How successfully run your first online election: Interview with CEO of Neuvote Matthew Heuman
On October 24, municipal elections were held in the province of Ontario, and Neuvote proudly and successfully hosted its first online Election in the Township of Ignace. We interviewed Matthew Heuman, CEO of Neuvote, to learn more about this incredible achievement and the future of democracy.
Matthew, please tell us a little about the town and the opportunity to bring online voting to its citizens.
Many people may not realize this, but Canada is one of the global leaders in online voting access right now. It's widely regarded as beneficial but very difficult to pull off. Luckily, in Canada, we have a very high tolerance for implementing online voting, and it's been used for about 20 years, and it's been growing at an exponential rate this last little bit. Every election cycle, it's been quite literally doubling almost, and this previous Election in Ontario, at the municipal level, there were just shy of 300 municipalities using online voting. So it's phenomenal to see that.
I do think the technology needs more work to be implemented at a higher governmental level, but we see some interest at higher levels.
Ignace, a small municipality in Northern Ontario, and our first job for online voting, which was b a long time coming, worked with the town of Ignace. And the main reason they reached out and wanted to implement this was that it's a community where many people work outside of town or are seasonal residents. So you had a town clerk whose job was to bring access to the vote. But how do you do that when residents permitted to vote don't live in the community; they live elsewhere? So they've done paper ballot elections historically and then experimented with a vote by mail in one instance. But while voting by mail is a great option, it provides many benefits and is easily understood by the electorate. The challenge for voting by mail is the chain of custody - ensuring the ballots are handled appropriately. Of course, the other big hurdle is just transmission of the vote; how long it takes for the votes to get from one person to another, from the post office to the voter, etcetera. So they ‘elected’ to try online voting.
This was their first use case of online voting ever. So they went from nothing to 100% online in this Election. And we're proud to say that it went flawlessly.
Several times, I flew out to Ignace to provide education services and help voters understand the transition. I know that this technology is perceived as a challenge or a period where people must understand how the process works. It's not just cut and dry. There are a lot of differences. There are a lot of similarities but a lot of differences in how people access the vote. They need to use a computer or some device to be able to cast their vote. And then generally, there's more or less instilling trust in how this marking of a ballot, which they're typically used to, converts into making the selections online and depositing it into a virtual ballot box.
The result was the highest turnout in Ignace's history. So that was, of course, an excellent indicator that we were successful. And the incoming Council was more than happy with the result. The town clerk seemed satisfied with the service, and we are ecstatic to have completed it. It is nerve-wracking. I'm not going to lie; doing our first primary binding Election but was a great experience overall.
How long did it take to organize the online Election fully? At what stages did it include?
Well, many may not realize this, but I think especially in the US, people have become increasingly aware that governments hire companies like ours to service elections. Though it is challenging, imagine throwing a party for your entire town that lasts for a day or so and needs to go off without a single hitch. Logistically, there are a lot of challenges. There are a lot of things that need to be brought together to provide an election. So the stages are typically a government or a council produces an RFP, a request for proposal, and then companies like Neuvote bid on those. With Ignace, we won the contract and started discussing with them how we will set up the different stages for instituting an election.
Typically, there's a planning stage where you go over all the elements you need to provide to the Township. The essential thing is disaster recovery and backups, what happens when things go wrong. So there's a whole stage for that. And then it's discovering all the different components, bringing them together. And you have done that for almost a year. And then you execute the Election on Election Day. So it's pretty logistically challenging, but once you get into the flow of things, it comes together quite nicely.
What was the process for Ignace citizens to vote online? How did they vote?
The online voting process is this; voters receive some “credentials” (a certificate that proves their identity), which could be delivered via mail or e-mail. They take their credential, log into the online voting service, and then cast their ballot. Now that can be done on, like I said, any Internet-connected device. They could go into the municipal office and cast their ballot in person if they wanted to, if they needed assistance or if they just felt it was more agreeable to them. They make their selections as they have been validated once they've confirmed everything; They review their ballots and ensure that those selections are exactly as they intend. Then you have captured the voter's intent. And this is important because, just like an in-person election, once that ballot is inside the ballot box, it's done, the vote is complete, and you can't reverse that decision. So similarly, in our online voting process, the voter is provided a validation code once the ballot has been submitted. Now, this is an element that we call in the industry “verifiability”. It is vital, and we feel essential, mission-critical to the success of online voting because it provides a mechanism for a voter to check the validity of their vote after the vote has been cast. It's an optional step, but once they validate that their vote has been cast accurately, that's it. And then every vote goes into the virtual ballot box. And then eventually, when the Election closes, just like in an in-person election, we essentially decrypt the votes, unpackage the ballot box and count the votes.
