Iowa’s App-ocalypse

Listen to this episode

S1: So what’s funny is that morning, the morning before the caucuses started, I sent a message to one of The New York Times internal Slark groups and I said, guys, I’m going to be keeping a really close eye on this app that they’re rolling out in Iowa, because everything I’ve seen about it makes me think that there could be problems.

S2: That’s Sheera Frenkel, who covers cybersecurity for The Times, and she’s been hearing things about the app that many precinct chairs were planning to use to report results to the Iowa Democratic Party.

S3: I had heard that they had security tested it. That tape’s pretty. Experts are really concerned because they don’t know any of the details. And already I’ve seen some people tweeting that they haven’t been able to download it onto their phones. I didn’t think it was going to be as bad as it was. I mean, we’d already filed an initial story saying this app could be your problem. We’re not sure. And then that really quickly turned into a yep. This app is definitely a problem. And we have no idea how big or bad this problem is going to get in the last two minutes.

S4: We’ve heard from the Iowa State Democratic Party. They tell us this. They say that they are doing, quote, quality control. The issue does seem to lie with this major coding error in the app.

S5: Caucus officials saying it was confusing and difficult to download.

S2: All of this has put the Democrats in a bind, not just over a complicated caucus results that have dragged on, but also over a huge push within the party to regain a technological edge over the GOP. Today on the show, how the app that failed in Iowa is just a symptom of something bigger plaguing the Democratic Party. I’m Lizzie O’Leary and this is What Next TBD, a show about technology, power and how the future will be determined. Stay with us.

S6: Before the app that we’re talking about came into play. How are things typically reported?

S7: It was generally done in a very sort of old school, the very efficient way in that people would call in results on a dedicated line. There’s some people that would even walk over results to the main Iowa Democratic Party centers if they were physically close enough to do it. We’re not talking about that many precincts at the end of the day. And so having a dedicated phone system with enough people to operate the lines was a pretty good way of doing this in the past.

S6: But this year, with concerns about election security, Iowa Democrats listen to advice from the National Party and decided to use an app to report results. How was this supposed to work this time?

S7: So there were supposed to be this app and it was a custom built fancy app, which everybody was going to download on their phones. And it would basically do the math for you because Iowa caucuses and because they need to given not just their first choice, but kind of like they’re downrange choices. They figured that if they created an app that would do the basic math for the precinct heads, it would simplify and streamline things. So this is an app that’s basically a calculator combined with like a text messaging service, which just then sent the results over to the main Iowa Democratic Party. So we’re not talking about an app that has to do all that much.

S6: Yeah, it seems pretty straightforward. You would think, you know, we should note that this year’s caucuses were more complex than they have been in the past because instead of one set of results, there were gonna be three. I mean, there was already a higher degree of difficulty here, kind of baked into the process.

S7: Definitely. And we watched a training video that the Iowa Democratic Party created for its precinct heads.

S8: Hello and welcome to Module 7. We are talking a lot more in depth about various scenarios that are with caucus now.

S7: They run through a good nine minute explanation of how did you the math, right, how to divide up all the votes they’ve gotten among all the various candidates and how to send those in. And so they knew the math on this was going to be a little bit tricky.

S8: So that way you’re prepared and we can run as smoothly as a caucus as possible when we are.

S7: But all that being said, the idea that they were going to create this app that was going to essentially do the math for people, I guess it made sense, but nobody talked about it. There was no training on it even in this nine minute training video. The app never gets mentioned.

S6: So like on a really granular level, if you’re a precinct captain and you’re trying to report your results on the app, what what were you seeing? What was happening? The people I spoke to all never got that far.

S9: Well, I think something like one fourth of the precinct heads managed to actually download the app onto their phones. And of those, an even smaller percentage used it to send in their results. So in all the reporting I did, I don’t think I’ve talked to a single person who managed to effectively use it.

S6: Reportedly about sixty three thousand dollars were spent on this. Can you put that in context?

S10: Because out of a normal amount of money, a little. A lot.

S9: That’s not much money. These kinds of things can cost four or five times as much. I spoke to engineers at Google and Facebook who said, look, you wanted a custom built app that was properly vetted and tested and had been approved to be in the Apple store, for instance, that people could easily download it. That’s a yearlong project which we would command, you know, half a million dollars for. That’s not the kind of thing you just throw like sixty three thousand dollars. I think part of the problem was also trying to do this very, very quickly. And in the world of technology, very, very cheaply.

