Dana  Chisnell 

Dana brings deep experience in civic design, starting with research at the National Institute of Standards and Technology (NIST) into the use of language in instructions on ballots (with Ginny Redish), and work on standards and testing for poll worker documentation for the Voluntary Voting System Guidelines (VVSG). She is also an expert in plain language and usability for older adults, including ground-breaking work at AARP that was the basis for several requirements in the Web Content Accessibility Guidelines (WCAG).

She teaches design in government at the Harvard Kennedy School of Government, and in recent years, polished off a 2-year stint as “generalist problem solver” for the United States Digital Service in the Obama White House, doing user research and civic design across agencies.

As the editor of the Field Guides To Ensuring Voter Intent, she has taught thousands of election officials how to improve ballots, websites and other election materials to ensure voter intent. She worked on the Anywhere Ballot, a ballot marking interface tested for accessibility by people with cognitive disabilities and low literacy.

Dana and Jeff Rubin wrote the Handbook of Usability Testing Second Edition (Wiley 2008), the seminal book on the topic.

Democracy is a Design Problem

Every great designed experience starts with the stories of individual humans. At the Center for Civic Design, Dana Chisnell and her team collected a thousand stories from U.S. voters over 5 years. The stories revealed two massive gaps in the process.

First, people who administer elections and voters have very different mental models on the process of voting. The second gap was between privileged voters and burdened voters. These gaps explained why it’s harder than it should be to vote in the U.S. and showed that policies meant to make things better had unintended consequences that actually make it worse.

Good morning.


Let's start with a story from among the thousands

of stories that my team has collected

over the last 15 years.


This is Bill.

When I met him, he struck me as somebody

who had been through a lot.

At first glance it was impossible to tell just how old he was.

In my head I was guessing that he was around 70.

He walked with a limp and his right arm swung

a little bit differently from his left.

He had a snappy brown leather blazer and he wore a big toothy smile.

He was cheerful and he was very flirty.


This was the fall of 2013 and

we were at the California public library.

We met there because Bill was in a study my team was doing

about information about voting

and elections that is published by California counties.

I asked Bill about the upcoming election.

He said he didn't know where to start.

I have been incarcerated ma'am.


OK I said, trying to be cool,

you know there is an election coming up, right?

He said yes, I said great, let's have a look at this booklet.

I gave him instructions to mark it up

and asked him if he had any questions.

Are you ready to start?

He said, “yes, but I have a lot of trouble reading.

I'm only just learning.

It takes me a lot of time”.

I said, that's fine, we've got all morning,

we can go through the booklet

or talk about whatever you want.


Bill was an adult learner in this programme at the library.

One of the first things he encountered that he liked

was the voters Bill of Rights.

This was a revelation to him.

Especially the part that said that he could vote.

Even though he had been in prison.

Ma'am, he said to me, is this document real,

is it true that I can vote even though I have been in prison,

he wanted to know if I had made this up for the test!


I said, no, this is real, it's true.

He said, “how do I get to do that?

How do I get to vote?” I said you just register,

it's pretty simple,

we can do it online here today when we're done.

He said, that's amazing that I can do that.

He was really energised now to look

through the rest of the booklet.

Later he told me he felt empowered.

We had a really great session.

He struggled through most of the 24‑page booklet.

I learned a tonne.

And he was true to his word.

It did take a while.


But when Bill finished and I asked all my questions

he wanted to know could he register to vote.

I had a partner there from

the League of Wales women voters and

she had already pulled up the website

where he could register to vote,

said come on over here I will help you.

In two minutes, Bill was registered to vote.

He practically floated out of the room on his way out.

He shook hands with everybody.


Later I found out that Bill was 56‑years‑old

and he had been in prison for 40 years.

But now, he could vote.

This is why I do this work, for people like Bill.

So, we got him registered to vote but

what's the rest of his experience likely to be like?

That was just one hurdle.

To understand the experience that Bill might have,

my team started with a question: Oh, technology!

Two, actually, two questions:

The first big question helped us find the beginning

of the journey that American voters go through.


How do they think,

what questions do they actually have

leading up to an election?

But we also wanted to look at

how election administrators think about elections.

