← Data-Smart City Pod
Data-Smart City Pod
Good, Responsive Services Start With the Resident
00:21:26
Good services don't start with the city's org chart or budget lines. They start by understanding the resident's actual journey — and all the hidden time, paperwork, and friction that comes with it. Host Stephen Goldsmith speaks with Dr. Kim Leary, clinical psychologist, professor, and director of the Good Services Lab at the Bloomberg Center for Cities at Harvard University, about how listening is an active skill, why selection bias shapes who gets heard, and how cities can use AI and resident-centered design to create services that actually work for everyone.
Loading summary
Transcript25 lines
- [00:03]
A
From DataSmart City Solutions the Bloomberg center for Cities at Harvard University, this is the DataSmart City Pod. In each episode, we bring you conversations with experts on public service innovation, the application of data and artificial intelligence, and responsive city leadership. Here's your host, Professor Stephen Goldsmith.
- [00:25]
B
Thank you and welcome back. This is Stephen Goldsmith with another podcast. And today I'm excited to Welcome a colleague, Dr. Kim Leary, to the podcast. Kim has titles that are longer than the podcast. She's a lecturer at the Kennedy School. She's an associate professor at medical school and a public health school. She's director of the Good Services Lab. She was in two White Houses, many other distinguished qualifications. So first of all, welcome, Kim, Steve,
- [00:55]
C
thank you for having me. And it's so good to be in conversation and in community with you.
- [01:00]
B
So what's the common thread among all of those? Multiple professions, multiple credentials? What motivates your work?
- [01:07]
C
Well, you know, I've been at it for a bit, so there is a long arc. But I started, as you mentioned, as a clinical psychologist, as a psychoanalyst, really doing what I think you would call and I would call as well, hyperlocal change. But as a psychologist, that hyperlocal change means working with one person, one family, one small group at a time. And in the years since then, over the last 15 years or so, so I've spent time trying to figure out how to do something comparable, but at the scale of organizations and systems.
- [01:41]
B
Well, you're certainly doing that now. You have this terrific new lab at Harvard. It's got an interesting name, Good Services. That's aspirational. What do you mean by that?
- [01:53]
C
Well, maybe I'll go back and say a little bit more about how you go from being a clinical psychologist to a good services lab. So you may know that for about 12 years I was the chief psychologist at one of the safety net hospitals within the Harvard system, the Cambridge Health Alliance. And that hospital serves immigrant, refugee and lower income communities. It also serves people who experienced political violence and political torture in some cases in other countries. That work about caring for people in a system did take me to the White House under both President Obama and President Biden, also to the Urban Institute, which is a think tank based in D.C. and now the Good Services Lab. And whether it's a patient or a population, I think the work is similar, helping people to navigate systems that may not have been built with them in mind and diagnosing those systems where intervention is possible. This is work your program does as well, so they don't have to fight so hard to. To be served. Now, one of the things I'll say is that the comparable value in those different contexts, and I know you know this from your own public service, is that listening is so critical and that listening is an active behavior, whether you're a clinician, a mayor, or an academic for that matter. You earn the right, I think, to be in certain conversations and to lead people by understanding where they are from first and what they want. That gets me to the Good Services Lab and to your question. Good services is plain and simple, what it sounds like in the context of cities. It's a service that a resident can find, understand, and actually use to get the thing they came for done. And it means they don't need to have specialized insider knowledge. They don't have to ask for additional help. The service provides them with everything they need if it's a good service in order to accomplish what they set out to do.
- [04:01]
B
I have a lot of questions for you about good services, but let me just go back to your introduction for a second. First of all, the idea of having somebody on this podcast I'm interviewing as a psychoanalyst seems quite intimidating. So I'm now very sensitive about the meaning of each one of my questions. No need.
- [04:17]
C
No need.
- [04:19]
B
Your hospital background leads me to a question. I was at a service event that Carney hosted where the governor of Idaho spoke right before I did. And his perspective was really fascinating, but I think it was shaped by his missionary service. And it feels to me, your comment about listening and your work in a hospital. Much of the problem with our polarization today at the city level is a lack of understanding of the context in which others live. Right. The struggles that they have. And so it feels to me like people who are fortunate enough, like we are, to have had experiences in multiple different communities have a perspective that causes them to want to deliver good services. So it's a long winded way of saying, what could we do to increase perspective and understanding and context? I think that would drive a lot of common civic aspirations.
