Get in touch

Postlight joins Launch by NTT DATA! Learn more.

There’s a feeling in the digital industry that working with the government is tedious due to endless bureaucracy, red tape, and too many stakeholders. As someone who worked in the public sector for years, Postlight’s Associate Director of Digital Strategy, Tait Foster, understands it better than most. This week Chris and Tait discuss the differences between working with public and private sectors and share tips on navigating software development for government stakeholders and, most importantly, their constituents. 



Tait Foster: Every time you want to try and do something new or try and expand something, it requires the corporate equivalent of the peaks of Westphalia. 

Chris LoSacco: (Laughs)

Tait: Like, you gotta get everybody involved…

Chris: (Laughs) Right.


Chris: Hello, and welcome to the Postlight podcast. I am Chris LoSacco, the president of Postlight. And today, I have a very special guest on the show: associate director of digital strategy here at Poslight, Tate Foster. Tate, welcome.

Tait: Happy to be here.

Chris: Good. I’m very excited to talk to you, because you are an interesting combination of strategist, product thinker and experienced government operative. And I want you to talk about that a little bit, because what I want the theme of today’s show to be is building software in a public sector environment, in a government environment, and why it doesn’t have to be hard. So first, let’s start… just tell us a little bit about your background, what you’ve been doing, both in the digital realm and in the public service realm.

Tait: Yeah, absolutely. So, yeah. So, I actually started my career in politics. And then from there, I basically worked in every facet of it. I worked in local politics in lower Brooklyn, in Queens in New York City, I worked in federal politics on Capitol Hill, I worked on campaign politics and state politics out in Washington State, and then worked for a nonprofit that did military and foreign policy, Track II diplomacy, back here in New York. And that was mostly with Russia, China, the United States. And again, Track II diplomacy, a little bit of military policy, a little at-PAC work and a little bit of climate work. And it was at that point where I sort of had tried every facet of politics and decided to try something new, and that’s sort of how my tech career started. I went to Bloomberg, I was on the editorial side for a hot minute, and then basically was given an opportunity to move into the product side, and… yeah. Worked at Bloomberg in the product side, New York Magazine. And then I actually went back into, not public sector but into politics. Into campaign tech. Did that for two and a half years, and I think that, you know, that’s why a conversation like this is so important, because I think there’s all these different distinctions that are necessary to be made between things that often are taken as interchangeable. So, politics versus government, and how a lot of the time I think it’s very easy for folks to think that those go hand-in-hand. And that… this might be getting a little ahead of ourselves in terms of our conversation, but I think a lot of it is just… there are extra considerations, because of how politics can inform government…

Chris: Oh, of course.

Tait: But that doesn’t mean that all tech in government has to be informed by politics.

Chris: That’s extremely well-said. I do think we should scope our conversation a bit. I… you know, my hope is that we can talk more about working in an established government environment, i.e. less about running a campaign and more about, once you’re in the building there are so many different government agencies, organizations, divisions, levels, right? ‘Cause you’ve got federal, state, local, you’ve got just affiliated non-government organizations who are trying to do work for the public sector. So… let’s talk about why it’s hard. ‘Cause I think there’s… it’s not a far reach to say there’s this feeling in the digital world, and the digital industry, that working with government is hard. Working with government, you’re never going to get anything done, you know, you have to go through, you know, fourteen layers of red tape, you can’t ship any software. And I would posit that it doesn’t have to be that way. But in your opinion, what are the challenges? Like, what makes it difficult, historically, to get things done?

Tait: Yeah, I think that’s 100% right. I think there is that sort of initial belief, understanding that there just… it’s bureaucratic, that you have to get through all this red tape. And I think that to a certain extent there are times where that’s true, I think, and also sometimes I think it’s just inherently confusing. Even, whether you’re talking about a city as large as New York, or you’re dealing with a township, you know, there’s always going to be that question of local versus state versus federal.

Chris: Mhm.

Tait: And I think a lot of the time, what is seen, or can be seen as bureaucratic red tape is actually just confusion, of who has the main stay here? Is it the New York City Board of Health? Is it the New York State Board of Health? Is it the Department of Health and Service… and, you know, that causes a lot of confusion and… you know, there are both interpersonal relationships of those entities, but I think then there’s also the inherent tension and rivalry that can exist with those.

