Visual Hierarchy and Composition
Podcast · Visual Hierarchy and Composition

Why the F Pattern Means Bad Design

by ZumiLabs · 52m
0:00 52m

Transcript

Made automatically from the audio, so it may contain mistakes. Click a line to play from there.

Host 1

So for years, pretty much everyone in web design and you know digital marketing, they've completely worshiped this one specific image.

Host 2

Oh yeah. The massive glowing red F.

Host 1

Exactly. Jakob Nielsen's famous F-shaped eye tracking heat map. I mean, if you have ever touched web design, you know exactly what I am talking about.

Host 2

You literally can't escape it in design school.

Host 1

Right. It looks like like a thermal scan of a house losing heat in the winter or something. Just bright crimson up at the top left, bleeding into this yellow and then fading into a cold dark blue on the right side of the screen.

Host 2

Yeah, the absolute gospel in the industry has always been, you know, this is just how humans naturally read on the internet.

Host 1

That we are biologically wired to read in an F shape. So we just have to design our websites to accommodate it.

Host 2

Which is I mean, it's just wild how ingrained that is.

Host 1

It really is. But what if I told you that the F shape isn't a natural human behavior at all? Like what if it's actually a symptom of a fundamentally broken design?

Host 2

It completely flips the script on how we understand user behavior, doesn't it?

Host 1

It totally does. And welcome to this deep dive, custom tailored just for you. Today we are jumping into season nine, episode one, which is all about visual hierarchy and composition.

Host 2

Right. Getting right into the invisible mechanics of the web.

Host 1

Exactly. And just to set the stage, if you remember our last deep dive, episode zero, seeing like a designer, we looked at those six levers behind a fifty millisecond first impression.

Host 2

Yeah, we covered the aesthetic usability effect, Jacob's Law, and we diagnosed all the massive baseline flaws in that fictional app we were fixing up, Canopy.

Host 1

Oh man, poor canopy.

Host 2

It is such a mess right now.

Host 1

It really is.

Host 2

So we are not gonna reintroduce that whole series map today. You already know the prerequisites.

Host 1

Today our mission is taking a massive stack of sources, stuff from the Nielsen Norman group, a list of parts, Smashing Magazine, the laws of UX, and we are unlocking exactly why your eye goes where it goes on a screen.

Host 2

And we're gonna hold off entirely on things like hue or you know, OKLCH lightness math or font classifications. That stuff belongs in episodes two through four.

Host 1

Right, because today we are establishing a strict audio first rule.

Host 2

Yes.

Host 1

We are going to explore deeply visual concepts using vivid physical metaphors like seesaws in grocery aisles.

Host 2

Exactly. And we will absolutely never read CSS selectors or syntax aloud.

Host 1

Nobody wants to hear us read Flexbox properties on a deep dive.

Host 2

I mean, I love CSS, but no. We are focusing purely on the psychology and the strategy.

Host 1

So let's go back to that F pattern for a second. Why is it an indictment of bad design?

Host 2

Well, the F pattern is not some inherent biological imperative. Like when you see an F-shaped scanning pattern in a user testing session, that heat map is actually screaming at you that your layout has left the user completely blind to the entire right half of your page.

Host 1

Oh wow. Blind to half the page.

Host 2

Yeah, users only resort to that F shape when they are confronted with this massive, intimidating wall of text and absolutely zero visual guidance. They are lost.

Host 1

So the F pattern is just their desperate attempt to navigate in the dark.

Host 2

Precisely. It comes down to cognitive load and our evolutionary wiring. I mean, human beings are fundamentally programmed to conserve energy, right?

Host 1

Right. The brain uses a ton of calories.

Host 2

Exactly. So when you land on a web page that lacks headings, lacks bold text, lacks bullet points, basically there are no structural cues. Your brain instantly calculates that reading every single word is gonna cost way too much energy.

Host 1

So it chooses the path of least resistance. It goes into survival mode.

Host 2

It absolutely does. You read the first line horizontally across the top, just trying to get the gist.

Host 1

Which creates the top bar of the F.

Host 2

Right. Then you drop down a little bit and read a slightly shorter line, hoping for a secondary clue. That's the middle bar.

Host 1

And then they just give up.

Host 2

Yeah. Your brain just gives up on reading entirely. You just drag your eyes straight down the left margin, scanning purely for a recognizable word or shape. That creates the long stem of the F.

Host 1

That is such a wild revelation. I mean, we've essentially been looking at a symptom of user exhaustion and treating it as a best practice.

Host 2

It's crazy.

Host 1

It's like watching people limp because their shoes are too tight and then concluding, oh well, humans naturally prefer to walk with a limp. Let's build more tight shoes.

Host 2

That is a perfect analogy. You are doing the absolute bare minimum to evaluate if the page is even worth your time.

Host 1

So if the F-shape is just users begging for structure, our job is obviously to provide that structure so they don't struggle.

Host 2

Right. We want them scanning efficiently.

Host 1

But to establish that perfect hierarchy, the sources suggest something that feels incredibly counterintuitive to me. They say we have to turn the color off completely. We have to design in pure grayscale.

Host 2

Designing in the dark.

Host 1

It strips away one of our biggest crutches.

Host 2

But color is the fun part.

Host 1

Color is like branding, it's emotion, it's how we make the buy now button pop. Why is color the enemy of structure?