- [05:20]
C
Yeah, I couldn't agree more, really. And a number of our colleagues, of course, are working on exactly that challenge. How do you have constructive conversations? How do you disagree constructively? I do think it starts with listening. And I said earlier that listening is an activity, and I think that's a good way to think about it. We tend to almost assume that people are speaking. Something is registering in our ears and we've heard something. Well, we may have heard something, but it may be actually our own presuppositions we may not have let exactly what you mentioned, the texture, the detail, the granularity of someone else's experience, actually register within our own ecosystem. And how do you prepare people to do that? I think sometimes our negotiation colleagues have a good model. You know, before they start to get down to problem solving, they often invent without commitment. They test out. Did I hear you? They play back what they think they've received, and they do a lot of that groundwork first before they jump into problem solving, before they jump into actually trying to negotiate a settlement or a better outcome. So I think the first thing is recognizing that there's the biology of listening, but then there's also just the experience of taking in something that's different and new and allowing that to register.
- [06:57]
B
I was a mayor once. I think you have to work at listing when you're a mayor because you know, the apparatus around you suggests that you have some special insight when you really don't. Right. So what do you teach mayors or senior deputy officials about listening and understanding their constituents?
- [07:16]
C
I hear what you're saying about mayors having an apparatus around them, that they have to be responsive to so many different constituencies. They may have to be responsive to their predecessors in office. And they most certainly these days engaged with a range of challenges that no one has prepared them for. And with the velocity of change, including the role of generative AI and where it's going to change all of our workflows and work streams, I think it's a lot to parse out and to make sense of all of the incoming information. The most important thing, though, that a mayor has, in my view, is a staff around the mayor, a staff who is busy trying to filter out what the mayor most needs to know so that when the mayor does meet with residents, the mayor can be aware of where that community is, where the pain points are in that community, and what residents most want the mayor to know. You may remember during the pandemic when the Bloomberg Harvard City Leadership Initiative had to change up its programming for a year, going from teaching a particular cohort of mayors to making programming available to any mayor across the country. I remember many of our programs at that time were sort of exploring what mayors needed to do under those conditions. They had to be an epidemiologist in a certain way, they had to be a public health professional, and they had to be the chief consoler amongst all of the challenges that were taking place. So I think what we try to do, what I try to do, the work I do with mayors, is to first help them to recognize that their constituents, their residents, are looking at them for cues, for cues about what the resident should do next. And that is sometimes to gather more information, sometimes it's to offer advice. And sometimes residents are looking for cues about whether the mayor has been willing to. To take in their story and whether or not that story has landed in that context.
- [09:35]
B
And good services. What are the one or two things that you think would best identify the gap in the city? What's keeping it from where it is and better services?
- [09:47]
C
I think that's a great question. We spend a lot of time, my team, working with cities, as we do in the Cross Boundary Collaboration Program, which is part of the Bloomberg Harvard City Leadership Initiative on the diagnostic process. What's the pain point now? What's the ache that the resident is experiencing? But then what surrounding systems are implicated? How do we begin to put together a picture not just of what's wrong, but how it came to be? I think we both appreciate that often a current pain point was yesterday's solution to some other problem. And so we have a process, we have a procedure, we have a way of doing things that made sense in an earlier context, but doesn't fit right now. Right now, I would say in many communities, residents feel that only certain people get listened to. The more organized, those that are most available to be a part of civic engagement processes, those that are most comfortable, most vocal, most eloquent at a public meeting when they grab microphone or are handed the microphone. So part of being responsive is recognizing that the people you're in conversation with as mayor or as senior leader may represent a selection bias. Not one that you had a hand in, but one that happens where the squeaky wheel gets heard and where the person who's most comfortable at the mic is the one who's often turned to and receives more airtime. So part of it is figuring out, how do you get more voices involved? And that's simply a design question. How do you know where a service is working while you listen to the people who are using that service? Who is it hardest for? You have to maybe go a little further to get their stories in the mix. And if you design a service for those it's hardest for, there's a good chance it's going to get better for everyone. You know, the curb cut effect.
- [11:47]
B
I used to have these meetings as mayor on, you know, important policy issues. You know, staff would come in and at the beginning of each meeting, I'd say, okay, what's the most important constituency that's not in the room. Who's representing them or have you talked to them? And there's always a key constituency that should be there that isn't just always there's a gap. What do you think the role of AI has in this conversation in helping a city listen better in AI in terms of responsiveness?
- [12:18]
C
So I'm fascinated by this and very much in a learning mode myself. I've run an exercise in a couple of my executive ed classes and mayor classes where we've actually asked city teams with respect to whatever issue they're working on to name the relevant constituencies. Then we've engineered a prompt that if they have leave to use generative AI tool in the city, because not all cities are able to do that, ask them. Well, okay, here's who I came up with. Here's what we're working on. Who have we not even considered? And even a short exercise like that typically generates organizations, constituencies, neighborhoods that just simply were not on the radar screen to mix metaphors, a bit of the city team. So I think generative AI as a tool can help us to expand our knowledge base. We have to check it out. Of course, an AI tool might give you a finding of an organization that doesn't actually exist. So there's some checks and balances here. But I think AI has a real potential for enhancing our understanding. The second thing I've become interested in is how the private sector has often used sentiment analysis and the reading of digital exhaust, if you will, to get a sense of many different kinds of data, to think about the product that they would like to sell or offer. Well, I think we can do that as well with cities. And some folks are already giving this a try. The Muhammad Ali Center. The Ali center in Louisville is starting to use a tool to identify what compassion might look like in cities. A compassionate city, we would imagine, might have other characteristics and might design their services a little bit differently, but we don't know that yet. We just have a reading of cities that can be classified along many different dimensions. But if compassion were one of them, that might be interesting for us to explore.