That is the significant benefit of online voting. After all, it can be done more expeditiously than regular elections because it's all automated.
But with end-to-end (E2E) verifiability, if somebody wanted to do a recount or contest the vote, you have a mechanism to go back and audit these digital votes, which is relatively new in the industry. It's a technology that is coming to fruition now. And we are the first ones who are using it.
Sounds impressive. What challenges did you face when organizing the first elections, and how did you solve them?
Oh wow. Challenges. The biggest one, I think, is education for the town clerk: informing, making them feel comfortable with the technology, and educating the voters on the technology, too, because we are talking about not only a paradigm shift but a substantial change in the procedures of how online voting runs—pulling the different components together. It was a challenge, but it was handled quite well. For example, we had to print off and send out pieces of mail for every registered or pre-registered voter to receive their credentials physically and then basically log the different voters that were coming in and registering after and then distributing their credentials; that was relatively easier because you're handling it digitally, but just keeping track of everything.
And honestly, it's got to go without a hitch. There can't be any disruptions. There can't be any issues. So everything has to be quality-checked along the way in a very delicate manner because we're talking about a municipality's direction and political leadership. If we extrapolate that to a province, or state government, like a federal government or a state, these have huge impacts. So we can't have any issues develop. We need to make sure that everything is complete and 100% accurate, and we need to be sure that there are no hitches in the election process.
You said that turnout was huge, but what was the exact turnout during the Election in Ignace, and how does that compare with their previous municipal Election?
So Ignace is a small town. They typically have above-average participation; I think it floats between 40 to 50% historically with their paper ballots or vote by mail. We got 69.11%, so just shy of 70%. We were trying to edge over that 70% number. I was there physically to help the clerk facilitate the Election. Compared to other municipalities like Toronto, where we are now, I believe Toronto's municipal election turnout was around 28%. So this is a big problem in elections and democracy in general when you have only a few people taking part in an Election; you’re not getting the real benefit of democracy, which is to vote for the people, using their voice.
How can we confidently say that a city and its town's representatives are being accurately elected if you have less than 30% of the populace participating? If you break down the winning candidates in an election, for example, 28%, maybe 30% of that number. Because all you need, in a “first past the post system,” is a threshold greater than every other candidate to win. So realistically, you have 10%, 15% of the population electing their leader. We feel like this is a big problem, but at the same time, this is why we got into the online industry.
We think online voting can provide increased access because of many of the challenges we've seen in elections; it's not just that people don't want to vote. People are busy. It's not always easy to get access to the voting process.
They have multiple family obligations, work obligations, and other issues that come up. To be able to open your phone and cast a vote within a few minutes is a huge benefit and reflects how we typically operate in modern society. Going online makes sense because you can't negate the efficiency of this online voting in a democracy, in general. We order our groceries online and go to the doctor online. Taking part in an election, we feel it should be handled online, but with the proviso that it’s held well and without any issues.
What was the percentage of voters that chose to vote online?
For Ignace, it was all of them. The entire Ignace Township elected to do the complete Election online. They could have done a paper ballot backup but decided to run the whole thing online. And because the online process was flexible, voters could vote at home and for their convenience or still vote in person, but just on a device in the municipal office. It provided a singular access point or tabulation point to pretty much everybody.
What was the best part about helping Ignace Township host its first-ever online Election?
For us, it was seeing the results and the highest-ever voter turnout. I can't understate that it was a vast, massive marker for us because it proves the theory that we've been working with for some time; if you provide online voting as a point of access to the vote, it will increase turnout, and our numbers now prove that out.
That's probably a hypothesis we've understood or thought about for a long time. But to see it be proven was perhaps the best aspect of the Election. And also, just seeing people run an online election, seeing the whole procedure unfold right in front of you as our first Election, was a massive milestone for us.
What feedback did you get about the new option to go online from Ignace Township?
Well, there was some pushback at first, 100%. Some people made comments that we should have offered paper ballots. Some people may not have technical literacy, but the one thing that we are lucky enough to have in Canada is this historical narrative of online voting. We've seen it in other municipalities for about 20 years now. To implement online voting at such a high level, we had to pay attention to the demographic numbers that said the older generation usually votes using paper ballots.
There were some cases where people were very old and may not have been very comfortable with technology. So they could go in and get help from the voter assistance center. But at the same time, I don't feel we should hold back the technology simply because some individuals may not be technically literate. I know every other service, including many other government services, is already handled online because you can't understate the efficiency and optimization of digitizing these services. For voting, some people were feeling: "I don't want to do this, I don't trust it, or I don't understand it, or I don't have a computer at home or something."
