human-centred design in identifying viable solutions that could encourage these behavioural changes in four district/city governments (Kulon Progo, Makassar, Pontianak and Mojokerto) and two provincial governments (Yogyakarta and South Sulawesi) in Indonesia. Step by step, this is how we went about it.
Understanding the Context
We kicked things off by brainstorming with the KSP team on as many exploratory questions as possible, narrowing these down based on KSP’s policy priorities, and then surveying a list of methods to address them. Are there any existing systems resembling a data governance framework? If so, what are the pros and cons of these systems? To get an understanding of the context, we invested a lot of time in observing a variety of data flows, from how data was being produced to how it was being used by local governments.
Understanding the context included researching, observing, shadowing and mapping the user journey (each method was selected based on ideas about the sort of insights we wanted to uncover). With shadowing, for instance, we wanted to look into the dynamics of user interaction within the health sector. We followed frontline health clinic workers collecting field data, observed how the data was reported to the clinic, and looked at ways in which health officials interacted with the data — this way we could get a glimpse of the everyday activities and challenges. In addition, with desk research we wanted to broadly examine regulations that already exist in the country and regulations that may be needed to strengthen the Satu Data Initiative.
Defining the Opportunities
Next, we proceeded to identify opportunity areas for structuring a useful data governance framework. As expected, some of these opportunity areas pointed to working hand in hand with Rancangan Peraturan Presiden Satu Data Indonesia (Draft Presidential Regulation on One Data Indonesia). While the results from our synthesis based on the Draft Presidential Regulation were constructive, it was evident that the draft regulation by itself would not be a panacea for challenges identified. So, Pulse Lab Jakarta together with the President’s Office created a prototype toolkit known as Wawasan Satu Data to help data personnel within the public sector understand data governance in a more holistic way.
Designing the Prototype
The Wawasan Satu Data toolkit was designed through a co-creation process with public sector representatives. It incorporates findings based on our analysis of the local government context. For instance, we found that users have a different set of needs according to their roles as data owners, data stewards, data consumers, etc.
For that reason, several versions of the toolkit were designed, each modified to meet the needs of its users throughout the data lifecycle. It was also important to consider the user mindset: Are users amenable to the use of a structured data governance framework? What are the existing user conditions? What conditions are needed for a new data governance framework to succeed? How does a government sector make a holistic transition from a non-existing framework to something new? And importantly, how do we measure the results and impact of such a transformation?
Testing the Prototype
We tested the Wawasan Satu Data toolkit in several locations in Indonesia, and recorded feedback from users that emerged from role-playing, documentary analysis, co-designing and going over worst-case scenarios. The latter proved to be particularly interesting when we tested the toolkit among local government officials who were not very familiar with data governance, and whenever we substituted our how-to-use facilitation for self-directed use of the toolkit. This was a practical way for us to evaluate whether the toolkit was straightforward, as well as to identify areas for improvement.
More than 200 users took part in the testing of the toolkit. We came to realise that the toolkit can be functional for both individuals and groups. Hence, to enhance the group usage experience, we developed supplementary content to assist group facilitators. We also made changes in terms of how the information was structured in the toolkit based on user observation and feedback. Some users, for example, were not clear on the directions in some sections of the toolkit, so we added a brief introduction in those sections for clarity. The use of visuals was also more appreciated than text-heavy content, therefore we ended up revising some sections with visualisations to simplify steps and make the overall toolkit more appealing.
Evaluating the Whole Process
After conducting a few testing iterations and assessing the impact before and after using the toolkit, our team noted several potential areas in which the Wawasan Satu Data toolkit can assist subnational governments with designing and implementing an effective data governance framework. Many of the individuals who participated in the toolkit testing mentioned being more informed (and indeed more interested) about the overall scope of data governance and its different components, namely the users, processes and technologies involved.
Representatives from the various district/city and provincial governments were also able to better outline stages for formulating a data governance regulation. Roughly two months after participating in the toolkit testing workshop notably, public officials from Kulon Progo, Pontianak and Mojokerto were able to build on inspirations and insights from the workshop to finalise a set of data governance regulations for their respective regions.
The Government of Indonesia is planning to release the Wawasan Satu Data toolkits on their data portal when the Presidential Regulation itself is formalised. Watch their space at https://data.go.id/
If you’re interested in knowing more about our HCD approach, drop us a line at firstname.lastname@example.org — we’ll get back to you in a jiffy.