Opening data for social services

Anne White

Anne White is a Senior Policy Advisor with the Privy Council Office at the Government of Canada. She has also been a Government Innovation Fellow with the Harvard Kennedy School of Government’s Government Performance Lab.
Listen and subscribe to the podcast on iTunes or Google Play.

Welcome to Countless Rebellions, the podcast series that explores the limits and potential of social innovation with academics, practitioners, and activists. This series is produced in Montreal, on the traditional and unceded territory of the Kanien’keha:ka or Mohawk peoples, a place which has long served as a site of meeting and exchange amongst nations.

In this episode, I sit down with Anne White, a federal public servant, to talk about data, facts and how constant incremental innovation can be even more powerful than the disruptive innovation we typically hear about. 

White: I am a social innovation advocate, I guess, and one-time practitioner — certainly, in the States. I’m trying to do that in Canada now as well. 

What would you say your project is? What are you trying to achieve through your work? 

White: My project… Well, I think the first one that I’m focused on at the moment in Canada is trying to unlock administrative data for use for social services. That means a couple of different things. On the one hand, we could just unlock it in a way that we understand in an anonymized way — a relatively micro-geographical level of what’s happening. Toronto’s done that at their city level — you can track a lot of stuff. You can do it for different infant mortality rates, or child abuse, etc. There’s a lot of power for social service organizations to understand their communities from that perspective, and to be able to understand where they might do greater outreach, for instance, to find individuals that need help. That’s an easier way, I think, of unlocking data, but what I would really like to see happen is the ability for Canadian Social Service Organizations and Public Health Organizations to partner with government to really understand whether or not their services are improving outcomes or impacting individuals.  

The kind of stuff I’ve worked with is infant mortality or child protective services — they’re important services. And you don’t want to just change everything on a dime. Yet we have all this amazing information that tells us with scientific rigor, that if we only do X, we’ll see vast improvements. It seems to me it’s a no-brainer.

What is social innovation?  

White: So, it’s a great question. I think everybody has a different definition. To me, it’s incrementally trying new things to make change. And sometimes that change can be not incremental, it can be absolutely radical and we can see big improvements, or it can fail, and we learn from that and go in the other direction. But I think it’s constantly dynamically changing the system — changing something that you’re doing to learn, hopefully, to make things better.  

And those little incremental changes, how do those add up to something greater?  

White: That’s a big question. I mean, it depends on the service provider and, in my work, it depends on the government partner, in terms of what they choose to do with it. Incremental change if it’s dynamic and the cycle doesn’t end — which is the ideal — is that you’re constantly refining, right? And so you understand what the best evidence is, whether it’s a human development model, all the way to the beginning of the theory: you amend your practices, you see how a person responds, and then you keep changing. As you do that, if you can feed that information back into government — let’s say if you were able to partner with them on a referral system — then they also will be tweaking that. So you’re really getting to a point where you know Jane just walked into the door, Jane would really benefit from this program with this potential add-on, right? We’re not there yet but I think that you start to optimize, right? You know people, you use the information that you have to know what they need and to respond to it.   

There’s a lot of focus on disruptive innovation, and I think, what you’re talking about is piecemeal or incremental, constant innovation. 

White: Constant innovation. Yeah, it’s interesting to think about whether or not it’s disruptive. And I think that depends greatly on where the service provider is when you start the initial assessment. Because if their practices are very, very different from what we know to be helpful outcomes — or helpful practices to improve outcomes — then that would be very disruptive for them because they would have to change a lot of their practices.  

For the most part, in Canada — so far anyway, in terms of what I’ve seen — the practices aren’t so far off of what they should ultimately be. But it’s the iterating, and the how you gather data, and the follow-up — it’s all the small stuff that needs to be refined. And, yeah, so it’s constructive, and it never ends. I think that’s part of it — that’s the part that not everyone is prepared for when they embark on one of these. Performance improvement is where it begins right, before you actually do the outcome validation-type evaluations. But it just never ends. You just need to get into this cycle of “We know this” and people change, right? So it should never end — using the information that we have to understand what the needs of the individual are, and then looking at our toolkit and saying, “Okay well, for this individual, we need to alter this service or think about calling this person that could maybe help”. Or if it’s an employment agency, making the extra step to find something that would work for that person. You’re starting to customize services on a population-based system, which is a bit tough, but it’s totally doable. Once organizations get into the habit of doing it, and have enabling structure — and they usually need a little bit of help to get there, it doesn’t have to be a fancy IT system, but just to start to collect the information that’s most pertinent, and use it in a way that allows them to personalize services a little bit — then you have very constructive innovation, and they see it, right? Because the people that they’re serving are happier, and they’re seeing better outcomes, and that’s where you get the real positive cycle they want to continue. It just takes a little bit of a boost to get there.

I think that the link between having these programs, or projects, or approaches of social innovation being one-offs versus having them reform the way we do public goods, frankly, is essential. People dismiss administrative data — it’s not the sexiest topic, I understand — but it’s the only bridge to make the two realities make sense.

There’s this one project where they were looking at expanding a service into an area that had high need, and they literally used Google Maps and some of the data, anonymized — nothing individual from that perspective — but just to see, if we’re expanding service in this area, where’s the need and where’s the population moving, and where are we seeing high levels of infant mortality? The service provider, this one nurse, sat and looked at this map, and she’s like: “Oh that’s where the new refugees are moving. There’s a church there. We could spend more time doing outreach there on Mondays.” And her eyes just lit up. That small bit of information you could say is constantly disruptive, right? It’s a different way of doing outreach. It means that people on the ground are going to knock on doors potentially, or spend time at the church, instead of what used to be more of a passive system where people would come in. Maybe that’s disruptive. But it’s exciting for people when it’s useful information. And we didn’t get into this, but a lot of the data that’s being collected at the moment is for different reporting purposes, for different foundations or government bodies, etc. And in the States, I think there are layers of hierarchy reporting all the way from the local level to the federal government. So, it’s a lot of different data points that are being collected that frankly aren’t useful for the service provider. Getting to a point where using that data and making it functional is exciting for them.  

