In this episode, hosts Brianna Lennon and Eric Fey speak with Jared Marcotte. He’s the founder of The Turnout, a company that focuses on the intersection of technology, software and security in elections.
They spoke about what motivated Marcotte to create The Turnout, as well as about the importance of election administrators being able to determine the safety and security of the non-voting system technology they use.
This episode is part of a three-part series exploring election technology.
High Turnout, Wide Margins Credits:
Managing Editor: Rebecca Smith
Managing Producer: Aaron Hay
Associate Producer: Katie Quinn
Digital Producer: Mark Johnson
Transcription of the episode is as follows:
Jared Marcotte: One of the things is really just being able to listen to the folks that you're trying to to work with, right? It's great if we take your data, we go off and we do all of these wonderful visualizations for it, but if none of that helps you from an operational standpoint, like, what's the point?
[High Turnout Wide Margins Introduction]
Brianna Lennon: Welcome back to another exciting episode of High Turnout Wide Margins. My name is Brianna Lennon. I'm the County Clerk in Boone County, Missouri, and with me is my co-host —
Eric Fey: Eric Fey, Director of Elections in St. Louis County, Missouri.
Brianna Lennon: And today we're really excited to talk to —
Jared Marcotte: Jared Marcotte. I'm the president of The Turnout, which is a company I started about, well, a little over a decade ago now, and we do a lot of very nerdy stuff for state, local, and federal clients.
Brianna Lennon: We are nerdy, as well.
Eric Fey: This is the right podcast for nerdiness.
Brianna Lennon: Our first question that we always ask people is how you got started in elections, and so, take us back to even before The Turnout, what you were doing that brought you into the election space and your path through it.
Jared Marcotte: I feel like everybody always says you end up here by accident, and so I was working at a startup in New York City, and coming from Vermont, which is where I was born and raised — me and New York City did not get along all that well. It was a little bit of a culture shock. By a little bit, I mean a lot, and, you know, anxiety runs high. And I was having some struggles working there, and my sister-in-law sent me a job application for the Voting Information Project as a software engineer, and she's like, "You should apply for this job because you hate yours,” and that's essentially how everything got started from there.
Eric Fey: Remind the folks that haven't been around as long what was the Voting Information project?
Jared Marcotte: So, the Voting Information Project, at the time that I was there, was a project in which the Pew Charitable Trusts and Google got together to try to answer how do we help people find out where they vote, what's on my ballot, and how do I navigate the elections process? And so, it was working with lots of state and local election officials to provide information to places where people go online to look for information, like search engines, social media sites, etc. so they can get official information directly from the source, but finding it in the places that they typically go. And so, I did that project for a few years, and then I was asked by Pew to come over and then run it on their side, and then I left to start The Turnout after that.
Brianna Lennon: The Turnout is a lot more — it's less public focused, right? It's more like an industry organization. So, can you talk a little bit about some of the projects you work on?
Jared Marcotte: Yeah, sure. One of the major projects we worked on is the Overseas Voting Initiative [OVI], which you know quite well, and that is a project between the Federal Voting Assistance Program, basically the the voting arm of the DoD [Department of Defense], and then the Council of State Governments, which is a membership organization that represents all three branches of state government. They get together to try to figure out how to help military and overseas voters because they're one of the most difficult voting constituencies to reach for lots of geographic reasons and infrastructural reasons, so we work[ed] on technology project with them specifically overseeing like the tech policy, like “What systems could you put in place to help election officials better serve these folks? How can you create more collaboration among the federal entities that oversee this work in order to again help military and overseas voters?”
So, one of the biggest things that we do is looking at the data and trying to figure out where are those places that we can ask more questions to say — “Is this an issue for a military and overseas voter? Is it the communication between international mailing systems and our own? Is the issue that people are not getting the information on time? How can we look at this data and really get better insights into what that, you know, progress to goal really is?”
Eric Fey: is. You've talked a little bit about what The Turnout does — we want to hear more about that, but I am perhaps most interested in how did it come about? Can you talk a little bit about the origin story of The Turnout?
