As discussed in the influential book “Information Systems Management,” open data represents the mending of the traditional separation between public organizations and citizens. “Mending” refers to integration of citizens using government services and the government agencies providing the services. With this definition in mind, what does it mean to have an open data program that can mend the separation? Publishing effective open data to eradicate the gap between organizations and citizens requires a framework to identify and make available datasets, as well as assessing the effectiveness of datasets after they’re published.
Effective open data starts with a clear view of the relevancy and use of the data being published. Relevant data denotes data that citizens desire to use for purposes of promoting citizen services and governmental transparency; once it’s available, the community will consume it. Useful datasets are valid and accurate, containing information that citizens consume for purposes of promoting civic engagement, entrepreneurship, and better government.
Once an organization has a view of the relevant and useful data, the next step is to actually find it. This is a challenging task, but programs that invest in this activity, and spend time understanding the data their constituents want, tend to be more successful and draw citizens to build new applications and drive decisions from the data. We’ve found that most relevant and useful data falls into three basic categories:
- Public services such as police, fire, rescue, and health: These datasets tend to directly impact quality of life and are in high demand.
- Citizen services such as inspection results, permitting, licensing, and inspection: These services contribute to a locality’s economic diversity and vitality.
- Financial transactions such as budget expenditures, capital projects, and contract expenses: Citizens interested in financial transparency, who want to highlight the use of taxpayer funds within the locality, typically use this data.
Once you have good understanding of your organization’s relevant and useful data how do you make it accessible to citizens? Keep the following key concepts in mind as you begin to publish data:
- Ensure the data is easily understood: Provide helpful explanations of the data, including human readable column names. Publish data in a timely and periodic fashion: Users who consume the data can depend on its accuracy, timeliness, and ease of use if it’s updated and published on a consistent schedule.
- Provide data in machine-readable formats via downloads or API integration: This allows users to easily consume the data for research or application purposes.
Is the Data Being Used?
Publishing effective open data to eradicate the gap between organizations and citizens requires a framework to identify and make available datasets, as well as assessing the effectiveness of datasets after they’re published.
Now that you have found relevant data, published it in an easy-to-use format, and made it available in human-readable terms, the next question to focus on is “how do I know if the data I’ve published is even being used?” Answering this challenging question requires analytics. Conduct analytics within your platform to continuously evaluate popular and trending datasets. Utilize services like Google Analytics to monitor your open data website and evaluate how users are interacting with the site. If you make your data available via APIs, monitor API usage for popular datasets and use case stories (e.g., the types of applications that developers are building).
Key to understanding if your data is being used effectively is to have a comprehensive governance process that continuously evaluates your published datasets for potential changes and updates. Effective governance programs also establish a process for citizens to easily suggest ideas for new datasets. Most important is to manage the publication process so that citizens can depend on accurate and timely information.