Lessons learned from talking #OpenData with our Municipality

Last year, in one of the first projects we undertook as an organisation, Open Data Durban, partnering with the eThekwini Municipality, Web Foundation, Palmer Development Group and OD4D, set out to assist eThekwini Municipality in creating an action plan for using open data in water management in the city. The idea was to get people from the water and sanitation department within the city into a room with environmental specialists, researchers and academics in the water sector, citizen-scientists and organisers of water-related CSOs and community groups, to assess the general level of awareness of open data in the group and address gaps in knowledge. The intention of this was to break down the gaps and challenges faced in access to and use of data in water management and create an action plan around how open data could be used to drive innovation and improvement in city water management.

Our methodology for identifying knowledge gaps included topic driven workshops and a number of one-on-one interactions with key people from the groups listed above. We were very intentional about going beyond just dialogue, and spent extensive time mapping and breaking down data needs and processes employed in the current applications and water management practises. We spent a lot of time talking and working with senior managers in the municipality and through this interaction, gained insight into the context of data analysis, data sharing, and open data within local government in South Africa.

From a process perspective, we engage with many city officials and cities on other projects, and from this project and others, our knowledge and insight into how municipalities and officials relate to open data has thus become more fine-tuned. Our conclusions and experiences, in many cases, confirm a lot of what we have read from other open data organisations working with government from around the world. We’d like to share some of the lessons we learned through the process around opening data in a South African local government context.

Firstly, a major lesson learnt was that if something does not line up to a KPI (Key Performance Indicator) it will be difficult to get someone to do it. There is a pretty rigid performance-based structure to job descriptions and functions in a municipality, driving employees to work within a system where they are directly rewarded for achieving KPIs. Performance outside of KPIs are considered unmeasurable and therefore, have the potential to be undervalued. Unfortunately, KPIs are difficult to change and not always aligned to the dynamic and multi-functional thinking required to not only perform but succeed in a role (which in my limited experience is no different to a corporate environment). On top of this, in a lot of cases, city departments are under-resourced, and staff procurement (hiring people) can sometimes take up to two years. Imagine having to perform a role where other critical roles (up to 50% of a unit in some cases) are vacant for extended periods of time, and where your KPIs (which are directly related to reward and promotion) are dependent on those vacant roles. Your only option then is to pick up the slack necessary, which in many cases means double or triple the work, and performing functions that take away from your role and are unmeasurable. Open data is not currently related to a KPI, neither is finding new ways of doing something or innovating on tasks. So where you find people within a municipality willing to try new things, we’ve learned that these people are your champions, as their mentality is to serve the city rather than just work for the institution.

Our next learning moment was that opening data carries some real risk and a lot of perceived risk for a city official or municipal staff member, and many times the perception is that it is much safer to do nothing (i.e. keep data closed). If your job and/or status and/or funding source is enhanced by opening data and cool civic technology this risk can be easy to forget. Currently, if a city employee or department opens data, there is a chance they will get credit or an award, but nine times out of ten either nothing will happen and it will just be more work for no reward, or they will share something which puts their job at risk, even if the data is not particularly risky but it happens to bring a negative mention or outcome for their department.

Overall, we found that challenges to implementing and using open data in government can sometimes be:

• A skills shortage challenge — as with most organisations in South Africa, there is not a widespread level of data science and technology literacy
• A capacity mismatch challenge — as some roles in local government, especially technical roles are overloaded, so although that person is well aware of the opportunities open data or data science or civic technology offer, has no ability to follow up on and implement them
• A resource challenge (the procurement issue mentioned above)
• A risk vs reward challenge

We also found that part of the challenge is in our own misconceptions in the way we think about government as the homogenous, inefficient entity responsible for the development challenges communities face. In my experience as a private consultant and now in my work with Open Data Durban, I can honestly say that I have met, worked with, and learned from some of the most impressive people in their respective fields working for the city. Seeing local government as a single entity and painting it all with the same brush is hugely destructive, as it just adds yet another barrier to those people within city departments that are fighting for change. Part of my personal motivation in founding our organisation was to be able to better collaborate with some of the people within government that are actively innovating and working for openness. At the same time, this is not to say there aren’t hugely problematic practices and people within government, but I would make the same assessment of the business and CSO sectors. It is apparent to me that we need to rethink our approach to government to be able to the innovators and change-agents better rather than leading with the preconception that government is inefficient and resistant to change.This change in thinking directly relates to how we want to frame open data in government to be more about advocating for participatory governance and democracy in action challenges with the same design thinking and user-centric practices we would any innovation project.

We also need to be careful of the impact of “open-washing” (meaning data publishers that are claiming their data is open, even when it’s not — but rather just available under limiting terms) in projects with local government (and any others). This project and others have taught us that municipalities are not blind to the fact that opening data alone does not give any guarantee of impact and that even with open data usage, engagement and impact is difficult to track. If impact is difficult to track this means it is also difficult to report, which translates to spending departmental budget and taxpayer money on something which is presumed to have impact (and trumpeted as “success” by organisations in our field) but where the impact can’t be proved. This is unlikely to incentivise municipal staff to do anything. What will incentivise the municipality to take positive action is to find how open data and use of data can link to improving KPIs, IDP and other development goals, or save time and stress in accessing, using or reporting on data.

For example, we noticed that there seems to be a lot of requests on the municipality to share data from research institutions, other government departments, and international organisations such as the World Bank. This is an area where some open data initiative could really help protect municipal resources against repeated work and overloading with requests. There is some scepticism within municipal units that a generic or automated process would work in reducing the ad hoc requests from these institutions, however. One example raised was that national government (for example, Treasury) requests data that doesn’t exist naturally or is/would be expensive to collect, and this causes a lot of work and stress for municipal departments. Another is that it seems easier for certain international financial institutions and organisations to get information from municipal departments than for a unit from their own department, and because of the (slightly strange) prestige that assisting an international organisation seems to carry, municipal departments prioritise this.

A final, parting thought from this project was that if you aim for open data, at best you get open data, which is not equal to impact. If you aim for civic technology, at best you get a tool (app, website), which is also not equal to impact. There needs to be a greater focus on framing these sort of projects within government (and in general) around accountability and participation, and with a focus on solving a citizen defined and centred problem. We have to stop thinking that data (open or closed), or technology, will solve a problem, but rather that these will facilitate the connecting and scaling of a solution focussed on solving a human problem.

We’re digging deeper into how we can address the space between citizens and local government and incorporating these lessons into the work we are doing. They speak to longer term initiatives that don’t necessarily have immediate, shiny outcomes, but ones we feel have a chance for actual impact.

*A google site was created for the project, which can be found here

- Rich

Richard Gevers is the founder of Open Data Durban and brews beer in his spare time.

Originally published at Open Data Durban.