I’ve spent a lot of time on state websites, both because I work on one and because in 2018, I built a site around clearer voting information for each state. (Will post a link for the 2020 version when it’s done.) As a result, I have a few thoughts on how we could improve election web content. This subject’s dear to me as a citizen, too. I think information about voting and voting procedures should be extremely, rigorously clear. It’s the institution most symbolic of Democracy – one of the ones we’re supposed to live up to.
Here, I’m focused on information architecture. There are lots of other things about election web content we could look at, but information architecture is the foundation. Structure and organization are especially important in public sector content, where information is both voluminous and complicated. My thesis here is that many state elections websites set up relationships with an audience who is either expert or interested in the voting process (or both). My underlying assumption is that their actual audience nonexpert and just trying to figure out what they have to do to cast a vote.
Into the bureaucracy!
Perhaps you are not intimately acquainted with the structure state governments. At the risk of presenting material that is so scintillating you’ll need to take cover, here is some context for what follows.
Election administration in most states falls under the purview of the Secretary of State (SoS). Sometimes it’s another elected official or a commission/board, but quite often it’s the SoS. The SoS also does other, non-election stuff, e.g. investigate corporate fraud, register lobbyists, help people start businesses, and who knows what else. So elections are just one part of their portfolio of responsibilities.
But this “just one part” is administratively very complicated. Elections involve literally everyone and have lots of moving pieces – multiple ways to register and to vote, candidacy filing info, press packets, identification laws, etc. You’ve got to train volunteers and manage & count ballots and do comms about the results. And then maybe recount ballots and do more comms.
Voters might have questions about this. So might candidates. Reporters definitely do.
So if you are a Secretary of State or a Board of Elections, you have this complicated process about which there is much to say. You’re concerned with 3 different audiences who need related-but-different information for different reasons.
Fortunately, you have a staff, and some of those people do the website, where you can put that information. Of course, therein’s the challenge for public sector content writers. What do you do so that people can find what they need? What do you avoid?
Old, nonresponsive sites and giant rails
This post isn’t about design, except when it is. And since this piece of the problem is the most obvious, let’s start here: many Secretaries of State’s websites aren’t mobile responsive, or they’ve packed all the content into a narrow, canyonesque column:
Iowa’s Secretary of State’s landing page
Now, this isn’t objectively bad. This blog, for instance, places most of the content in the center of the screen. And that’s fine if you’re just reading text. But if you’ve got multiple columns of content, some vertical and some horizontal, and lots of links, and images, and banners, and so forth and so on, then squishing everything together is trouble. It’s like trying to move 4 bedrooms of furniture into a studio.
Louisiana, Massachusetts, Wisconsin, Missouri, Oklahoma, and Illinois are a few of the others that follow this trend.
Oklahoma’s Board of Elections landing page is 957 px across.
To be clear, not every state elections site is this way. New York’s Board of Elections is definitely not slick, but it is functional. Oregon, Colorado, and Ohio make better use of space, too.
But many don’t. And the problem with this trend becomes more obvious when you attend to just how many links state election sites like to include on their landing pages.
“With just one click!”
It used to be common practice to make everything reachable in 3 clicks or fewer. And I bet “put it all in one place” still works if your content is something people are hyperliterate about. Amazon and Gap can get away with a hundred of links on their landing pages. But I’m not so sure it works for state election sites
Here’s New Hampshire’s elections landing page.
We’ve already seen the Big Rails trend. But now, check out the volume of links squeezed into the center content. It’s like they’re creating a site map. There’s plenty to snipe at here, e.g. could one replace those 39 left sidebar links “Past Election Results” archive? But the underlying issue is the chaos. There’s a lot to parse to find what you’re looking at. Here’s a zoom-in of the top links in the center column:
Links for all 3 audiences have been mixed together and organized apparently at random. I have no idea what Master Election Ballot Column Rotation is, but I am sure that someone does and is looking for it. I am also sure that whoever that is does not have all those FAQs about voting. This is a grocery store with just 2 aisles: Good luck readers.
Other states that have site-mapped their landing page include Kansas, West Virginia, Virginia, Illinois, Wyoming, and California.
Virginia Secretary of State’s elections landing (beneath the banner) is a giant, sub-headed list of links.
A related trend: instead of listing every link on the landing page, some states provide you with a… robust…navigation system. South Dakota is the most extraordinary example of this. First, it has 3 nav bars:
3 nav menus
And then, when you “click one of the boxes and SCROLL DOWN” (as instructed), you open up 2 more navs.
2 “surprise!” nav menus
Indiana has – well, I’m not sure how many navs to call this. More than 3, I think.
And don’t miss the “featured” carousel in the bottom left. Hope this isn’t that important.
Multiple navigation systems are confusing – possibly even more than are site-mapped home pages. It’s like being on a game show and having to pick what’s behind the door – except that instead of 3 doors, there are 35. Also, it’s not fun. This information architecture that doesn’t reassure voters they’re in the right place, doesn’t anticipate what they’re looking for, and doesn’t make an complex system easier to parse.