As for the vast majority of voters, though, they were more than happy to have this access. People, I think, are inherently attracted to convenience. When digitized services emerge, they typically last because it just works.
How can you be sure that your vote was counted accurately?
Simple. It is our E2E Verifiability. Let's imagine you vote in person. It means we removed the online part. So you have a ballot that you put into a ballot box. Well, how, as a voter, can I be sure that the ballot box wasn't manipulated somehow? There wasn't anything going wrong.
Our End-to-End verifiability or “E2E verifiability” is essentially a mathematical algorithm, a protocol, I should say, that provides a digital receipt attached to every vote.
Now that means you can track it through several stages throughout the process, whether your vote was cast as intended, recorded as cast, and then counted as recorded. So what that means is my vote was registered by me accurately, it was deposited into the ballot box accurately, and it was calculated accurately. This code, this string essentially that's left over the receipt, if you will, allows the system to check the results and ensure that everything was working mathematically.
The critical thing here, though, is you can't reveal who a voter voted for. That's the big thing. It's essential for voter privacy. That's why the protocol is cryptographic; we use a level of encryption here to ensure that voter privacy is not breached but simultaneously prove without a shadow of a doubt that this all happened accurately. So that complexity element is very different from what people are used to. But in what we're seeing in the US right now, where they don't have online voting yet, we're seeing this element of essentially having to ensure that each vote is counted accurately like that, that people need to trust the election results more and more.
So this E2E verifiability component is a new addition to online voting, and they're working very hard on instituting this in the US election system. So the voter can be confident in their election results even if they vote using paper ballots. That's what needs to happen for online voting. We have brought this into the online world because we feel this is essential. Voters need to be 100% sure that whatever happens throughout this process, the actual voting process, they are 100% confident that their vote was registered accurately. And again, being able to have this protocol deep and analyzed by a third party or somebody who says I don't believe in the election results is a huge benefit to online voting.
What lessons have you learned from this Election?
The biggest takeaway was, of course, the need for voter education. Make sure that people feel very comfortable with the technology. The other one, I think, is regarding E2E verifiability. Our website has several videos on the importance of E2E verifiability. I'm sure you've looked at them, or we've talked about it at length about what this means, and it's fascinating when you get into it, but there's a little bit of complexity to it because when people think about online services, they're used to justifying it working. You go on Amazon, buy something, put in your credit card information, and it goes, that's it, it goes through. With online voting, we have this element that we had to put in, or as we did, we put it in. Still, I should say the academics and the people thinking about this technology were the ones who thought about this around 20-30 years ago. And the idea is, and this doesn't just go for online; this goes for any vote.
Was the online Election for Ignace cheaper than traditional in-person voting four years ago?
Oh, it's much cheaper. From the staffing perspective, to run an election like the one at Ignace Township, a smaller town, they needed 18-20 people. Because you have different places and people need to be worked with staffing there and everything, you're reducing your staffing costs to reduce that to one central location.
A big thing we've seen in elections recently is the number of people volunteering to work Election turnout is getting smaller and smaller. It goes back to the issue of people being busy. It takes a lot to volunteer and essentially works an election, especially when you're counting by hand. You could sometimes work until the middle of the night; many people don't want to do that. So to be able to reduce staffing costs significantly, that's massive. You're saving all those people who must be recruited and paid. You're holding that cost by instituting the technology.
There is an upfront fee; it is like renting a digital, optical scan tabulator or some other machine to help process the votes. But in Ignace’s case, you're not spending as much time because you're reducing 18 people needed for in-person voting, organizing 18 people, all the different locations, and all the different logistical challenges that go into running an election. Instead, you're taking that multiple-location complexity away and bringing it into a central place. So logistically, it just massively reduces the cost. As for the technology, it’s about the same as running any other digital service; the tech is already built, and you're not creating anything new. You’re just offering a service.
The experience of organizing the election in Ignace was incredible. Where do you see opportunities to introduce online voting except Ignace?
Right now, in Canada, we see tremendous interest in online voting. Some provinces are looking at it. It's coming. And I think people are starting to realize, and COVID highlighted, that if we don't improve voting access and convenience processes, people will not participate. Toronto's recent municipal numbers were 28% participation, and Ontario's provincial numbers were the lowest turnout in the province's history. You can't claim that democracy is running efficiently or as it should be when the dismal numbers are right in your face.
