Development phases for building services on open data

Previous article described various obstacles commonly related to service development based on open data. Obstacle is the first dimension of a model intended to gather and categories problems related to third-party development based on open data. The second dimension of the model concerns different phases that third-party developers goes through when creating services.

Our studies shows that some obstacles cause problem in beginning of the development process, while others cause problems in later stages. An example of this was the license agreement which initially could be difficult to read or understand concerning what applied regarding further processing and storage of data. Once the third-party developer overcame license agreements specifics the obstacle it diminishes in later stages of the process. Other obstacles such as lack of quality or flawed data, was not a major problem until later in the process. This was due to various reasons, such as that third-party developer did not have time to get acquainted with details of large amounts of data, or flawed details could not be detected until more extensive testing and usage of had been carried out.


The seven obstacles explained in previous article, has different implication in different phases of the service development process. Which entails a way to capture and measure the implication of the obstacles over the whole life cycle of creating and maintaining services built on open data. For inspiration of which phases to include in the model,  a combination of Agile software development and IT – service management bests-practice (ITIL) was used. The table below describes the four phases that relates best to interviews,  software and service development processes.

Design and ideationIdeas for a new service or apps comes often from self perceived problems or identification of business opportunities connected to availability of open data
ImplementationDuring the implementation the service normally undergoes several versions, from early prototypes to release candidate, and potential business models are refined and finalized
Launch and marketingBefore the release of the service infrastructure needs to be prepared, and attracting presumptive users
Delivery and continuous developmentWhen the service is released the process of ensure quality and continuous development of the service begins

Creating services on open data is slightly different from the conventional software development since data for the service comes from one or more open data owners or suppliers. This means that there is a dependency between data owners and third-party developers . When data is supplied without interruption and without flaws everything is fine, but usually also data owners need to maintain and update their services . Which leads to that third-party developers also needs to adopt their services encompassing chances. Last but not least third-party developers needs to continuously improve their services to create better added value for end users.

The next article will explain the last dimension of the model, which depicts the different categories of third-party developers that was identified in previous studies and what conditions these categories have to create viable services based on open data . The phases of third-party development is an article in the focus area of open data.

Icon made by Freepik from, licensed by CC 3.0 BY

Leave a Reply