The Person on the Other Side of the Screen

What A Web for Everyone taught me about accessibility as a designer

A person opens something we designed because they want to do something. Send a message. Complete a task. Learn something. Buy a ticket. Watch a video. Get through their workday.

Then sometimes our design gets in the way. A control they can’t operate. A video they can’t understand. Text they can’t comfortably read. A form that doesn’t give them enough time. An instruction that only makes sense if they can see the colour we’re referring to.

Nothing about that person’s goal has changed.

We introduced the barrier.

That’s the idea from Sarah Horton and Whitney Quesenbery’s A Web for Everyone that has stayed with me most. It changed accessibility from something I thought about as accommodating a person’s needs to something that also asks me to examine the environment I’ve helped create.

As a designer, that changes the question. Instead of asking, How do we accommodate someone who can’t use this?, I find myself asking:
How can we make this work for more people from the start?

It sounds like a small shift. It isn’t.

1. People First: Design for different ways of experiencing

Imagine two people opening the same laptop. One scans the screen, recognises the hierarchy, finds an icon and gets to work. Another experiences that same interface through words and sound, relying on information about where they are, what they’re focused on and what they can do next.

Same computer. Same goal. A completely different way into the experience.

That reminds me of something that’s easy to forget when I’m staring at a Figma my design tool canvas: the interface I see isn’t necessarily the interface.

I design from what I know. I can see the hierarchy, so it feels obvious. I understand the terminology because I’ve been immersed in it. I recognise an unfamiliar icon because I helped choose it. My own experience is an essential design tool, but it’s also a blind spot.

For example, Windows Narrator is a built-in screen reader that reads information on the screen aloud. People can customise aspects of the spoken experience, such as speed and volume, and use it for activities including reading and browsing.

What interests me isn’t simply that a screen reader exists. It’s what has to be true of the interface for that experience to work well. A button can’t rely only on its shape or position to communicate what it does. The interface has to carry enough meaning beyond its pixels for someone to understand where they are, what something is and what they can do with it.

That’s what People First means to me. There isn’t one standard human being waiting on the other side of the screen. People arrive with different abilities, knowledge, preferences, technologies and ways of experiencing the world.

I can learn accessibility standards and imagine different scenarios, but I can’t fully imagine my way into someone else’s lived experience. Sometimes the most useful thing I can do is stop saying, “I think this will work,” and ask someone to show me.

I ask myself: Who else could I learn from before I decide this works?

2. Clear Purpose: Good design asks less of people

Imagine opening a page because you want to read. Around the content are options to share, review, comment, export, translate and print. Every one of them might be useful, but not right now. Right now, you just want to read.

Then the interface grows quieter. The content gets more space. Your attention no longer has to negotiate with the interface before reaching what you came for. Nothing important was added. The design simply remembered your goal.

An example that comes to mind is Reading experiences in Microsoft Edge. Microsoft Edge provides reading and accessibility capabilities that support different ways of engaging with text, including Read aloud and text-related customisation.

What I like about this example is the focus. Making something easier to use isn’t always about adding another accessibility control. Sometimes it’s about removing the things competing with the person’s goal.

As designers, we value simplicity. But even good ideas can start to pile up. One more feature, one more card, one more action, one more clever interaction. Eventually, the person has to understand our product before they can accomplish what they came to do.

Clear Purpose has made me see restraint differently. There is empathy in making an experience quieter.

I ask myself: What can I simplify so the person can give more of their attention to their goal?

3. Solid Structure: Design what it means, not only how it looks

When we design a heading, we want people to recognise it for what it is. So we make it bigger, give it more weight, add some space, maybe a different colour. Eventually, nobody could mistake it for anything else.

Visually.

But there’s another question underneath that: Is it actually a heading?

A real example is Document structure in Microsoft Word. In Word, I can make a line of text 24-point, bold and blue, and it will certainly look like a heading. Or I can identify that text as an actual heading in the document’s structure.

Those two versions may look almost identical, but the second carries meaning that technology can interpret rather than simply display. That distinction matters when someone isn’t experiencing the document purely through its visual presentation.

The same idea applies to more complex information. Imagine a chart in Microsoft Excel where colour, position and height make the story immediately clear: one product overtakes another in July and continues to pull ahead.