- [14:28]
B
Think about what you just said. We have this project with Robert Wood Johnson foundation and Knight foundation where we're trying to use AI to help cities and communities engage more effectively. Because much of government community engagement today is just trying to convince a community that you're right about something you've already decided. So if you thought about responsiveness and good services and engagement the way you've talked about, how could we assist communities in being partners in this solution?
- [15:00]
C
Well, I think it may start first with how you frame the challenge before you even get to the tool. So in a lot of my work, we focus on inclusive resident impact to take account of those missing voices. But we're also returning to teams and to their engagement with communities, not just as a consultation exercise, though consultation is important, but with the assumption that people who are living in neighborhoods know things that you and I may not by virtue of the fact that we live elsewhere and, and, or the role that we have. So one of the most wonderful projects my lab took on this past year was studying the Youth Climate Action Fund that Bloomberg Philanthropies initiated about two years ago, taking an interest in how young people might serve as a relevant constituency to their city's climate action plans. So it was a youth micro granting project. Cities would get a certain amount of funding and then they would regrant that to young people to do projects that were aligned with the city's climate action plan. But what was fascinating about interviewing those young people was that once they had a sense that the city leaders were actually interested in their point of view, they started sharing more information. And we came to see that young people as a constituency, like moms in a community or like teachers or like small business owners, that they have a perspective and a point of view and expertise that if you layer that and if you integrate it, you get a very different portrait of where a challenge exists or where there's a burgeoning solution that just hasn't been noticed yet. I think AI can multiply that. I think it can also help search out solutions that exist in comparable cities working on similar challenges, and maybe even identify sister and brother cities that you may not have recognized as having useful information and models that you might be able to benefit from. That's a little bit different than having a solutions bank, which I think is really important. It's also about trying to customize a little bit more. A mid sized city or with a mayor who believes this, who was that in his or her previous work life, and a community that had a natural disaster and recovered from it and, or fill in the blank. I think it allows for that layering and that customization. And I think that's one of the exciting things about how we might be able to use AI models as we go forward.
- [17:53]
B
Well, your work's so interesting and there's so much opportunity to help. We have state and local officials who listen to this podcast if they want to deliver good services. What are one or two pieces of advice? How would they measure whether they're delivering good services and what could they do that would have the biggest impact.
- [18:12]
C
Two thoughts. The first is one that we know but could probably do a more robust job of bringing to life and that's organizing your efforts around the resident's journey, not the city's org chart or the city departments, which is how most of us think. Right? We're thinking through a departmental lens, we're thinking through the org charts, we're thinking through particular budgets, which are clearly important. But we're not always thinking about the resident who woke up to get their kids ready to go to school and didn't say to himself or herself, I'm going to go and use the transportation sector to get my kids to daycare. I'm going to go to the department of education that's running or plays some role in their schooling. Instead. People are thinking about how they want to get through morning to evening. They want to know that they're safe and they want to know what they need to do next and what paperwork they need to bring, what kind of attestations will be necessary. And a lot of the time we're not thinking about just that time tax that's involved in all of the work of getting through day to day life. So a second piece of advice would be to start efforts around a high stakes service, one that really makes a difference to a lot of people. In the federal context, it's often called a high impact service provider, but a service where people need it, people use it and for that reason they're going to be willing to partner with the city in order to make it better. The last thing, I guess a third piece of advice is to rigorously measure to measure at the beginning so you have some sense of what a resident's irritation actually looks like in a measurement scheme and in the middle and then at the end so you have some sense of what's improving. And when I think when you do that combined with all the other things we've talked about of really including residents of mayors and city leaders who are committed to listening and to listening to be surprised, I think that's a winning combination, that set of factors.
- [20:29]
B
Well, Dr. Kim Leary, you were spectacular. As I anticipated a lot of good advice for local officials. Good luck with good Services Lab. We'll be eager to promote your terrific results. Thanks for your time.
- [20:42]
C
Thank you so much, Steve. And thanks for the terrific work that you do with this podcast and with your own work on service improvement in cities.
- [20:52]
B
You're welcome. Back to you.
- [20:57]
A
Thanks for listening. Do you have a question for our Mailbag episode? Send us an email @datasmartks harvard.edu. this episode was produced by me, Betsy Gardner, and there are more great conversations on our other episodes. Just search for the DataSmart City pod on your podcast app and don't forget to subscribe so you never miss an episode.