Chris: Yeah. Totally.

Tait: But I think, you know, I think that’s one. I think, fundamentally, I think that’s one piece of it fundamentally on the government side. There’s also just the fact that it’s not… it’s often been seen… and I think this is both fair in some cases and deeply unfair in others… that government has to move slowly. And people put it as slowly, you know, “Oh, they have to move slowly,” or “Oh, this could be done faster in the private sector.” But I think a lot of that is tied more to the fact that in government, you have to be more considerate. You know, there’s a level of care that has to be put into things.

Chris: Yes.

Tait: You know, it’s one thing if you have a bright, shiny new startup that’s trying to disrupt, you know, health insurance and there are doing things a little bit fast and loose. It’s one thing for, you know, someone trying to find a niche in a market. It’s another thing when it’s the New York State Health Department.

Chris: Right.

Tait: You know, there’s different levels… and that might just even be a scale question. Beyond, sort of, a legal question. So I think that’s another one of just sort of, you know, the impact that those kind of… and you see that with the launch of, the initial launch of the Obamacare website, and how that led to such confusion and distrust because it wasn’t taking into account, you know, how people would actually use it, and thinking about it more as a consumer product, rather than a government website.

Chris: I want to dig in on that. Because I think that is a very instructive case study, right? So if we think about the rollout of, and the fact that you could buy health insurance policies directly from a federally sponsored or a state-sponsored entity, if you qualified. There’s a political question, which we’re not gonna touch right now, and then there is a technology question. And it is fascinating to me to look back at that, and to think the same kinds of problems that we see in big client environments, big enterprise environments when we go in, existed… maybe even, you know, multiplied during the first approach to, where they said “We need this to be a big team, we need to invest in the infrastructure and make sure that we are building this giant thing, we need to go with a trusted entity,” and it got caught up in this kind of snowball of having to check all the boxes and be developed against a unchanging set of requirements, instead of being developed around the users. And what ended up is, you had something that is incredibly unreliable, incredibly hard to use, it was crashing all the time and it was incredibly expensive to run. ‘Cause you had this big team, and you had this expensive infrastructure, because it was built, again, looking at, you know, a long requirements document instead of looking at an experience. And the way that it was fixed, and there’s a very… I mean, there’s several, actually, sort of looks into what happened, where they assembled a team of ex-Silicon Valley people to rebuild the thing from scratch. And they took a completely different approach. They said “We’re going to build a minimum viable product. We’re not going to have all the features out of the gate. We’re gonna build it on a cloud infrastructure.” They used Amazon web services. They focused very much on user experience, and making sure that it was really easy to sign up. They prioritized reliability and the whole team owned the quality of the software, versus throwing it over the wall to a, you know, an outsourced QA team. And the difference was night and day. They launched it in a fraction of the time for a fraction of the cost, and it was this very stark before and after example, where if you bring a heavyweight process to something that really doesn’t need it, you’re gonna end up in a way worse position than if you bring a lean team to go attack a new problem, prove something out, and then shore it up as the platform lives on. And I think that that exact approach could be applied to so many other areas in the public sector that it just hasn’t… that it just hasn’t been done yet.

Tait: Yeah, I think that’s 100% right, and I think a lot of that stems from… you know, we see this… you know, as you said earlier, with sort of large enterprise clients that are in the private sector, where it’s just you have… they’re at such a massive scale, they have so many different sort of internal business units or teams, and you end up with these RFPs and these discussions that have way too many people involved, as opposed to sort of… you know, I think the very basic question of… and these are product questions, of “Who is it for? What’s the path that they want to take? Why do we want them to use it? Why do they want to… why do we want them to want to…” You know, these sort of very simple questions. And I think that that was a lesson for government, and some people have learned it. I mean, you see the founding of digital offices. The New York City Mayor’s Office has one, the White House has one that carried from Obama through Trump.

Chris: Yes. Yep.

Tait: It’s in the Biden White House now. You are starting to see, cropping up, these entities within government that are trying to bring some of those learnings, and you know, the importance of a MVP, the importance of a laser-focused team. The managing stakeholders in a way that is logical for what you’re trying to do, not to make sure that someone doesn’t get angry at the end of the road.