If the alt text simply says, “Bar chart showing sales over time,” someone who can’t see the chart gets the object, but not the insight. A more useful description carries the meaning with it: “Sales for Product A surpassed Product B in July and continued to increase through the rest of the year.”

The goal isn’t simply to describe what something looks like. It’s to make sure its meaning isn’t lost when the visual isn’t available.

I find that idea surprisingly powerful because interfaces are full of invisible relationships. Proximity suggests which things belong together. Size suggests importance. Position suggests sequence. A card suggests that several pieces of information belong to the same idea.

Solid Structure makes me think about whether those relationships exist underneath the visual treatment too. I’m not only designing what something looks like. I’m helping define what it is and what it means.

It also makes accessibility feel like something that belongs at the beginning of the work, not something we attach once our screens are finished. When structure is considered early, accessibility becomes part of the creative process rather than something we come back to afterwards.

I ask myself: How can the underlying structure reinforce the meaning I’m communicating visually?

4. Easy Interaction: Let people find their own way through

There’s a button on the screen. You can see it, you know what it does, and it’s exactly what you need. But you can’t reach it.

Maybe you’re navigating without a mouse. Focus moves from one control to another and then skips straight past the action you need. Someone looking at the screen could say, “But it’s right there.”

For you, it isn’t.

A feature can be right there on the screen and still be out of reach.

Microsoft Teams supports multiple ways of interacting and participating, including keyboard shortcuts, screen-reader support, voice typing, meeting controls and hand raising.

That shifts the question for me from Does the feature exist? to Can someone actually use it?

Clicking, hovering, dragging and swiping can feel universal when they’re the ways we interact every day. They aren’t. People also need room to hesitate, take longer, make mistakes, change their minds and recover without losing their work.

This is where Easy Interaction starts to feel deeply connected to agency for me. The ability to choose, explore, recover and accomplish something independently isn’t a nice addition to an accessible experience. It’s part of feeling in control of it.

Sometimes the most respectful interaction isn’t one perfect way forward. It’s more than one good way forward.

I ask myself: How can I give people enough choice and control to interact in a way that works for them?

5. Helpful Wayfinding: Predictability helps people feel oriented

Imagine stepping out of a lift in a hotel you’ve never visited. There are three corridors and no signs. The doors work, the lighting is excellent and the handles are easy to use. Nothing is broken.

You just have no idea where your room is.

A digital experience can be exactly like that. Every control can work and someone can still feel completely lost, because being able to move through an experience isn’t the same as understanding where you are within it.

A real example is Outlook’s information hierarchy. Think about the familiar mental model of an email product such as Outlook: Account → Folder or mailbox → Message list → Individual message.

That hierarchy turns a huge amount of information into a space we can mentally navigate. When I’m reading a message, I have some understanding of the larger place it belongs to and how I might get back there.

That’s what connects Outlook, for me, to the hotel with the unsigned corridors. The problem in the hotel wasn’t that the doors didn’t work. The problem was that I couldn’t build a map in my head.

Helpful Wayfinding has made me appreciate predictability differently. Not every part of an experience needs to be new. Consistency can quietly say, You already know how this works. You can find your way back.

I ask myself: What familiar cues can I provide so people feel oriented and confident about where to go next?

6. Clean Presentation: Design systems that bend

Imagine a dashboard with twelve colours, six font sizes, four icon styles and three kinds of cards. Almost everything looks important, so nothing really stands out.

Quiet some of it down and suddenly there’s somewhere for your attention to land. The heading becomes clearer. Related information feels related. Important states stand apart.

Then change the conditions. Enlarge the text. Zoom the page. Change the contrast. Use a smaller viewport. The design should still make sense, even when someone experiences it differently.

Windows provides contrast themes that use more distinct colours to help make text and applications easier to see.

As a designer, what interests me is what happens to my carefully composed interface when someone needs to experience it differently. If I relied on subtle shades of grey to separate primary from secondary information, or on a faint colour change to communicate state, does that meaning survive?

Colour and hierarchy aren’t decoration sitting on top of an experience. They participate in communicating it. But the information shouldn’t become meaningless because the presentation changes.

That’s how I now think about Clean Presentation. Not minimalism, but clarity that can adapt. We design the default. The person gets to make the experience work for them.

I ask myself: How can the visual design communicate clearly while still giving the person room to adapt it?

7. Plain Language: Make the product sound like it’s trying to help

Something goes wrong and an error appears: “The operation could not be completed due to an invalid configuration.”