30,000 election jurisdictions in the United States,

they are mostly run in counties.

We actually started to look at the second question first

and we did this in two phases,

the first was looking at 145 county election websites.

This was very deliberately chosen sample

out of the 5,000 websites we could find

and it looked at different sizes of counties

in terms of area and population, density, race,

educational levels, things like that.


So, we literally catalogued what was on the pages.

The second was, we invited people to come do a usability test.

We asked them as our first question

what questions do you have leading up to the election?

And then asked them what their most important question

and asked them to try to find the answers

on their local government websites.


Here is what we learned:

This is how election officials think about

the process of voting in an election.

It starts with announcing that there is

an election coming up and then goes to registering to vote.

Ends with marking the ballot and checking the results.

This is largely chronological based on

lots of federal and state deadlines that they are given.

It outlines the chronological process.

But when we talked to voters,

when we asked them what their questions were about

how elections worked and what was going on in their heads,

this is what we heard:

What's on the ballot was the number one question.


This was the driver for everything that happened afterwards.

That's actually several steps into the

chronological process, about number eight out of ten.

So, this finding was a big deal.

It rocked the world of elections administration!

And when we went out talking about the questions

the voters have, and how they are

and are not being answered on websites that

are coming out of election offices,

those folks, those election administrators were shocked.

They really hadn't thought about it this way,

but now they knew why they were

experiencing some of the pain that they had

endured in trying to support voters.


So, it turns out this is not exclusive to websites,

this problem of the mismatch of mental models

as we tested voter information and education and

lots of different channels including print.

Voter starts by asking:

What is important enough for me to be

involved in this election?

For me to invest in this effort?

Now, this is a kind of a thing in the United States

because we ask a lot of our voters,

direct democracy means that a lot of work.


So, voting in elections in the US is kind of different.

I want to give you some examples:

So, this may look familiar.

This is the Brexit question.

One question, two choices,

the Catalonia referendum from 2015.

Stay or go out.

This is the UK ballot from 2017,

where you are choosing a party that you want to run the country.

I could give you lots of examples.

I will not show them all to you today!


Let's look at how things work in the US.

In 2016, the world was pretty much aware that

there was a presidential election going on in the US.

But in San Francisco, California, where this is from,

there were also elections for US senator,

Congress person, state senator, state assembly person,

a judge, board of education, college board,

director of the bay area transport.

I don't know why elected and the board of supervisors,

only half, I think there were 11.

It was either five or six that year.

But wait, there is more.


There were also more than 40 questions on the ballot.

And this is just a check-list.

This is not the actual ballot.

This is what ballots look like in the United States.

This is a typical ballot design for a system where

you are still in a bubble and push

the paper through a scanner.

That tallies the votes.

This one from this area of California is relatively short.

Each paper can be between 28 and 48cm long.

Printed on both sides.

So, now you can see that this is actually

a pretty complex problem.

There are a lot of elements, a lot of stakeholders

a lot of constraints.

But this is exactly the kind of beautiful,

wicked challenge that my team has been working on

for a long time with election officials as our partners.


So, we mapped this experience out.

There are going to be two views.

One for privileged voters and one for burdened voters.

So, here is what voting in America looks like

for a privileged voter.

Across the top is the institutional process,

the one that election administrators

think everybody goes through.

And those of us who are kind of geeky

about elections see as the process.


Below that I'm going to show you the voter’s actual path.

Like I said across all of our research the

common question people start with is: 

What is on the ballot?

So, great, I decide, as a voter - I am in,

there are things I care enough about to invest.

But what are my options for taking part?

The privileged voter gets a voter guide mailed.

It includes a ballot to mark.

He can drop it in a mail box,

a drop box or take it to a polling place.

But he actually loves the idea of voting in a polling place.

He has been voting there for decades.

It has been the same location all that time.

It is close to home, easy for him to find.

Basically, zooms through steps 2 and 3.

Next is how do I mark the ballot?

The voting system for better or worse in his town

has not changed for 15 years and he is an avid voter,

so he feels confident he knows how to mark his ballot.

The voter guide reminded him what to do and how to do it.

It is typically about here where voters start to

wonder whether their voter registration is up-to-date.

Our privileged voter was automatically registered.