Chris: I want to pause there again, because I think that’s worth a little time too. The stakeholder management with a government entity is a little bit different, right? And it’s probably more important… I mean, it’s important in any scenario when you’re building software. You’ve gotta get the people who are sponsoring, so to speak, your effort, to feel really good about what you’re doing, and be an advocate on their behalf. But when you’re working with… when you’re working with a public sector entity, you’ve gotta think about not only who are the decisionmakers, but what consensus do they need to build? Who do they need to have onside? And how do you help them get people onside? And that’s a complicated thing, right? I mean, one of the things that we always do is, we say “Let’s default to working software.” So rather than showing people Powerpoints and requirements documents, let’s prototype something, let’s have a design, a clickable prototype in Figma. Or maybe it is an alpha version that only has 20% of the features, but shows you how the thing is starting to come together, and communicate about that early and often, so that people can rally around the key stakeholder that is sponsoring the effort because they see progress being made. Versus sitting in a meeting… you know, how many meetings have we sat in where it’s like, 60 people, and you’re running through the Powerpoint slides, and it’s like “Okay, we’re on track, it’s month two of the nine month discover phase,” and you’re like “Oh my God. What… when is this ever gonna make a difference to citizens or users,” you know? So showing software is key to stakeholder management and building broad support within these groups. And I think, you know, a point you were making to me earlier is it goes maybe even one step beyond that, because you have different motivations even amongst the stakeholder groups, right?

Tait: I think we overstate the differences, a lot of the time. But this is one of those areas where I think government does have certain wrinkles that don’t exist immediately in the private sector. You know, an example in the private sector… you might assume that that VP, or director, or head of… you know, or C-suite individual is trying to get to the next tier.

Chris: Climb the corporate ladder.

Tait: Right. Or they’re trying to set themselves up. A lot of the time in government, you have people, it being reported that they’re trying to do that. (Laughs) The way that I always looked at it is that you sort of have… you have electeds, you have appointees, and then you have your sort of career governmental workers.

Chris: Mhm.

Tait: And, you know, what’s tricky about that is, the electeds… not gonna delve too much into the politics side, but they’re always kind of looking, like “What’s the…” You know, “On the federal side, what’s the next committee assignment? Is there a larger… am I gonna go back and run in the state side? Am I gonna…” You know, and that trickles down. I’m a mayor here, do I run for the legislature there? There’s always sort of that consideration.

Chris: But are you saying, Tait, that a software team has to kind of play that chess game on their behalf? Because I think it might be simpler than that. 

Tait: (Laughs) Thank you for keeping me in line. I think the key is that it gets very easy to get plugged into that in that world.

Chris: How so?

Tait: Well, just that you’re hearing this, you know, I think right now… I think there’s one person who’s declared for the presidency in 2024, but there’s already eight other people’s names floating around?

Chris: Swirling around. Yeah.

Tait: Yeah, and if you’re… you know, on the Republican side if you’re in the Florida, or Virginia, or any of those places, you know, you’re hearing that name pop up. What’s interesting about that is that, depending on where you are in the governmental hierarchy, you might be talking about your boss’s boss’s boss. You know, and that, if you’re at another publicly traded company you’re probably not hearing the same type of rumor mill, or just impacts you differently than the way that it’s being reported on the governmental side. So I think a lot of the time that can feel very imposing, but I think, to what we’ve been discussing before, I think it… understanding the stakeholders and what they’re trying to do, I think, can be much simpler. Because fundamentally, as much focus as there is on elected or principles, the people that matter are the constituents.

Chris: Yeah.