In the community of social innovators, where system disruption and radical idea-generation is revered, the idea of constant, incremental change is not a common practice. Why is incremental change considered an outsider’s approach?  

White: I don’t know. It’s kind of funny. You’re making me think — what’s creeping up in the back of my mind is — maybe I’m risk-averse. Right? Maybe there’s a part of me that kind of goes, “We got all this great information, let’s start there. You know, we don’t need to reform everything and make everyone take risks.” Because it is. The kind of stuff I’ve worked with is infant mortality or child protective services — they’re important services. And you don’t want to just change everything on a dime. Yet we have all this amazing information that tells us with scientific rigor, that if we only do X, we’ll see vast improvements. It seems to me it’s a no-brainer. It’s an opportunity to improve without actually laying it all out there and, you know, ripping it up and starting over. It’s really focusing on what we know we’ll improve.   

What is tough about this kind of work? 

White: For me, I’ve seen in experiences that I’ve had so far that people tend to respond with a no more often than a yes. It doesn’t even matter what it is, right? But people want to be comfortable with what you’re proposing and until they figure out whether or not they’re comfortable with what you’re saying, they’re going to say no. That can just be wearing after time. You know, it’s fine the first meeting in the day, and it’s fine the second, but when it’s the sixth time that you’re working with the same people, and they’re still saying no, I think that’s challenging. It’s easy, or relatively easy, to cut through that if you have the right tools. So, back to the example of showing this nurse the data and she could right away see the church and the outreach activities and got really excited, that cuts right through it, right? Because now you’re not talking about disrupting, you’re not talking about new processes. You’re showing them very useful information and they automatically create their new process themselves because they want to. That’s powerful.  

The challenge really is where you don’t have that yet — where you’re trying to sell something that people can’t imagine. I find that I’m in that world a lot these days. It doesn’t matter how many white boards I put out and how many times I draw it out, until the individual you’re speaking with can go to the service provider, or imagine it, or spend time and understand this is what we’re asking them to do and here’s how it ties back to local government, to the province, to the federal government —  until they can see that … it’s almost like you need to take them through the factory, then you’re trying to sell something that’s theoretical and, as a result, it’s not concrete. It’s very, very hard to sell things that are not concrete. I find that tough. That’s what will drain my energy. But on the flip side, the service providers always bring it [the energy] back, so that’s where you get the fuel to go back to the front line.  

I was curious to find out what Anne thinks is a common misconception about social innovation.  

White: I think that the link between having these programs, or projects, or approaches of social innovation being one-offs versus having them reform the way we do public goods, frankly, is essential. People dismiss administrative data — it’s not the sexiest topic, I understand — but it’s the only bridge to make the two realities make sense. The challenge that we have in terms of accounting for prevention is that our system is set up such that you need to hurt yourself and have a symptom before we’ll invest in you. Any service provider that works with families closely will know — and many have evidence —  that we need to do more. There’s opportunity — when a family is going in a certain direction, or health is going a certain direction, etc. Our accounting system in government is created on a yearly basis, so it doesn’t allow for that sort of thinking: “Yes, we’ll invest today in order to save them three years from now.” On top of that, it’s all siloed in different departments. If department X invests and department Y gets the benefits, that doesn’t work either.  

We have these systems that perpetuate challenges and create these services that we have, and they constrain our service providers. I don’t think that people think about that enough when they’re trying to reform the system. On the one hand, you could say, “Well, forget about it, we’re just going to build a parallel system and eventually this one will fall.” But I’m not sure that that’s practical. It seems to me that you need to bridge the two, and you have to get the one system to enable the other to rise up. That creates a world where our services are more effective, our outcomes are better, and it’s all cheaper in the end. It costs a lot less money to do it that way, using the evidence, using the data. And, that’s not fun. It’s not fun to think through all those sort of accounting problems —  like, “Oh, how boring”. But I think it’s really, really powerful. It is the one thing that I wish I could find a way to talk about with greater energy or pizzazz or make more interesting for people because fundamentally I think it’s our biggest barrier to really creating change.  

After working in this field for many years now, Anne is hopeful for the future.  

White: Like everyone, I read the news, and it’s pretty negative. That can feel challenging. But my experience actually makes me feel quite positive about the world. In the States, I worked with folks who were on very different places on the partisan ideological spectrum. At the end of the day, when you get very concrete about a service for a person, when you’re looking at real data and it’s no longer theoretical  — “If I do X, this person will have something” — when you’re actually looking at the data and you say: “I helped this person. And hey, look! I can see they used the emergency room less.” Or the interactions with child protective services were less versus this other group. Well, there’s no more room for guessing. And ideological differences evaporate at that point because it’s real. You can see the differences. And you can attach costs, and savings, and benefits. You can have a very concrete conversation. And it’s a lot easier to deal with some of these challenges that we’re trying to solve in the world when you have that sort of very concrete sort of understanding about what’s possible.  

[Outro theme song]  

Thank you for listening. My name is Scott Baker, this is the Countless Rebellions Podcast brought to you by the McConnell Foundation. To hear more, be sure to check out the other episodes. And if you heard something interesting, share it with someone whom you think might like it. Countless Rebellions is produced by Adjacent Possibilities in collaboration with Brothers DePaul. To learn more about the McConnell Foundation and the work of their grantees, visit