S6: You know, we’ve been talking about the specifics of what happened in Iowa. And I guess I wonder, should we be looking at this as a mistake, a one off, or is it representative of more systemic issues within the Democratic Party?

S1: So maybe both. The only other caucus that we know that was going to use the same specific app is the Nevada caucus. And they’ve now announced that they’re not going to use it anymore. So this particular mistake probably won’t happen again. But at the same time, I think it is symptomatic of a bigger problem when it comes to the Democratic Party.

S11: A lot of the challenges around election technology go back to the very nature of campaigns. They’re cyclical. And for years, the Democrats in particular would pour a lot of money and resources into election year activity and then kind of let things stagnate in between.

S12: I remember back in probably the early 2000s at the DNC, I mean, they would almost literally shut the lights off in between elections.

S10: That’s Nancy Scola. She covers technology for Politico and specifically, she pays attention to the ways tech is used in campaigns. But before her career as a journalist, she actually worked in politics.

S12: I remember at one point, you know, this is that memory is fuzzy. It’s been a, you know, a bunch of years. But at one point, the DNC decided to hang on to I think it was a videographer between cycles. So between, you know, two year elections, election days. And that was a big moment. They were actually going to fund, you know, one or two staffers between. Election, so that’s you know, when you build an app, you try it out for a couple days and you put it on a shelf for two years. It’s not the best way to go about building technology.

S6: The Democrats recognize this as a problem. And in 2004, things started to change with Howard Dean’s campaign.

S13: Today, I announce that I’m running for the presidency of the United States of America.

S12: The Dean campaign really experimented with using the Internet and sort of a major way for the first time for a presidential campaign. They ran online fundraisers using very what in retrospect with very rudimentary graphics. But people got very excited about this kind of new interactive, our mind fundraising and talk about a payoff.

S14: Howard Dean supporters are still counting receipts from a weekend online fundraising effort that amassed half a million bucks.

S12: So people came off that campaign thinking, huh? These tools are actually really powerful. We should figure out what to do with them. A lot of the folks that were part of that Dean digital effort actually, you know, went away for a couple of years, built up digital technology, strategic firms, and then came back for the Obama campaign four years later and really kind of powered that effort, taking some of the lessons learned from the Dean operation and those lessons worked well.

S6: The Obama campaign raised money online. They use tech to organize and go after persuadable voters. By the time Obama made it to the White House, his team had collected both a massive database of supporters and a reputation for innovative digital thinking.

S12: They looked at that and said, wow, these tools are just like, you know, incredibly powerful. We’ve done something really amazing and maybe a little bit rested on their laurels for the next couple years and didn’t really invest in thinking about, OK, how do we build these tools when we’ve a candidate who’s not as compelling as Barack Obama in 2016, Scola says.

S6: The Democrats kind of got complacent. They failed to recognize that the Trump campaign had built a successful digital operation, especially when it came to social media all the way up to pretty much all election day.

S12: The tone the Democrats were taking towards the Trump campaign was, oh my gosh, how badly they have bungled technology. They didn’t seem to have any sort of data operation. They didn’t seem to have any sort of turnout operation, even really sort of online persuasion operation. What we realized as reporters and as is people and people, the Democratic Party realized this, too, was that they had quietly built a little bit of a digital juggernaut, that sort of tip of the spear. There was Donald Trump himself, his use of social media. There was a deeper layer, which pretty much consisted of Facebook ads. And again, this was something that was mocked in Democratic Party politics. This isn’t how you use these new Internet tools to win elections. When he did no God sort of had to kind of take a step back and say, you know, we’ve maybe kind of been riding on the past glory of the 2008 Obama campaign, assuming that we have some sort of fundamental advantage that isn’t going to go away. And you just really couldn’t ignore the fact that after Donald Trump rode the road, these things, the White House, that that advantage wasn’t there anymore. So what you saw in Democratic Party politics is a number of people who are veterans of the Clinton campaign, the Obama campaign, all these different Democratic, you know, sort of very high profile campaigns, take a step back and say, OK, what do we do? What do we do to regain that advantage to do that?

