This opinion piece was written by Greg Jordan-Detamore, technologist for the Sunlight Foundation’s Open Cities Team. Through the project’s he helps cities across the US become more accountable and transparent through open data policies. This piece also appears in our digital government newsfeed.
At the Sunlight Foundation’s Open Cities team, we challenge cities to ask whom their open data is for. If it’s for the public, then it needs to be centred around the public’s needs, goals and feedback. If governments want people to use their data, they need to solicit feedback from their constituents to determine what would help make open data more useful to them.
While a number of governments around the world have done a good job publishing large amounts of data, many have little to no idea who is actually using it. Businesses? Academic researchers? App developers? Policy advocates? Different users have different needs — by figuring out who these users are, what they want, and why they want it, governments can tailor their open data programs to meet user needs.
In the last year, we’ve shifted from helping cities pass open data policies to helping them facilitate community use of open data. We’ve developed a set of actionable steps and strategies to help governments investigate data users’ needs and publish data that meets them.
A community-centred process
Around the United States, we have been piloting a new approach we call Tactical Data Engagement (TDE). TDE is a four-step process for governments to facilitate community use of open data:
- Find a general focus area
- Refine use cases
- Design a plan
- Implement an intervention
We piloted this approach in the city of Glendale, a mid-sized city in the Southwestern US.
Glendale sent us data on all public-records requests they received in 2016. We cleaned and analysed this data and found that the vast majority of requests were being handled by their Development Services department, which deals with building safety, code compliance and planning.
Sunlight and the city worked together to interview people who frequently requested development-related data to better understand their needs: who they are, what they were looking for and why they needed city data.
After synthesising responses from the interviews, we recommended the City open several datasets. Among these were building permits: many respondents were real-estate researchers completing multiple appraisals or assessments per week, regularly submitting and waiting weeks for responses to their records requests. “The longer an appraisal gets held up, the more that loan to that prospective homeowner gets delayed,” one respondent said.
“To make open data useful, it’s critical that government staff go out and talk to people”
While Glendale made use of public records requests to address data needs, cities have other options to consider when seeking a specific focus area. Some of these may include: going to public meetings, looking at results of community surveys, consulting strategic priorities or using other problem-scoping methods.
Conducting user research
When determining how to publish data for the public, government workers should be mindful of their own limitations and biases in representing the needs of potential data users. City staff alone cannot hope to represent the diversity of needs, desires and viewpoints of the public. This has an important implication for open data programs — they cannot just be a reflection of what the program manager or even the mayor wants.
If they want to make open data useful, it’s critical that government staff go out and talk to people rather than make assumptions about what they might want. As we often say, you don’t know what you don’t know. This is why user research must be central to any user-centered open data program.
A common component of user research processes is the creation of user personas: fictional characters that represent general types of users. User personas can help communicate community needs across the city in ways that are tangible and humanised. By showing a potential data user’s background, occupation, skill set, attitudes and desires, decision-makers at the city can visualise who their open data serves. Using personas allows governments to look at their data users holistically, and think about what kind of support they can offer different groups.
When working with Madison, a city in the Midwestern US, we spent two weeks conducting 36 user interviews in collaboration with Reboot, a design-research firm specialising in design research for the public sector. We created a set of six polished user personas for the city government to use as a foundation for its work connecting neighbourhood-based data users to open data.
When thinking about data users, it’s important to consider not only current users but also potential ones. When Reboot worked with New York City, they created several personas of potential data users:
Potential data users, in particular, may be less tech-savvy, so it’s vital that cities consider how their needs can be addressed.
A word to the wise: one barrier to finding potential users for open data can be the word “data” itself. Officials in Boston found this out when speaking with community members at local libraries. “When interacting with Bostonians, we found that most people seldom discuss their everyday concerns by requesting more access to city data,” wrote the city’s open data project manager in a guest post on our blog. “When asked if there was data about Boston that people wanted to see, they rarely had any requests.”
Asking about what types of “information” people want — rather than “data” — may help get around this issue. Understanding residents’ information needs may also require asking other (seemingly unrelated) questions or making additional observations. By taking a more holistic look at their needs, cities may find some community issues with an “information gap.”
Publishing open data is an important first step toward creating a transparent government and an informed community, and should always be celebrated. But we can’t stop there.
By conducting user research to ensure that our open data programs are targeted toward community needs — then acting on the insights gained from that research — we can help make open data a catalyst for action in our communities. — Greg Jordan-Detamore
(Picture credits: Unsplash/Antenna/Reboot)