The computer seems to know what happened. The people who built it probably know what happened. But the one person who needs an answer is still left guessing.

Now imagine the message simply tells you what happened and what you can do next. The technical problem hasn’t changed, but your relationship with the product has. One message reports a problem. The other helps you move forward.

Microsoft Editor supports people with aspects of language such as spelling, grammar and clarity. Here, comprehension is part of usability.

The same is true inside an interface. Buttons, headings, instructions and errorsare all part of the experience.

People also encounter our words while distracted, tired, rushed, learning or working in a language that may not be their strongest. Clear language doesn’t underestimate them. It respects their time and attention.

That’s what Plain Language has come to mean to me. Not simply using shorter words, but reducing the unnecessary distance between what the product knows and what the person understands.

I ask myself: What could I say more clearly so the person understands what’s happening and feels confident about what to do next?

8. Accessible Media: Protect the meaning, not the medium

Eleven people are in a meeting. Someone says, “Let’s move the launch to Wednesday.” Ten people hear it. One doesn’t.

Maybe hearing the conversation isn’t an option for them. Maybe they’re in a noisy environment. Maybe their audio simply isn’t working. Different circumstances, same problem: important information existed only as sound.

In Live captions in Microsoft Teams, the conversation also appears as captions. The information hasn’t changed. What changed is the number of ways it can reach someone.

That’s what I love about captions as a design example. They aren’t useful in only one narrowly defined situation. Different people, in very different circumstances, can arrive at the same need.

The idea extends beyond captions. Colour, imagery, sound and movement are powerful design tools, and accessibility doesn’t mean stripping that richness out of an experience. It means not trapping meaning inside one way of experiencing it.

If colour disappears, can I still understand the state? If sound disappears, can I still understand the message? If someone doesn’t experience the image, can the important information still reach them?

The job isn’t to protect the medium. It’s to protect the meaning.

I ask myself: How can I preserve this meaning across different ways of seeing, hearing and experiencing it?

9. Universal Usability: Aim higher than “usable”

Imagine arriving at a beautiful new building. Straight ahead is the main entrance: wide stairs, glass doors, everything drawing you towards it. Then there’s a sign pointing around the corner: Accessible entrance →

Technically, everyone can enter. But the architecture has already said something.

One entrance says, We expected you.

The other can feel like, We thought about you afterwards.

That’s a difference I don’t want to lose sight of in digital design.

The Xbox Adaptive Controller is designed to support adaptive gaming setups for players with limited mobility, including configurations that can work with external switches, buttons and other controls.

What interests me isn’t one piece of hardware. It’s the assumption underneath the system: there doesn’t have to be one “proper” physical configuration through which everyone participates.

Different ways of interacting can still belong to the same experience.

To me, Universal Usability doesn’t mean making everybody interact in the same way. It means making flexibility part of the idea from the beginning, so different ways of interacting are part of the experience, not exceptions to it.

A person can technically succeed and still be exhausted by the experience. They can complete a task while feeling that every choice was designed with somebody else in mind.

I ask myself: What would help more people feel that this experience expected them from the start?

Design for the person who actually shows up

Every interface contains assumptions about how someone sees, hears, moves, reads, understands, concentrates and interacts. It contains assumptions about what technology they use, how much time they have and what they already know.

What I’m taking from A Web for Everyone is not another checklist to apply at the end, but a different way of looking at the work while I’m doing it.

Standards matter. Reviews matter. Automated checks matter. But they can’t completely tell me what an experience feels like. A checker can tell me alternative text exists; a person can tell me whether it helped. A test can tell me something receives keyboard focus; a person can show me what it’s like to actually get there.

And sometimes a specification can tell me everything is correct while a person is still saying:
“I can’t do this.”

My job in that moment isn’t to defend the design. It’s to listen.

So before I call an interaction intuitive, I want to stay curious about who it feels intuitive to. Before I consider a design finished, I want to come back to the simplest question:
Does this design help the person on the other side of the screen do what they came here to do?

There are people, arriving with different abilities, contexts, technologies and ways of experiencing the world. Our designs need to make room for them.

Have you read Sarah Horton and Whitney Quesenbery’s ‘A Web for Everyone’? Share your thoughts in the comments below!

The Person on the Other Side of the Screen was originally published in Bootcamp on Medium, where people are continuing the conversation by highlighting and responding to this story.