So, he didn't have to think about a thing.

Now, step 5, because he got his ballot in the mail,

with his voter guide he was able to mark at it

at home in his own time using the internet.

But this guy took himself to the polling place,

marked up his ballot, boom he is down.

Step 6.

He goes online to his county website which is excellent

and gets to watch the results coming in.

This is about as frictionless as it gets in the United States,

even though there are lots of steps but

this is not the experience, that most voters in the US.


Most voters experience voting as a test,

in terms of what they have to know about the issues,

who is running, can you say 24 presidential candidates

for the Democratic Party and just to get through

the gauntlet to get a ballot in their hands.

Some of the steps are well-intentioned.

They want to make sure only people who are legal

and eligible get to vote.

Giving lots of options for taking part makes sense too.

How hard can be to mark up a ballot.

What I would like to show you is how most people

in the United States experience it.

If you are a designer in an organisation

you may recognise this process in your own design,

in your own services.

Even if you don't work for Government.

If you work for a bank or an insurance company,

for example, or biotech company, whatever,

are you doing these kinds of things to your customers?

Think about that.

So, again, the Government's chronological process

across the top, with the voter path below,

this time I want to look at the burdened voter.

Let's say this is Bill.

Registered to vote.

What is the rest of the experience like?

He again, even though he is super psyched to vote,

makes a decision on whether to vote based on

what is on the ballot of the now Bill,

in addition to his other life experience,

doesn't get a voter guide.

 He doesn't live in a place where one comes to him.


 A voter guide would be in plain language and

would include all the information about

everything that is on the ballot,

 the candidates and questions,

 along with how to carry out his franchise.

 That would be helpful.

 But, as of about 2014 when we last looked at this data,

only 9% of the 5,000-voting jurisdiction sent out

information ahead of elections and

trust-worthy information is really hard to find online,

outside of the US and websites.


So, what is on the ballot?

Now deciding how to take part is tough,

because no information is coming to him about

early voting or voting by mail.

And he missed all the deadlines to get to do that.

He's just moved to this town.

He has never voted at his polling place;

it is far from where he works.

He goes to where he thinks it might be,

but it is not clear that that is where he should end up.

Now, Bill also needs photo ID which he has,

hooray, but that was a process, even though,

by law the card itself is free, getting it is not.

He had to particular time off work to

go to the Department of Motor Vehicles.

He had to wait for hours.

He had to take transit to get there.

In Texas, on average, the closest DMV.

Department of Motor Vehicles, is 25 miles away,

where you can get this.

This is a whole day expedition.

So, not free.

Note that we are, just now,

step 5 in the place where election officials think

you should be when you start this process.

Our voter had registered to vote but he has moved,

so he had to update his voter registration

to his new address in a new county.

Now, he can't do this online.

He has to go to a website, download a PDF.

Print it, fill it out, sign it, fold it up,

put it in an envelope, buy the stamp,

put the stamp on it and mail it in.

Any one of those steps are steps that could

mean somebody drops out.


Now, he has had no chance to do any homework

because it is just hard to find the vital information.

He doesn't know who is already in office.

He doesn't know what is going to be on the ballot and,

and when he gets to the voting booth

all of this will be there.

He skips these steps of learning what is on the ballot

and how to mark it because he has

run out of time and run out of resources.

Finally, he gets to the polling place on

election day and he is gobsmacked by what is on the ballot.

Tonnes of things he didn't know anything about before.

There were contests and candidates

he had not heard anything about until this moment.

And he is super confused about how to mark the ballot,

he has never seen a system for ballot like this one before.

He leaves hoping he has got it right.

I know people with Masters degrees in political science

who are unsure if they got it right.

And he hopes that it was all worth the work.

He is really interested in knowing the result,

but his county doesn't have a very good website,

so he doesn't know any other source to get them.

There is no local newspaper any more.

So, I have to ask: 

In the spirit of making some things easier,

are we actually making other things hard?

To take the example of early voting,

this seems like a really great idea but

you have to learn where it is,

what days it is open, what hours it is available.

It's not at the usual polling place and

it is not always at the town clerk's office.

It is not simple and straightforward to do that.


So behavioural economics playing out at every step.