S15: Democrats invested now in the central party, but in a series of incubators, non-profits that were designed to boost Democratic tech. One of those groups was acronym, a digital consulting firm, acronym did a lot of things, including funding the company that made the Iowa app, a company called Shadow.

S6: I know it’s a lot of names, a shadow acronym, groups like these. They were supposed to be the things that helped Democrats get their digital edge back.

S12: One of the ironies of that this whole acronym situation is they were exactly designed to be the answer to a lot of these questions that we’re discussing to a lot of these challenges. Right. So the idea there was let’s have a digital strategy firm. Let’s have folks that build campaign apps and campaign tech. Let’s have a part of this outfit that only focuses on Facebook ads and how you use them against Donald Trump. So the idea really there was to marry all these pieces together. The whole point of this was to build a lasting infrastructure for Democrats. The whole point of shadow, the name actually refers to the idea of being a shadow infrastructure for a party that wasn’t willing to build it. So that’s sort of the irony of some of this, is that this is exactly the problems that acronyms and shadow were meant to to fight.

S6: One of the things that strikes me about some of these operations is that there seem to be multiple ones for the Democrats and for the Republicans. It’s really centered around the Trump organization. How does that split shake out? So it’s.

S12: In a lot of ways, completely not surprising to see Democrats approach technology in this way because this is how the Democratic Party operates, or at least has in kind of the last few decades, is that when people think they have a good idea about how the Democratic Party should function differently, they do it outside the party. So the thinking is let’s innovate outside. The party eventually will push the official Democratic Party structure to adopt this sort of best practices, lessons learned that we’ve come up with. You see that with things like Catalist, which was some clintonworld veterans, kind of came up with this idea of let’s build a different way of handling data than the DNC was doing at the time, sort of voter data that caused some tension in the party because you had the sort of official operation. And so it sort of outside funded operations you saw with the Obama campaign in 2008. The Obama for America transition to Organizing for America. They tried to awkwardly bolted it onto the DNC at some point, never really worked very well. So this is just how Democrats seem to operate when it comes to technology. It’s just sort of beginning again. Doesn’t seem to work particularly well.

S6: Why do you think it doesn’t work well?

S12: It’s sort of the fundamental challenges of building campaign technology. So two things. One is there’s not a great deal of money to be made in those sort of tools because some of the other tools that work in campaign, the campaign environment, say fundraising, you can build that tool once tested on the ground in a campaign, then go sell it to corporate America. Go sell it to non-profits. Go sell it to a bunch of other organizations. It’s not a great deal of market for organizing digital tools. So people really don’t want to kind of get in the business of building those things. So that’s that’s one challenge. The other challenges that the people who are your sort of test clients, that people are going to give you feedback. The people that are going to fund the development of some of these tools are competing against each other. Right. You know, the Sanders campaign and the Clinton campaign have famously feuded over data and technology. So it’s really hard to kind of be in that environment where you say, you know, people aren’t necessarily going to say, hey, this is how the tool worked for me in the field, feed it back to the developers and try to improve upon it. So it’s a really tough environment to build technology.

S6: Is there a possibility that this whole experience in Iowa actually sets back people who were we’re making some progress in their minds to close that technological gap between the parties?

S12: I think absolutely. I think if you look at the Nevada Democratic Party was set to use the same shadow up in their own caucuses that are coming up and they’ve said, no, no, no, you’re absolutely not going to go and do that. And that’s how you make technology like this better, right? You test out in the field and sort of real life environments. And that’s not going to happen with this app because of this debacle.

S16: Nancy Scola, thank you so much. Thank you. As a Nancy Scola is a senior technology reporter for Politico. Sheera Frenkel, who you heard at the top of the show, is a cybersecurity reporter for The New York Times. OK. That’s it for today. What next? TBD is produced by Ethan Brooks and hosted by me. Lizzie O’Leary. And it’s part of the larger what next family. TBD is also part of Future Tense, a partnership of Slate, Arizona State University and New America. Mary will be back on your feet on Monday. If you didn’t catch her show about Mike Bloomberg’s campaign, go back and listen. You will not regret it. All right. Thanks for listening. Talk to you next week.