Jared Marcotte: Yeah, I think working on the Voting Information Project gave me a better set of insights as to — we always talk about the trenches, right? And everyone has a different view of what the trenches really are in elections. And I always look at the data trenches and what's missing, and I think there's a lot of low-level infrastructural work like data standards, and, you know, ensuring that the information that gets shared with different systems is actually well constructed and constrained in a way that you don't lose any information, right? You talk about signal loss in certain situations, and so you want to make sure that you're getting the strongest signal from one system to another. And so, things like looking at this data that's coming from the military and overseas voters from the local administrative systems — we want to make sure that we're actually comparing apples to apples in a lot of things, right? I think when you try to interpret the data underneath, and then provide a survey response to something — there's something lost in translation. However, if you're looking at the low-level administrative data, and you've got a strong set of definitions as to really what you're looking for, and that you know that it's fairly hardened to collect those things, I think you can get better insights into what's actually happening at a local and state level, so it's not just, I think, the election space where all of those those trenches could be dug, but it's been basically where we've been working for a while, for quite a few years.
Brianna Lennon: I'd love to hear your thoughts on maybe some of the differences that you've noticed because the data that we're talking about is in the public sector, and we, as election officials, don't have the capacity or even the expertise, or people around us with the expertise to be able to really analyze data in any particular way. Versus the private sector, where, as you mentioned, AI is being leveraged for all kinds of things because they're profiting off of it, and so they're going to do more data analysis because it's going to make them more money. And we don't — that's not why we would be doing data analysis. Has that been a challenge? Has that been something that you've had to kind of deal with since it's not really a business thing?
Jared Marcotte: Yeah, I think one of the things is really just being able to listen to the folks that you're trying to to work with, right? It's great if we take your data, we go off and we do all of these wonderful visualizations for it, but if none of that helps you from an operational standpoint, like, what's the point? So, if we're not talking to you and saying, you know, “Where are your pain points?” Like, “What is the issue that you're trying to solve here? What are the questions that you actually want answers to?” You know, if we come to you and say we have a number of insights on your military and overseas voters, we could show you 10 charts, and you only like one of them. It's like, “Okay, well, let's talk about this one, and like, how does this help?” Because we do want to reduce the noise to signal ratio to a point where it's it's understandable and it's approachable, and I think that is really the most important part is, is making sure that what we're doing is impactful for the folks that we're really, we're talking to, and you know, yeah, data data analysis is great, but it's got to be data analysis that's operational in in some way.
Brianna Lennon: Because our audience is mostly election officials, if somebody were to listen to this episode and want to learn more about their own offices or about processes they're doing and the data that comes out of it, what should they be doing? How should they be thinking about data analysis?
Jared Marcotte: Yeah, I think a lot of it comes down to process in general, right? We do a lot of work on process analysis, as well as data analysis. I mean, it all, right — in my head, it's all data in the end, right? We're looking at how you do things, right? And process analysis and process improvement is a big part of it. I mean, I think that looking at what you do right now and starting from a “What does the ground truth look like” first, and then talking about “Okay, well, how do we improve from here?” There's really a whole discussion about “What do you have to do, right? What's your statutory environment say? What are you doing right now that may be outside of your statutory environment that you still need for operational reasons?” And then, “What are the things that you can effectively reduce to make sure that you're doing only really what you need to do,” and you're trying to save time, you're trying to save the effort, you're trying to save money, which is a big thing. And honestly, these are some of the most impactful things that you can really do because it doesn't involve buying a new tool, it doesn't involve like doing a huge RFP [Requests for Proposal] or changing your system, you know, paying for vendors to do those sorts of things, and you can get actionable insights from that information, but you can also change like how your office functions in a a fairly accessible way that is responsive to your own needs.
Eric Fey: So, you mentioned OVI earlier. What's another one or two things you guys at The Turnout are working on?
Jared Marcotte: So, one of the other major things that we've done is a, it's basically a project called RABET-V [Rapid Architecture-Based Election Technology Verification]. I'm not going to explain what RABET-V actually means. We've stopped exploding that term out because it just is very long and ridiculous, but, in short, it's a way to look at non-voting election systems in a way that allows you to better understand what you're purchasing, right? It's a way to test the security maturity of that system, and figure out that, “Oh, this organization has good controls in place to create good safe software. It looks like the product itself is built well,” and then doing a penetration test and a compliance audit will let you know, “Is the system I'm actually about to buy going to live in my environment and be safe for me to use.” So, we've run the system, we've run the program, we developed it with the Center for Internet Security back in 2019-2020. We ran it through two pilots, and it became operational in 2023, and we've been running it since then.
Brianna Lennon: So, the thing that I appreciate about RABET-V is, I think that it identifies something that I have concerns about, which is that there is, I guess, so many options for different software for things, like you said, that is not technically elections related — it's not the tabulators. It's not the things that go through the EAC's process for verification, but they still touch our systems and they do still introduce weaknesses if they're not super secure. So, things like poll pads, but also things like election night reporting foundations and things. And there are a number of organizations that are little mom and pop that have gotten to know a county because they used to work with the county in some other capacity, and we see that, not just in elections, but in other county offices where one office will buy a piece of software and it will not interact with any other piece of software that is necessary for the functionality of of something to happen, and then it causes all these problems, and that's a common data format thing, but it's also security risk. So, what motivated the creation of RABET-V for you, and what's been the reception from election officials?