Tait: Are the citizens of the state, of the municipality, wherever you’re working. And I think that, you know, that’s where… we’ve seen a little bit of it, but you see more of it, of sort of saying “Well, what’s the answer? What can we provide them?” And this is one of those areas where I think fundamentally, government does an uneven job of sort of saying or displaying to constituencies what they can do for them. Like, I remember this when I worked in congress. You know, my portfolio when I worked in congress in local politics was mortgage modifications, immigration casework, healthcare issues. And people would call us and they’d be like “I need help with a mortgage modification.” Now, I wasn’t a banker, I didn’t work at it, but I was basically an arbiter, in a way. Of like, I was someone in their corner who could send the paperwork over to the bank to basically say, like, “Look, the congressperson is paying attention to this. We’re involved.” We couldn’t sway the bank to do anything, but it was a way… you know, not dissimilar from immigration casework. It was just, we were a point of contact with these huge entities for our constituents. And I would explain and tell that to people, and they were like “Wait, you…” you know, “You did what? Your portfolio is what?” ‘Cause people don’t, you know… it’s very important stuff like that, and it’s, you know, you also can get like a free tour of the Capitol if you call your congressperson. I think that a lot of the time, government, because of that feeling that it’s so big and it’s so hard and everything has to be worked out in RFP, loses sight of the like… the simplistic. I mean, going back to the, of like, what can we do to simplify what is a complex system without… And you can simplify something without breaking the law or breaking the rules or anything like that.

Chris: No, of course. Yeah.

Tait: You can comply while also simplifying.

Chris: I wanna tease out something in what you’re saying. Hopefully people are listening to this and they can take away some concrete advice, right? And we’ve given… you know, I think MVPs are important, right? You can build the simpler version, like you were just saying, and that can work. You can embrace modern technology platforms. Amazon web services is government certified, so that is a clear example, and there are a lot of examples where you can use open-source software and modern tech to accelerate what you’re doing. You don’t have to keep building on the COBOL platform that was there from 30 years ago because that was there. But the… this idea that stakeholder management is a little more complicated… I think you were sort of touching on a point that I think is really good, which is that the ambitions are a little bit different in public sector, because it is not about maximizing revenue. Right?

Tait: Exactly.

Chris: The goal is not “How do I make the most money?” The goal is, “How do I serve my constituents to materially improve their day-to-day lives, and, if I’m thinking a little selfishly, how do I show what I’ve been able to do?” Right? Whether or not I’m trying to build my political career, whether or not I have aspirations…

Tait: Right.

Chris: A lot of times the “success,” quote-unquote, in government, is what I was really able to achieve. And that, I think that’s like a… it’s a back door. It’s a Trojan horse into, how do you align stakeholders? Because as a digital team, right, as a software team, we can say “We’re gonna work quickly to put working software in the hands of people, and that is going to be something you, stakeholder group, can point to and say ‘We got something done.’ Right? We put a working platform in front of the people of X, you know?” And I think that can be a very powerful statement, versus “We pumped this budget into some IT line that’s a row on a spreadsheet without really having much to show for it.”

Tait: It also is a way of thinking that kind of nullifies that political question or that political uncertainty that can be incumbent in government, which is, you’re trying to provide services that already exist. You know, you’re not worried about “Alright, we’re trying to introduce this whole new concept.” You’re like “No, these exist.”

Chris: For the most part, yeah.

Tait: For the most part, it’s existed… you know, yeah. Exactly. For the most part. That’s an interesting quick point, is that either you’re already trying to provide the services that already exist, of like… you know, we saw this recently with COVID, in terms of contact tracing, trying to get that off the ground.

Chris: That’s what I was thinking of.

Tait: You know, it was like, trying… you know, that in theory, that’s a new thing, where it’s like “Alright, how can we react to a crisis quickly with the data that’s incumbent, and basically what we need to do to keep our citizens healthy and safe.’’

Chris: Yes.

Tait: There’s the other elements of it, which are just, you know, like “Can we make the DMV more efficient? Are there elements there?” Like, I recently officiated a wedding in New York City. And it was fascinating to me, you know, I had to fill out… had to print out a form, fill it out, I had to send them a money order.

Chris: Hm.

Tait: It couldn’t be a check. And it’s these sort of things where I’m sitting there being like…

Chris: (Laughs) 

Tait: You know, I had the certification that I’d gotten from the internet. Like, this should be 15 minutes.

Chris: Right.

Tait: Here’s my credit card, here’s the religious title that I’ve chosen, here’s your $15 and we’re off to the races.

Chris: We’re good to go. Right.

Tait: But instead it was, go to the Kinko’s store to print out the form…

Chris: (Laughs) 

Tait: Go to the Western Union to get a money order,” which I’d never done in my life…

Chris: Yes.

