Q. Tell us about the start of Code for SA?
Code for SA grew out of Odadi — the Open Data and Democracy Initiative — which started in 2011. We ran a big hackathon called Coding for Democracy. It was one of the first times, for me, that open data made it beyond a few discussions. It was exciting that a large number of varied groups participated. The problem was no one turned it into a career, and it fizzled out. And so, we founded Code for South Africa to focus on open data and transparency. This is not the the traditional sense of “transparency and accountability” (such as the right to information), but rather open data and proactive release of data. We spent the next two years or so exploring different approaches, focusing on the demand side. A lot of the work that had been done, particularly in the northern hemisphere, was around supply. Certainly in the South African context, we needed to drive demand. Even if we had perfect supply of data, I don’t believe we had enough of a data literate society to consume that data.
Q. Who are the stakeholder groups that you work with?
The first is media, and here we focus on data journalism. Journalists have a massive platform, and if we could get them to engage with audiences, to use data appropriately, they could deliver their stories more effectively. One of the jewels in our crown is a story called Living Wage which looked at domestic worker wages. Our focus was how we explore new ways of telling stories and how we engage audiences to encourage their participation.
Then, we work with civil society organisations (CSOs). These are the people who fight the good fight, but they don’t necessarily use data in their work. Depending on the organisation, they are often working at the grassroots level. The team at Open Up come from commercial backgrounds. This is our first foray into the non-profit world, so we still have a commercial, IT, and tech perspective. I think this is refreshing in this space. It makes for an interesting intersection. One project in this space is the Parliamentary monitoring group website. We made over a terabyte of data from Parliament publicly available. On the face of it, you would say this is a website. For us, it is more of an information retrieval system, and it is making more information about Parliament available to the public than Parliament itself.
Our work with the Black Sash focused on how you merge high tech and low tech in the same project. This project was done with the support of Making All Voices Count. Many people are cognisant of the digital divide, and that often when you speak about open data, you are excluding people who don’t necessarily have access to technology. This project proved that data is still usable by communities. You don’t need apps or high tech phones. We can still use data to promote advocacy and discussions, and empower communities. It was a cool example of thinking laterally around technology, and that is effective even in places where you might think technology doesn’t necessarily make sense.
We also work a little bit with government. This is more on the supply side. I participated in the City of Cape Town’s open data policy discussions as a public representative, and I was invited to sit on the city steering committee for the open data portal. We worked with National Treasury on a site called Municipal Money — providing access to info about municipal budgets and expenditure. Here, again the emphasis is not technology, but how people use the information made available by technology so that they can make better decisions and better engage with government on budget issues.
Finally, we are also doing community type work. Codebridge is our base of operations for that. We try to bring in not just programmers, but anyone who is interested. We bring people into same space through events like hackathons and data visualisations events.
Q. What changes does the new name (Open Up) signify for you? How does it reflect your shifting goals?
Our original name (Code for South Africa) aligned with the global movement, with organisations like Code for America, Code for the Caribbean, etc. But that name speaks predominantly to code, which means there was a dominant focus on technology. While that is the window through which we look at the world, technology doesn’t solve problems. What it does is help facilitate virtual spaces for people to engage. For us, technology is only 10% of the solution. Transparency by itself is completely useless. You can tell people what’s going on, but if you don’t give them the tools to take action, it’s actually frustrating. You’ve just ruined their day. Our focus now is inform, empower, activate. “Inform” is the open data stuff. “Empower” is reducing the friction to making things happen, such as giving people template letters to participate in the budget formulation. “Activate” is something we are exploring. We feel that we need to find a way of transitioning people from empowered citizens to active citizens. We can’t speak to that right now, we are still experimenting. Lastly, language is significant. If you talk about “Open Up”, even internally we think about ourselves differently.
Q. What advice would you offer to civic tech projects starting out? Or, what would you do differently if you were starting over?
It is hard to say because sometimes you just need to stay afloat. You won’t last very long if you can’t pay salaries. But I do think you need to think about these things strategically: is this going to derail us? Try to focus on a single issue if you can. We were shooting in all directions initially. If you look at a single core issue, it may be harder to get funding. But you are more likely to converge on something meaningful, than if you worked on dozens of projects in a shallow way. Try to go as deep as possible, as quickly as possible. The second you show impact, you won’t have too much difficulty in finding funding.
Q. How do you nurture and promote innovation in your organisation?
Yes, operationally, we have taken a “fail fast” approach. We work in two week sprints, and we try to do things as cheaply as possible. I’ve got a favourite project that was designed to fail, designed for us to learn from. You need to make space for that… I also try to push our staff to go to conferences and to write blogs. In this way, we are trying to develop thought leadership within the organisation. It also gives people more responsibility and ownership of projects.
Q. How do you manage this tolerance for failure on the client or donor side?
From a donor perspective, we’ve been very fortunate. We pitch ourselves as experimenting. We don’t deliver outputs, as much as learnings. We try to be clear about the fact that we are trying things and will reflect on them. In our consulting work, we insist that we work on an agile basis. We work on a fixed budget, variable scope model. So we work together as hard as possible to get the best value out of a fixed budget. Even if it’s not what you think you want, you’ll get what you actually need through that process.
Q. How do you document your learnings?
With each project we do, staff write a closeout report. This is focused on the learnings and insight from the project, rather than just a summary. They are also required to put together a few slides from each project. This focuses on outcomes and learnings, so that these already exist if we need to go present them somewhere. And then the final thing — and this is a new element — is taking the reports for donors and making these available to the rest of the world. That is on the pivotal “to do” list for our website. Across these, the focus is less on comprehensive reporting, and more on what was important.
Q. How do you measure success of projects?
Our approach to measurement is maturing. Initially, we established a set of organisational targets around engagement and data literacy, and the projects themselves need to speak to one or more of those targets. We needed to hit particular KPIs within each project. Now, I want to go deeper. What I really care about is: “Have we made a change in the world?” Because, otherwise, what are we doing really?