Jared Marcotte: Yeah, so at the time, in 2019, there was no program to certify non-voting election systems. The EAC, obviously, has a program for voting systems specifically, but in 2019 there wasn't anything else, and that was one of the biggest concerns that CIS had was these systems are largely, as you said, sometimes you're not really sure what the development process looks like. They're developed by small folks who may not be or small shops that may not be effectively at a security maturity level that you would want them to be, and they're usually internet connected — either directly or indirectly connected to the internet, and that, of course, changes the game when it comes to the threat environment that they have to live in. So, we developed the program for that purpose, and we wanted to make something that would be effectively universal from a security standpoint. That you could be RABET-V certified in one state, and that if you're looking for that same certification in another state, then you don't have to go through the process twice.
We also wanted to make sure that the certification process actually mapped to the development process because people can develop things very, very fast — both good and bad. But we didn't want certification to be the thing that held them up before they could get back into market, right? So, you wouldn't have to go through a year or more of certification before, then you could release your next version of the software, which then would have to go through its own set of certification controls, right? So, we wanted to map this so it was just as rapid as you could develop.
I think the reception from election officials has been good. I think communicating what the program does and how it helps is one of the most difficult parts. I mean, again, we're nerds. We're not, you know, communications experts, and trying to talk about all of these things, about security controls — it's a large educational piece that we're still working through. That doesn't mean that we haven't had some major successes in some states, and that people haven't really appreciated this kind of rigor and speed. But, you know, we're still climbing. We're still going.
Eric Fey: I'm glad you addressed the issue of speed of this process because I know one criticism that has been leveled against the VVSG [Voluntary Voting System Guidelines] and all that, is the kind of glacial pace of reform or you know updates to VVSG, and that it can have the effect of stifling innovation in the election technology space, so how do you balance that? I mean, do you worry that this RABET-V process could, in some way, stifle innovation or stifle responses to, you know, things that pop up, or or is it a good thing, in that regard, to make sure you know if you're responding to some issue that's popped up somewhere, is it good that there's a system for it to go through?
Jared Marcotte: It's a balance, right? Trying to weigh its effect on the market, and we all know that it's a little bit of a fragile market, so adding too much burden to the folks that are out there is going to be difficult. But we also have to make sure that there is a baseline for security and that we know what the threats are out there, which is largely why we took a maturity model approach to RABET-V. So, there are three levels of security controls that you have to make sure you go through. There's level one, which you have to get 100% of. That's just the pure baseline. If you don't, then you know maybe you shouldn't have a product that is out there and potentially dangerous to the environment that you're actually operating in. But then we also look at the level two and level three for basically, like the more advanced controls that we hope that you have in place, and that you have to meet some mixture of 50% of those, because eventually, as things change and things change rapidly, those level threes are going to become level twos, right? And those level twos are going to become level ones, and then there's going to be a whole new set of level threes out there. So, you have to continue to show improvement, right? That's why there's 100% of level ones you need, and then showing that you're actually moving towards something even more mature, more stable, solid, secure, and then eventually that you're — this is a continuous process, right? It's iterative. It's just like your development process, right? You're always improving the system. You're adding features. You're adding changes, but you also have to make sure that you're taking into account the security behind that because you want to make sure you're protecting your clients, as well.
Eric Fey: What would you like election officials to know, perhaps, more about this system? Because I don't know that there are a lot of election officials when they're going out to buy technology that are thinking about asking a potential vendor if they've gone through this process, or if they did, what were the findings of this process? I mean, do you have any recommendations in terms of — as granular as what questions should an election official ask a vendor about having gone through this process?
Jared Marcotte: Easiest question is, “Have you gone through RABET-V?” And we have a public, I mean, we ask the vendors, "Would you like to be listed publicly on the RABET-V site? And we hope that that's a starting point for a lot of conversations. That's the first question, and then being able to have a conversation of, “Okay, show me that report. I want to see it,” and we digitally sign every report so you know it comes from us, that it's official. I would say the other part of it is is if you have an RFP and you have the ability in your RFP process to be able to ask for things like a security review that that has either the RABET-V characteristics or is RABET-V itself, that's the best place to start because then the vendors get that signal of, “Oh, you, this is going to be a concern that I should work toward,” right? And again, it gives you the best baseline because some folks that I talk to are just like, “Well, we pen test everything.” I'm like, “That's great, that's a good start,” right? Ours is just a little bit more involved, and again, it will give you a lot more insight on what the health of that organization is that's developing that product, and really how is that product maintained and built and constructed, right? It's really about the design decisions they made going into it.