Tait: Go back to the Kinko’s store to mail it, and then the entire time sitting there being like, “Alright, this wedding’s in two months, I hope they got it.”

Chris: (Laughs) Totally.

Tait: You know, there are elements like that, where that service exists. And it’s a silly one, but it’s like about simplifying that. And then I think you get to beyond that, and this goes into sort of making it easier for the workers and the folks who are working at the DMV, and the ways that you can streamline this and make their lives easier. And thus also make lives easier and happier for your constituency. Then I think you get to this really interesting area, and it’s the point that I, like, have these sort of fever dreams about, which is… how can you start having these different governmental entities and agencies, whether they’re local, state or federal, start interacting with each other in a way… in front of a constituent or in front of a citizen or a user, that actually makes sense and is helpful.

Chris: Mhm.

Tait: An example that comes to mind is, let’s say I’m going and visiting family in Idaho, and I’m going to go and get a fishing or a hunting license set up. Wouldn’t it make sense that Idaho Department of Parks, that these other… the Department of Tourism, have these things that are like “Oh, okay! You’re looking to go fishing in Idaho.” Doesn’t it make sense that it’s like, “Here are some great places to go fishing”? Here are some other attractions in our state.

Chris: Oh, I see what you’re drawing here.

Tait: Doesn’t it make sense for the federal government to say, like, “These are some national parks… you can’t fish in them, but maybe you’d have an interest… You know, preparing to do an outdoor activity, maybe you’d be interested in them.” There are these interactions that I think… you know, and this is a little bit of a self-reflection that I think government has to do, but what are the ways that constituencies don’t expect to be served that could be served. Beyond the, like, making things easier, which, like, look. Even just the table stakes in making things easier would be huge.

Chris: So, to me, my honest reaction is, it feels like you’re like four steps ahead. (Laughs) Because I actually think that just focusing on something like you were talking about before, which is pure, what I will call efficiency gain, right? There is massive opportunity there. And part of the problem is, the incentives are not always aligned. Meaning, you’ve got an apparatus that is set up, and people learn how the machine works…

Tait: Mhm.

Chris: And in some cases I mean that literally, like, you know, they learn the interface that’s in front of them, and it is very difficult to start to change how it works. From a personnel perspective, from a training perspective. Like, these are the things that you have to really take care with, when you are thinking about introducing a revolution. I mean that in a lowercase-r kind of way, right? 

Tait: (Laughs) Yeah.

Chris: Like, a brand-new way of working, versus just an incremental improvement over what’s there. And I think that it seems off the table because it’s too hard, but I think what we are here to say is, that’s not off the table. You just have to be considerate of it, right? You have to understand that part of the job is not just whipping up a new interface using the latest and greatest tooling. Part of the job is sitting with the users, understanding what they do today, and figuring out a… I don’t know. A transition plan, or a, you know, a training period, or an overlap, where you can get people comfortable with the fact that it’s going to take them fifteen minutes to do something that took them three hours before. And my limited experience is that sometimes, vendor… this is where vendors fall on their face.

Tait: Mm.

Chris: Because they think, “Oh, I can just… as long as I’m checking all the boxes for this new thing, we’ll be good to go.” And they don’t think about, this actually has to get in front of real people. And you’ve got government workers who have… and they’re incredible civil servants, right?

Tait: Mhm.

Chris: They’ve spent 20, 25, 30 years, maybe more, doing a particular job to serve their co-citizens, you know? And what I think we try to do, and the best digital agencies try to do, is they really internalize that. And they say “User education and user adoption is part of this journey,” right? And if we’re thinking in a product-oriented and in a platform-oriented way, we’ve gotta be thinking about how we get the users to use this new thing, and not just foisting it on people. Do you know what I mean?

Tait: No, I think that’s 100%… I mean, it goes to, I think a fundamental question that sort of wildly isn’t asked enough, which is “Why?”

Chris: Exactly.

Tait: Why are we doing this? Why does this exist? I think this is actually something that government workers are actually quite good at, is asking the why. You know, and this is… look, I think this is partially why this is a fruitful conversation, as we started it. There is this kind of preconceived notion that government is hard, and that it takes too long, and I think a fair amount of that stems from… there have been a lot of different instances where, you know, someone comes in and says “Oh, this is… I’ve got the solution,” but that solution isn’t taking into account the experience of the government workers, isn’t taking into account the constituencies that they’re trying to work for or what they’re trying to accomplish. And so it ends up kind of being a boondoggle, as opposed to… you know, and this is what we were talking about earlier in terms of MVPs, the importance of asking these…

Chris: Start small.