This is, by the way,

what is happening in everything we design.

People are making decisions about

whether to move on, take the next step or

drop out and try to do something else.

Add to this the buzz of the ecosystem

in an overall electoral system,

where our presidential campaign goes

on for about two years and the whole thing

can feel pretty overwhelming and just be exhausting.

And so, this is the context that I want you to know about.

Because, there is this notion that people who

don't vote don't care and that they are apathetic.


But the thing that we have come to realise

through our research is that,

there might not be such a thing as voter apathy.

We believe that Americans really do care about voting.

Almost all of them are really, really excited about it.

But for some, it's harder and

for some the trade-off at each step means dealing

with the problems of the now: 

Do I have to take time off work?

Do I have to make sure I get my kids out of day care?

At the expense of weighing in about an unknowable future.


Now, my team never asked the question about

whether people care about voting.

We asked - what obstacles do people face

in casting a vote the way they intend?

And what we learned is that people try really,

really, really hard to vote.

In fact, there is so much shame attached

to not voting that people lie about it.

We know this from survey results.

So, what we did see is that a system,

in trying to make things more convenient and

easier for voters, actually it can make it harder.

Because there are a lot more decisions to make.


And this kind of thing is true across government,

in my experience, whether you are trying to

vote or applying for food assistance or

buying health insurance or using the health system,

paying your taxes or trying to get a business licence.

If you think about the private sector,

let's look at how hard it is to get the

phone company or cable company to do

what you want when you need them to do it.

There are many more steps to these

processes for users than you think.

As a product team or a team delivering a service,

there are usually more steps than you ever realise.

So how the organisation thinks about

the process and activity is way different

from how the people going through it experience it.


This is true of your products and services, too.

Users make rational trade-offs at every step.

The team at the Center for Civic Design could not

have laid out a research plan to reach this point.

Now the map that I showed you,

the journey map I showed you,

it's just an artefact, it is not the product.

We don't do this work in the service of

making things like that.

We use the artefact to track where we are

in how to understand the problem space.


We did the research to begin to understand the larger system.

This is what the main research agenda looked like.

Not only did we hear about a thousand stories

in five or six years, it took a lot of people to collect those stories,

design is a team sport after all.

The first one we did needed 30 researchers and 40 participants.

The next one took 100 interviews by my co‑director

and I, that we did on the street in various cities.

The next one gave us more than 35 voters

with low literacy and then worked with hundreds

of workers and their bosses across the country and

2016 my team tested messaging with about 40

participants and did a diary study with another 50 people.

Then they worked with a couple of dozen

county election officials to implement some of

the things we had learned to field test in pilot testings.


In 2018 we repeated parts of these studies to

further our understanding,

had things changed, had voters changed.

Had voters changed how they think and behave and

how they inform themselves.

What new burdens were they encountering,

or were they the same.

We couldn't possibly have done this alone.

We had a lot of partners,

a bunch of volunteers here and there,

some grad students, a lot of help.

And we involved a bunch of different methods to do that.

Because it turns out our problems are hard

and this is why I love working in civic design.

There's plenty of work to do.


UX design is getting a lot harder and way more nuanced

than it was when I started doing this work a long time ago.

It's beyond what any of us has had to think about

in terms of UI just for a "digital experience" of

the now we have to think about culture and context

that things are being used in and the

privacy perimeters and the business and political constraints.

We had that first question,

what questions do voters have about elections

and you know that if you do research finding

answers to questions leads to more questions

and as you peel away the relatively easy problems,

you reveal several issues and harder problems

and deeper complexities and we came away

from doing all of that with what we think

of as pretty major insights along the way.


First, while our findings are about voting in elections,

they are about civic engagement generally

I think and I think back to whatever you are all working on.

People want to know what's actually happening

in the election and how it will affect their lives,

this is the deciding point.

They want to know whether this activity is worth

investing in, what will be different,

because these people got elected or that blog is passed.


People want to vote but if they can't relate to

what the outcomes are going to be,

how their daily lives or the lives of people

who they are close to are going to change,

they are not going to show up.

If they don't know what to do,

they are not going to show up.

We talk a lot about getting people to the polling place,

a lot of people don't know what happens after that.