You are going to need some technical expertise to ensure that you can understand the RABET-V reports because, right, it does get very involved — reading threat models, reading architectural diagrams is not necessarily the easiest thing. So, if you have folks that you can lean on, great. If you have tech staff — I know that's not the case for everybody, right? So, that is one thing that I think we're still trying to figure out what is the right solution for. We could potentially lean in more, but then that feels like a violation of a concern that we would have between us and the vendor who would pay for the program. So, I'm still working through that part of the equation.
Laughter
Brianna Lennon: There are offices that are afraid to embrace technology, so they don't, and there are offices that really like being early adopters and become, you know, the innovators and people follow them and emulate what they're doing. But I think there's also a sense, like we're having this conversation with the assumption that all of this technology is needed in the election space, and I think it's almost become a self-fulfilling prophecy that because all this technology exists and so many of us have adopted it — that there really is no way to go backwards. You can't not have an election results website now. You can't not have a poll book. Once you adopt it, you have to keep doing it because your community gets used to it, it becomes your new best practice, like you cannot go backwards. Do you ever — when you're talking, especially to, maybe, smaller entities or very new to technology type of offices. Does that ever come up? That, like, “You know what? Maybe you are small enough that you don't need to buy anything. Maybe you can just keep doing what you're doing.”
Jared Marcotte: Yeah, and I think that comes back to process over everything else. Like what works best for your office over anything, like maybe you don't need to purchase this thing because you operationally just need to make a couple of tweaks or changes, and you're still able to to achieve the same level of throughput as you always have. And I think that's how you right size the solution space for any given office, whether you're, I'll just throw out, in L.A. County to you know, or Boone County — everybody, like even on those extremes or the outsides, I think it really just comes down to what works best for you? How do you best serve your voters? And what is the right solution for you? And that's, we're not, like we don't sell a product, right? We're just trying to improve the space at the end of the day, right? We want to make sure our work counts for something.
Brianna Lennon: In the space that we're in right now, and so much scrutiny about, like making sure there's integrity in our elections and the public having lower levels of confidence or trust — Is there space? Is that a consideration for things like RABET-V and The Turnout, in general, to try to bridge that gap and improve the confidence and the trust from the public's point of view — even though you're not directly working with the public, does it give election officials the tools to make that communication?
Jared Marcotte: Yes, I think that it is baked into all of this. I mean, even just the the data analysis on military and overseas voters and the issues that they face, and and showing that, by and large, if you can meet people where they are and you can provide them with the services that they need in order to be able to get their ballots to vote, return them on time — that helps those voters to feel like their votes will count. Communicating how the process works, you know, ballot duplication and all the other aspects of the election administration around that, I think help give those voters confidence that their ballots are coming back, that they will count, that it will effectively work for them. And noting that the election officials are doing everything they can to make sure that this work[s] — that they're feeling like their needs are being served, and with RABET-V, right? It's looking at these systems and seeing how they can be improved over time, right? And again, RABET-V is iterative. We expect it to constantly be run against these systems regularly, and so, again, we're matching the speed of development. That we find some sort of concern in a system and communicate that immediately, and so the vendor can then make a change and put it back through and say, "Is it still working? Is it good? Are we seeing what we expect?” And being able to note that those changes are happening over time and they're happening rapidly and, I would hope, I mean, I would like to know that at some point we get there to say, like, “Oh, this system is RABET-V verified.” That that does lend itself some confidence. I mean, it probably does a little bit to say that we've been through the security program, and, therefore, it's at a higher level than than most other systems that are out there, but, you know, obviously, to say that RABET-V verification gives the the public some confidence, we're not quite there yet. Hopefully, we'll get there eventually, but you know, until then.
Eric Fey: You've been listening to High Turnout Wide Margins, a podcast that explores local elections administration. I'm your host, Eric Fey alongside Brianna Lennon. A big thanks to KBIA and the Election Center for making this podcast possible. Our Managing Editor is Rebecca Smith. Managing Producer is Aaron Hay. Our Associate Producer is Katie Quinn, and our Digital Producer is Mark Johnson. This has been High Turnout Wide Margins. Thanks for listening.
