Combining empathy and data to transform San Francisco

This interview is part of our policymaker interview series. In this series, we talk to policymakers from across the world about their policies, policy-making and life in government.

As befits a city located so close to the global tech powers of Silicon Valley, San Francisco has long been seen as a city that has pioneered the use of data and digital technology to strengthen citizen services. That's partly why Joy Bonaguro joined the city and county as its first chief data officer - that and the opportunity to take San Francisco's use of data to the next level by embedding it into all levels of decision making.

Through her work managing DataSF, she and her team are ensuring that city leaders and employees are using data to continuously improve services and operations - but how are they doing so? What barriers are they coming up against? Why does she think empathy is so critical?

We recently caught up with Joy where she told us about empowering the use of data - San Francisco style…

Can you tell us a little about your role?

My role started with two broad mandates. One was to reboot the city's open data programme. We'd had a programme since 2009 with lots of successes, but it needed some love and nurturing. And then the second was a broad mandate to improve the use of data in decision making across the city.

The first one was pretty straightforward in terms of coming up with a game plan. There were some things that we needed to do. But the second was a broad mandate that needed to be shaped and formed. And so we went through an engagement process where we worked to understand the barriers against increased use of data in decision making across the city. Once we had diagnosed and articulated those barriers, we needed a strategy to systematically dismantle them. So that's what we did.

As part of our engagement, we did a lot of interviews and then we ended up essentially quantifying that via what is now an annual survey. And then we shaped our strategy to tackle those challenges progressively over the last four years.

So, what are the barriers to increased use of data in decision making in San Francisco?

The challenges broke up into three buckets.

First for context, the city is highly federated. In the US, we're both a city and a county. It's very unusual to have both a city and a county be equivalent contiguous geographies. We are the only one in California. Usually, multiple cities are contained in a county. So we have both functions. This means we're larger and more diverse in terms of our functions. We joke that we run everything from A to Z, an airport to a zoo.

Two of the key data barriers are simply knowing what data exists across the county and having efficient and effective means of accessing that data. We have federated technology systems that grew up in their own silos and people don't always understand why data is needed to be shared across different departments.

And then there are legal issues to accessing data. In the US, our privacy frameworks are essentially sector by sector privacy frameworks. We lack an omnibus privacy law. Because privacy rules vary, it is hard to figure out the right legal pathway for data sharing across departments. And the reason why we care about data sharing is that when departments are serving the same population, if we don't understand the mix of services and how well they're working, we can't serve them as effectively as we would like.

Part of our strategy works on tackling that from the goal of making data easily accessible, which is both about the infrastructure, but also governance in terms of making sure that that data sharing is appropriate, responsible, secure and respects privacy.

The last bucket of problems is around what we call the ‘ability' problem. If you have data, you need to have the skills and capacity to use it effectively. One of the things that we worked very hard on was lifting and levelling data skills via Data Academy, so that we can have similar skill sets across departments. And then over time, we continue to evolve and advance those skills.

Ultimately, solving knowledge, access and ability challenges is in service of putting data to work. In addition to increasing skills and capacity, it's changing capacity. That's where we're working on ways to structurally free up different parts of the city so that they can start working on higher-order analysis as well as providing data analytics services. So, for example, our most recent service is DataScience SF, where we provide data science modelling and advanced analytics on public policy problems with an eye towards implementation change.

Do you feel like this programme has been effective, and if so, what does that mean in that context?

If we're going to be data driven that means we need to evaluate ourselves. For any new service we offer, we seek to evaluate the impact. We have a logic model about our work relates to improving quality of life, economic opportunities and city services.

But obviously, it's very hard. Some of the programmes are harder to measure than others. One of the very first things we did was develop an evaluation plan for open data. We liked to use the results-based accountability (RBA) framework for all of our measurements, which helps you define metrics around quantity, quality and impact.

What we've done for open data is carve out a core value proposition of open data that was a little bit different from the transparency movement that spawned it, and focus more on driving internal use of our open data platform. Because that's solving our knowledge and data access problems.

If we drive it as an internal tool, it will drive investment. It will increase the quality and quantity of data, and make it more of an institutional tool and not an external policy agenda. That was a very explicit strategy. I think that was something new that we brought to the field. No one was talking about that at the time. I think it's pretty common to talk about it that way nowadays.

Have you learned any lessons about how to influence others to drive effectiveness?

I don't know if it's lessons or more of a philosophical approach. We never came in and told people this is how you should be doing your work. I find change agents or political appointees to be unhelpful when the message they send to people is that they're doing things wrong.

A strong philosophical approach that we adopt on our team is empathy. We indoctrinate everyone into design thinking, which is about empathy and being centred on the user and where they're at. We very much try to understand where people are at, understand what their issues are. We try and meet those before we expose them to new ways of doing or thinking.

So I think a humble approach is best, where you minimise the change and the difference between you and the people you're working with. I think there are too many overly “sexy” people trying to change things in government, and I think they actually mostly annoy the civil service class. If you can instead approach change from a position of humility, I think that it's a) more effective and b) more appropriate.

I am stunned every time I read a news article where someone gets appointed to a position and talks about how they're going to shake things up. And I'm like, wow, you've just alienated everyone who's working for you. Policymakers, I think, need to be careful about the people and the personality types they hire into these positions. It'd be interesting to see if someone could quantify the personality, tenure, and impact of the average quote unquote change role. In any complex organisation with legacy systems and people, change takes usually at least five years.

Why do you think empathy and effectiveness are linked?

Why do I think empathy and effectiveness are linked? Because fundamentally you are doing change management. In order to change, you need to understand the people and the systems in which you're operating while keeping in mind opportunity and the broader end goal. So you need to understand the perspective of the people who work in government and why change is hard.

I think most people sort of dismiss and hand wave as opposed to listening. Listen and then make them part of the problem-solving. Empathy is a really a tool to understand where people are coming from and it's actually about effectiveness. If you actually want to be successful in delivering your value, and where you want the organisation to go, you have to bring people with you. So I think that's the link to effectiveness.

Why do you care about government being the best it can be?

When I moved down to New Orleans, I discovered true poverty and realised that I was mostly a product of the privileged circumstances that I had been born into. At the same time, I was a philosophy major, and became familiar with concepts of fairness, opportunity and the veil of ignorance. The veil of ignorance is the concept that your society is just if you are willing to be born into any part of it. Our society is not just.

In government, we're dealing with some of the toughest problems of our age. I love problem-solving, and I love juicy complex system-wide problems. Anyone who thinks these problems are simple, or there's a solution just waiting to be picked up and grabbed, is kidding themselves. It's so important to strive for a more just society - and that's exactly what I hope to do.

This interview is part of our policymaker interview series. In this series, we talk to policymakers from across the world about their policies, policy-making and life in government.