To be clear, it’s not always terrible for sites to clutter their home pages with links. Lots of successful sites do: Buzzfeed and HowStuffWorks for example. Pinterest is just a giant mosaic of links. Amazon fills up nearly every pixel with links.
It’s an issue of audience.
IA and audience
Web content – all content, I think – always posits a relationship with its consumer. The NFL.com headline “Mock Draft: Murray to Cardinals” assumes you’re following NFL news enough to know what a mock draft is and who Murray is and why it’s newsworthy that he’s headed to the Cardinals. Cooking blogs that spend 2000 words on how much the chef likes apple pie and how much the chef’s family likes the chef’s apple pie before getting to the recipe assume you care about the chef’s opinion (or food-as-social-bonding-agent, or something). This weather site thinks its users must be pretty irreverent. I think you’re interested enough in civic engagement or writing or communications that you’re willing to wade through a couple thousand words on state elections websites.
So, to restate my thesis: What feels off to me about many elections websites is that they set up a relationship with an audience who’s either expert or mainly interested in the voting process or both. I’d be willing to bet that this is not the audience they actually have. This is similar to the point that researcher Dana Chisnell has made that the paths election administrators think constituents take to casting a vote doesn’t match their actual path. Most voters aren’t that engaged with what they would be voting for in the first place, let alone the process by which you cast a vote: 40-50% of the country doesn’t vote in presidential elections, and 50-60% in midterms for pretty much as long as we’ve been tracking the data. People don’t always know who their congressional representatives are, or where they stand on the issues they care about. And then when they do try to engage, the front door might be this:
It’s easy to blame someone for low voter turnout. Schools, millennials, and parents are the usual suspects. But my take is that you should meet your audience where they are. And where they are is that the process is complicated and only relevant insofar as it leads to voting, and voting itself doesn’t seem like such a prize when politics – at least nationally – is a shitshow. There’s got to be a better way to invite constituents into the process, to make it something they feel ownership over, and not alien to.
And fortunately, some states do this well.
Constituent-friendly elections sites
It’s probably just a coincidence that many of the constituent-friendliest elections sites also had the smallest pictures of the Secretary of State – or no picture.
More importantly, the landing pages have distinguishable, prioritized links, and they imagine a relationship with a nonexpert audience for whom the point is voting, not the voting process. Or perhaps it’s better to say, they imagine an audience who comes to them asking, “I’m trying to vote. What do I have to do?”
Minnesota has 4 featured links beneath its banner: Get registered, find out where to vote on election day, other ways to vote than in person, and what’s on the ballot.
These are basically the 4 questions every constituent has to ask. (A lot of people have to ask about ID, but you don’t need ID in Minnesota.
This doesn’t mean that Secretary Simon’s site doesn’t have the same information as any other SoS’s site. Just like with New Hampshire, you can get data about past elections here. They’re 2 intuitive clicks into the architecture: first to data and maps, and second to election results. This makes sense. “Past election results” is a specialized desire for a specialized audience, whereas “register to vote” applies to everybody in Minnesota.
Montana has a similarly intuitive front door (as long as you make it past the truly massive banner). There are more links here than Minnesota has, but not so many that its crowded. Plus, it’s currently January of 2019, only a few months after the midterms. The primary audience at this particular moment in time might be reporters looking for voter participation numbers and election results.
Louisiana’s site, though dated and not mobile responsive, is still thoughtful about its audience. It provides 3 doors for each of its 3 audiences:
Even though there are quite a few links, they’re nicely sectioned off. You only have to navigate small group of them – the group that applies to you – to figure out where you want to go. Also, in spite of its lack of responsiveness, these links look much more screen reader friendly than some of the flashy-but-textless links that many of the other sites sport.
These aren’t the only sites that have a better take on IA. Nevada and Hawaii both deserves shoutouts. Mississippi and New York are cut from the same, totally satisfactory cloth. Alabama does well, if you can look past the banner/. Ditto Tennesee.
One of the most profound things I’ve read with respect to quality state government digital content was — because of course it was — a Tweet: “Good digital work in government is a million silent nods of pleasantly surprised approval, not one loud round of applause.” I’m not sure there’s much of an incentive structure for creating good web content about voting. Does anybody vote for their Secretary of State based on their digital content’s quality? (Could I have ended that last sentence after “State”?)
That said, it seems clear that there are public servants thinking hard about audience, and designing information architecture accordingly. This is good, since laying out complicated information in a clear ways seems like one of the most important jobs that governments have. It affects everything from health insurance delivery to home ownership.
And voting, of course. It’s one of the oldest institutions in the U.S. It was innovative in its time, and I still think it feels innovative now. Let’s hope that someday, the country is, without exception, just as innovative in how we communicate about it.