The data doesn't lie; people are not taking part in elections. You could propose a bunch of different hypotheses as to why that is. At the end of the day, though, if election officials don't offer alternative solutions, how many elections will we see where the turnout is even more meager? You can't claim that you're running a representative democracy and not representing anything more than a handful of people with the time and availability to cast the ballot. So I think we will see, at least in Canada, a high number of online voting initiatives springing up.
The challenge for companies like ours is that we need to ensure that they are appropriately handled because the absolute last thing we want is an issue to emerge where something happens and people get scared about it. Some online voting companies have had problems, and some countries have tried online voting but experienced issues. Switzerland is a great example of online voting that went awry. Switzerland had an online voting initiative that would run its entire country off of online elections. Of course, Switzerland has a different form of democracy. Still, they vote four times a year, so for them, the goal was again reducing the logistics, making things more accessible, increasing access, et cetera. Unfortunately, the vendor that was providing that service screwed up in a way that was so substantive that Switzerland had to pull back on the initiative. So, I take that to heart and think about these issues every day because if all it takes is one major issue, that's it.
Trust in a democracy right now is already fragile. The trust in the system. That's why Neuvote exists. And this is my life's work.
What makes Neuvote online voting different from other voting systems?
Right off the bat, we believe in providing optimum security. Security is our number one thing, as we think that if we do this accurately, we can grow. My life thesis, the path I've followed for the last six years now, has been based on what happens when we improve access to the vote and more people are voting right and making sure we're improving the system of democracy.
I tattooed democracy on my neck so that people understand what I'm about the second we meet. I truly believe in this and in my company. When we were looking at providing the online voting solution, the first hypothesis we developed was that the paper ballot is probably a mechanism that needs to stay. In Ignace’s case, we didn't use a paper ballot election. They decided not to do it. But we are building a remote-hybrid paper ballot solution for higher-level elections that combines paper balloting with online voting.
When things are contested at higher governmental levels, which we're anticipating, we've seen the ability to audit the digital and paper ballot votes as critically important. It is instrumental in ensuring that an Election is handled smoothly, with accurate results and that people trust the system. The other thing we've learned throughout building this technology is that the E2E verifiability component is probably just as critical as a paper ballot. Having the ability to have a physical ballot and a digital auditing mechanism that people can rely on to receive these results, the combination of those two things, is just phenomenal for system security. Because it provides two mechanisms for trust, when we do a digital count in a physical count, we have a one-to-one match, so we can be sure that the Election was handled without any issues.
And that is the most significant benefit of what we do versus others: optimization, security, and efficiency.
What were security processes in place to ensure the election results were accurate and no tampering could occur?
Our code is tested by third parties (independent analysis). That is one thing that's a little bit different from us. I learned early on there has always been in this industry a weird dichotomy between the academic and research side and online voting companies, as they're usually butting heads. Because if the researchers find a problem in a company’s online voting system, they don't want to be embarrassed; they don't want their clients asking them tough questions. But why wouldn't you trust these academics to test and find these systems? For example, if somebody finds an issue with Google, Google pays a considerable amount of money to those people to tell them about the problem.
At Neuvote, we want to work with academics and researchers. This year, for example, before the Ignace pilot, we handed our system over to Canada's leading academic for review. He said he wanted to test it. We got the thumbs up from him, saying no vulnerability found. So for us, that was phenomenal. It was a pleasure to work with him. We were at a conference in Austria talking about this issue when somebody asked me why the academic reached out to all the companies in the space and what the response was. He said we were the only ones to participate other than, I think, Switzerland, which I talked about, is working on its own online voting thing.
Sure, if somebody could find a vulnerability and mess with the Election, that vulnerability led to a failed election, and now we're dead. Like no one's going to trust us. We're not going to be able to provide our service anymore. So for us to have this frustrating relationship with the people just trying to do things right and ensure that nothing is threatened to democracy doesn't make sense. These people are passionate about their work. We don’t have to criticize them.
As a citizen of Ontario, where this technology is emerging, I want to ensure that if I cast a vote online, I'm electing the right representative as that system selects the right representative. It's not in my interest to have a manipulated system because I can't trust it. Money isn't the driving factor. I think what we do is ensure that we're developing a system that genuinely works. It benefits society more so than anything else.
From this conversation, I understand that you enjoyed organizing the online Election in Ignace.
I did. I got into this business because I had an interesting theory. Not only that, but I eat, breathe, and sleep to solve this problem (organizing online elections). You see it all the time in the news; democracy is under threat, blah, blah, blah. And, whatever your political affiliation is, we're not political. We don't care. We simply think that access to the vote and security needs to be improved, and that's it, right? It's very simple. We want to make sure that that access is handled appropriately. Ignace allowed us to be the first to organize online elections there. It was a moment that I'll never forget.