“What is the use of open data that makes it relevant for everyone?” This candid question was raised at the East Africa Open Data Fest held in Nairobi in August. People from a range of backgrounds and organizations were asked to assess the International Open Data Charter and exchange views on how to strengthen open data policies in Africa by sharing experiences about what has or has not been working in the region.

Some great questions were raised about the meaning of open data: How do you say open data in your local language? How would you explain open data to your grandmother? Does open data have any relevance to the daily bread and butter issues that African citizens face?

There were an equal number of challenges raised around open data, including the manipulation of data for political purposes; the limited use of African open data by Africans; and the need to publish data in formats that people can use.

As manager of the HDX Data Lab in Nairobi for almost two years, I have had to grapple with similar questions about the use of open data in the humanitarian sector. Partners often ask us to “do something with the data” once it has been shared through the Humanitarian Data Exchange platform. Even to (small) data owners, data is abstract and messy. Learning to ask the right questions is important when brainstorming around possible data uses and applications: Do you understand who your users are? Is there a difference between what they want and what they need? Once you have built an application or product with open data, how do you measure its impact?

Our work with Twaweza East Africa, for example, was a successful collaboration. We partnered with them to visualize and communicate results of their Uwezo survey – Africa’s largest annual citizen assessment of children’s learning levels across hundreds of thousands of households in Kenya, Uganda and Tanzania. Their goal is to communicate these results to ‘a discerning and curious public’ to prompt collective action towards educational reform, turning the the data into a useful public good that can enable decisions to be made faster (see visual below).

Twaweza Uwezo Data Viz

But there have also been occasions where we haven’t been able to create value from shared data.

Most user-centered design practitioners will be familiar with Steve Jobs’ quote, “A lot of times, people don’t know what they want until you show it to them.” Or the adage attributed to Henry Ford, “If I had asked people what they wanted, they would have said faster horses.”

I sometimes wonder whether a similar problem is facing the open data user community – that there is an unidentified or unarticulated user need. Participants at the Open Data Fest suggested that open data needs to be “socially readable,” not only machine or human readable. There is also an important role for ‘infomediaries’ – intermediate consumers of open data who create valuable products that make data more accessible (such as journalists and app developers). Services like Map Kibera and the Development Data Hub by Development Initiatives are such local examples.

It is time to go beyond looking at open data as specific file formats in greater quantities and to place ourselves in the shoes of open data users, working to better understand their needs. A participant offered that open data is “love made in numbers.” Some say that love cannot be defined, but I’d like to think that we’ll know it when we see it.

I will be attending the International Open Data Conference next month, sharing impact stories from our work in East Africa. I look forward to continuing the conversation around the value of open data in Africa and globally. You can follow the work of HDX on Twitter at @humdata