Tait: Start small, but also the importance of iteration. It’s funny, because it’s sort of inherent in government, this idea of pilot programs. That’s a well-known concept.

Chris: Great point. Yes.

Tait: And it’s wild, because it doesn’t seem to be applied to this kind of work a lot.

Chris: Right. It’s… that needs to translate to digital, you’re exactly right!

Tait: And the funny thing with pilot programs is, more often than not the way that they’re discussed is, “We’re gonna do a pilot program.” And it either doesn’t work and it can be used as a political football, which… Or, it does work, and it’s like “Great. How do we do this everywhere tomorrow, or yesterday. And I think this goes to something that constantly we’re talking about from what we see, in terms of like… again, starting-small iteration, but then like doing those things with a consideration – not a focus, but a consideration – to scalability.

Chris: 100%.

Tait: And I think that’s where, you know, I think you can… in the past, and currently, sort of run into these sort of problems where, if you are doing something small, there hasn’t been a thought to…

Chris: Where it’s going.

Tait: Yeah! How does this… like, you know, one of the things that I always have said, this is a sort of maybe trite thing from product, of like, you know, you start with a bicycle, goes to a scooter, goes to a motorcycle, goes to a car. And that’s how you… that’s one way of thinking about product development. And what I’ve always sort of said, whenever I’m talking to product folks, the way I’ve always approached it, was I’d start with a scooter. I have three or four different ideas of what a bike… what the motorcycle or car thing could look like. They might not be where we go or end up.

Chris: Right.

Tait: But having that level of awareness was enough to sort of give me a little bit of peace of mind, of “I’m not just putting all of my eggs into one basket here. I have a rough sense of the spectrum of where this could go.” And I think that that’s something… like, what I was sort of saying earlier, of wouldn’t it be great if you could get a hunting license and it was keyed into the department of tourism. You know, those are ideas and thoughts that people in government think about and are dealing with, because more often than not they know the other people in their departments. Like, they’re aware of each other. The funny thing about governance as a whole, and governments as a whole, rather, is that more often than not the sinews of it are the people that we don’t talk about. Congress is run, sure, by the speaker and by the… you know, the congressional folks and all that. But it’s congressional staff. You know, folks who have been there, you know… same thing with state agencies and departments and all that. And I think that a lot of the time, there are those inherent connections and all we’re sort of thinking about and talking about is, you know, taking advantage of those and then sort of trying to make sure that they’re not reliant on a single person or a single interpersonal relationship or all that. And thinking through, how can you take some of that knowledge, some of that awareness, and turning it into something that a constituent… you know, you’re not just lucky that you ended up talking to the right congressional staff member, that you’re not talking to the right person at the Department of Health…

Chris: Well, the software should help you.

Tait: Right.

Chris: A digital platform should help you navigate, you know, so that if you don’t have these personal connections you can land in the right place. I feel like this is… this is a good place to close it. You know, there’s so much more that could be talked about with this stuff, but we’ve touched on a few of the key things, right? We embrace a modern way of working, start small, ship a minimum viable product and iterate, use the cloud, use AWS, customer experience matters, and that’s whether you’re talking about the citizens or whether you’re talking about government workers who are trying to do their best. And they, you know, in my experience largely are really wonderful people. 

Tait: Mhm.

Chris: Just doing really critical work for all of us. And above all, you don’t have to go slow. You don’t have to internalize the bureaucracy around you. You can work within it and figure out ways to align your project team’s motivations with the motivations of your stakeholders. So, we hope this was helpful. Tait, thank you for coming on, talking about this. We love public sector work, and if you’re listening to this and you’re like “I’ve got a project like this and I would love to talk with these people about how we could bring some of this thinking to my group,” please reach out. Me, Tait and the team, we love getting those kinds of emails, and we would love to talk to you and hear about your situation and figure out how we could maybe make it better. Thank you, thank you for listening, and we will talk to you all soon. Bye.