So, people who are new to a process

don't even know where to start asking questions.

It feels like an exclusive club and they are not part of it.

They will have to take a test to get in.

Again, we hear this over and over.


Think about somebody learning about investing

for the first time as a parallel or

somebody who has been diagnosed

with some serious health issue.

Those people have to become experts

and that is what we are asking of voters as well.

Even on apparently simple questions.

People drop out because every on stacking

is cumulative, it is not as if you get through

one gate and all of the burden of that is removed,

what happens is you carry some of that to the next,

and then the next, and the next.

It diminishes the investment people are willing to make.

It is exhausting.

The burden is great.


In the US, and this is true in Canada and

the UK as well, nearly half of adults read

at or below grade 6 level.

This is often paired with low digital literacy.

So, we learned to always include people

with low literacy in our studies,

in fact in some cases we only have participants

who have low literacy.

That was a monster revelation to me.

I have been working for decades writing

what I thought was plain language and

we got people into a usability study situation

and they just kicked our butts.

Learned a tonne.

Voters are frustrated in their attempts

to separate the signal from the noise.


2016 showed us this in ways he hadn't heard or seen before.

There was more noise than ever and it's not going to get easier,

it's only going to get harder and

this set of issues in social media and

regular broadcast media is going to continue.

This erodes trust in elections,

it erodes trust in government generally.

If users, if voters can't tell what's important,

if they can't understand the messages they are getting,

they are just going to opt out.

They will stop trusting the system.

They will exhaust their willingness to invest

in the greater good and they are going to trade

that off for what they can control, right now.

For future, that they can't predict.


I think that sounds like a lot of the questions

that we have dealt with not just in the US,

but in the UK and elsewhere over the last few years.

So, everything that we put out there has to

be useful and useable accessible, clear,

plain and relatable.

And I think that's the hard part.

So, taken together I think those tell quite a story.

All along the way though, all the stories of the users,

of the voters, show just how

hard it is for people to vote.

There are dozens of hurdles.


One of the things we have learned that there is still a lot we don't know.

These are just the ones that we do know about.

Others for people in the military or citizens

who live abroad, there are more for people

who live in under‑served areas,

there are more for people who have

limited English proficiency.

And imagine if you have a combination

of those issues and you have disabilities.

So now we have a research agenda for the next ten years!

I am going to tie this up altogether.


People want to know whether the activity is worth

their investment in time and energy,

this is true not only for voting,

not only for civic engagement but

anything that we put out in the world.

They worry that they don't know enough to

make a really good decision.

The burden adds up across the experience.

And it's difficult, if not impossible in some cases,

to connect campaign promises,

this sounds like advertising and marketing too,

to real‑life and how my life will be different,

how my life will be better because there's

a trade deal or a trade war.

So, there are things that you can actually do

and I encourage you to do them.

If you want to change how things work for

people here are steps you can take.

First, register a vote, make sure your voter registration

is up‑to‑date and help somebody else get registered to vote.

Second, sign up to be a returning officer,

there are never enough of them and it's amazing field research!

Sign up to be an official observer,

there are instructions on websites, it's not hard to do.


When there are important issues or

confusing events going on among candidates,

say you have ten candidates running for a party,

you can hold a neighbourhood meet‑up to

have a discussion about who the candidates are,

what their issues are and where they stand on them.

One of the things that we have heard about

in places that have all vote by mail,

like Washington, is that people hold voting parties.

These are not parties for voting and they are

apparently is no coercion going on,

but when you get a ballot like the San Francisco one,

people split up the assignments across the questions

and issues and then they come back together

at a dinner party a week or two later to

share what they have learned.


Next, probably most important,

is make a plan for yourself to get to the polling place

and cast a ballot.

And if you can help somebody else

make a plan and get them to the polling place,

that will make a huge difference.

Finally, at least in the US we know from anecdotal

evidence from stories that we hear from participants

in our studies that the most engaged folks

were taken to the polling place as children by an adult.

So, take a kid with you to the polling place.

Rent one if you have to!

 If you want to change how the system works,

how government works,

how the company you work for changes and works,

you have to be present.

Decisions get made by the people who show up.

Thank you.