Open data value generation cycle

As the last part of my PhD project, I have tried to estimate where I have been contributing to research on the open data value generation cycle and where I have not. I hope to develop these ideas into a research agenda. My insights regarding this cycle are based on a literature review, a review of mass-media open/big data discussions, and finally a triangulation between a) case study from the supply side perspective, b) a case study from the demand side perspective, and c) a collection of secondary (open) data from various organizations. I used the quantitative data to create and test a model that could help me identify the most relevant relationships between open data and value over 76 countries.

I use critical realism as a philosophical foundation. Critical realism predicts we can observe events in the empirical realm (it is not a given that we do observe them, but we can) and that those events are "manifestations" of underlying causal powers that are called mechanisms. We cannot observe the mechanisms directly, but we can use a method called retroduction to hypothesize about these underlying, enduring powers from our observations in the empirical domain.

The cycle below is supposed to depict in a very simplified manner how we can proceed from the decision to make data openly available towards an evaluation of impacts. As this is a cycle it has no beginning and no end. Which is in itself problematic :)

However, I will start with the digital strategy and vision and emphasize that as most of the open data out there is government data, this "event" happens mostly within the public sector. The public sector perceives that data are valuable to society and creates a strategy - a vision - on how these data can be "put into play". The next event is the planning and deployment of the Open Data Initiative (ODI), but what happens in between is what I call (for now) decision making mechanisms.  These mechanisms vary greatly between different public organizations at different levels. There is a body of research that looks at  policy making and the barriers and enablers related to the decision to open data. I have not contributed much to this corpus as it is a bit too geared toward public governance which is not my area of expertise.

The Open Data Value Generation Cycle

After planning and deployment, the dissemination of liquid open data starts. I call the causal powers that link these two events governance mechanisms. These mechanisms describe how the stakeholders involved in the Open Data Initiative (ODI) establish processes and tools so that they can provide sustainable liquid open data of high quality and maintain oversight and accountability in a loosely coupled organizational structure (which is often necessary as the data comes from various independent organizations). My (unpublished) supply side case study on the Danish Basic Data Program discusses the challenges related to ODI governance mechanisms.

Regarding dissemination of liquid open data, I have created a taxonomy with 5 dimensions and 7 attributes that describes the liquid open data construct. I hope this construct can be used to evaluate the "openness" of individual datasets (and here it is not alwasy the case that the more is better, it depends on the vision and strategy). This construct has a 1) strategic dimension (open by default?), 2) legal dimension (open licenses?) 3) economic dimension (free or marginal cost?), 4) conceptual dimension (interoperable?) and 5) technical dimension (are the data usable, accessible and discoverable?).

The mechanisms that are the driving force behind reuse of data (via intermediaries or Multi-Sided Platforms/MSPs) that are in various states of being liquid and open are the engagement mechanisms. Here is a variety of unanswered research questions, to which I have NOT contributed. These questions revolve around areas like who is engaging with the data and how does it happen? Or, in some cases, a more important question might be: Why are people NOT engaging with the data. Should the public sector do more to engage people, or should their main concern e to make the data liquid and open and let the market take care of the rest (by identifying business opportunities and implementing platforms). There are some few, but interesting, research papers that have already started to address these questions.

I have written about use of open government data, how platform intermediaries are arriving and what are the main elements in their business models (unpublished book chapter based on use cases) and how these data are being used to generate sustainable value (case study on the use of a combination of OGD and smart meter data in the Energy realm). My main contribution lies in this area, where I have identified the relevant value generating mechanisms and the contextual elements that I propose will influence these mechanisms. I have a number of papers positioned there. Furthermore, I have spent some time on conceptualizing what I call sustainable value, emphasizing the need to move beyond monetary measures and concentrate more on other types of impacts.

A well know example of a value generating mechanism is the market mechanism, which essentially utilizes the forces of supply and demand to determine the appropriate price and quantity of goods and services bought and sold in free markets and thus improves market efficiency and creates opportunities for commercializing innovations.  While it is never easy to evaluate how much open data is worth, this mechanism makes is at least possible to look at monetary measures for economic value like market size or profit of companies that sell open data products and services on the market. However, while the market mechanism has served us well, the trend towards openness and sharing is slowly diluting the effects of this mechanism.

If we consider the use of open (free) data to generate information which is shared and consequently used by an individual who becomes empowered and makes better decisions, we face a huge measurement challenge. A large part of my research has been centered around a type of a value generating mechanism, which I call (for now) the information sharing mechanism. This mechanism can be further broken down into lower level mechanisms such as transparency mechanisms and civic engagement mechanisms. We often assume that these types of mechanisms create value but we do not quite understand how it happens. At least not in the same way as we understand the value of generating a new product that is sold on the market or the value of implementing a new efficient process that reduces costs. And my proposition is that we REALLY need to understand the value of information sharing, and offer my modest contribution there.

The final set of mechanisms are the evaluation mechanisms, which drive strategy and vision. My concern is that if we cannot evaluate - quantitatively or qualitatively - the impacts of opening data, interest will fade and investment will disappear, which I believe is detrimental to society. Some of our early stage evaluation mechanisms rely on qualitative evidence like case studies and use cases - storification. But as we are still firmly rooted in a market economy, I fear that we need more quantified methods to showcase the value of sharing data and information in order to spur more interest, investment and use of open data.

I have written earlier about the Open data value paradox  which I think is a great research opportunity. We need to further examine the relationship between the proportion of "unmeasurable" social value based on information sharing and the willingness to invest in open data infrastructure. However, as I have suggested, I think the MSPs have in a way solved this paradox without necessarily suggesting any evaluation methods for information sharing. These platforms tend to utilize the shared information to generate "secondary" profit and therefore show how the information sharing and market mechanisms can feed upon each other, creating valuable synergies.