Host 2

I know. Designers hate hearing this. But color can easily mask severe structural flaws in a layout. There's this brilliant piece from a list apart called Contrast is King.

Host 1

Okay. I love a list apart. What do they say?

Host 2

They bring up this fascinating observation about childhood development. When kids are playing with crayons and construction paper, they have this intuitive, flawless understanding of contrast.

Host 1

Oh, like they know what works and what doesn't.

Host 2

Exactly. A five-year-old knows instinctively that a yellow crayon is not going to show up on a piece of white paper. They don't even try it.

Host 1

Right. They immediately grab the dark blue or the black crayon.

Host 2

Yeah. They understand that for a shape to be seen, it must possess luminance contrast against its background.

Host 1

But then we grow up, we become digital designers, and we get a palette of 16 million hex codes.

Host 2

And we get totally seduced by hues.

Host 1

We start thinking, well, the background is white, but this button is neon green, so of course people will see it.

Host 2

And that is a fatal trap. If your design only works because your button is a bright, screaming red, your design is structurally weak.

Host 1

Because you are relying entirely on chrominence, the color itself, rather than luminance, which is the actual lightness or darkness.

Host 2

Right. And that becomes a massive accessibility issue immediately.

Host 1

Oh, for sure. If we rely on the hue of that red button to tell the user this is the most important thing, what happens when a user simply cannot perceive that specific hue?

Host 2

The whole hierarchy just collapses. The sources emphasize how critical it is to design for the full spectrum of human vision, which includes colorblindness.

Host 1

Right, like deuteronomy.

Host 2

Exactly. Red-green colorblindness, which is incredibly common. To someone with deuteronomy, that screaming red button and the bright green background might both look like the exact same muddy shade of brownish yellow.

Host 1

So if they share the same underlying lightness, the button is practically invisible.

Host 2

Yes.

Host 1

Let's push that analogy. It's like designing a fire alarm for a building, but the alarm only flashes a specific narrow wavelength of pink light and doesn't make any sound.

Host 2

Right.

Host 1

If a person in that building can't perceive that pink light or just isn't looking directly at it, the alarm is useless. The building burns down. A functional alarm needs the strobe light, but it also needs the piercing siren. It needs multiple structural ways to communicate danger.

Host 2

That is a great way to put it. And that is exactly why the sources strongly advocate for forcing your design environment into pure grayscale during those early stages.

Host 1

How do you even do that practically?

Host 2

Well, they mention using web tools like GreyBit to strip the color out of life sites or operating system level tools like Nocturne or Sim Daltonism.

Host 1

Oh, to force the whole monitor into grayscale.

Host 2

Yeah. Because when you remove the hue, your brain is no longer distracted by how pretty the brand colors look. You are left staring at the raw bones of the design.

Host 1

The structural skeleton.

Host 2

Exactly. The text actually points to the Interact website as a gold standard example of doing this right.

Host 1

Oh, the Interact site, yeah. If you visit that in full color, it looks great.

Host 2

Yeah, you hover over a link in the nav, the text changes color to let you know it's clickable. But the brilliance is revealed when you switch your monitor to grayscale.

Host 1

The hover state still works flawlessly.

Host 2

Because they didn't just rely on a color shift. They engineered the contrast structurally. When you hover over an interact link, the text might change color, but it also gains a thick, dark underline.

Host 1

And the actual shape of the element shifts slightly, right?

Host 2

Yeah. They layered a structural change underneath the chromatic change.

Host 1

So bringing this back to canopy, the app we're trying to fix today. Our first step is to strip away all of Canopy's brand greens and blues.

Host 2

We build a monochrome grayscale skeleton first.

Host 1

But once we are staring at this gray layout. How do we objectively test if the structural hierarchy is actually working? Like, how do I know what the user is going to look at first?

Host 2

Well, the Nielsen Norman group heavily endorses a classic design school technique for this. It's called the squint test.

Host 1

Oh, I love the squint test. It is so beautifully low tech.

Host 2

It really is.

Host 1

You literally just sit at your desk and squint your eyes until the screen gets blurry. Or if you want to be precise in Figma or whatever, you apply a five to twenty pixel blur over the whole layout.

Host 2

And the mechanics of why this works are fascinating. When you blur a design by five or ten pixels, you destroy the legibility of the microelements. You can't read the text anymore.

Host 1

You can't see the crisp edges of the photos or the delicate icons.

Host 2

Right. All that is left for your optic nerve to process are these abstract blobs of light and dark. It simulates that exact microsecond a user lands on your page before their brain parses the content.

Host 1

Let's unpack the feeling of that. Imagine walking into a massive pitch black museum exhibit.

Host 2

Okay, a dark museum.

Host 1

Right. The contrast in your design is the museum lighting system. If every single painting, every exit sign, every fire extinguisher has a massive blinding spotlight pointing directly at it, you aren't creating a hierarchy.

Host 2

No, you're just standing in a brightly lit room with no focus.

Host 1

Exactly. Everything is shouting at maximum volume, so your brain processes it has white noise. Nothing is heard. The squint test reveals exactly where your museum spotlights are currently pointing.

Host 2

That's a spot on metaphor. The Spotify mobile app is actually a perfect case study for this.

Host 1

Oh, yeah. Those sources highlight that one.

Host 2

If you take the Spotify interface and apply a mild five to ten pixel blur, a very clear structural hierarchy reveals itself. The recently played section, you know, those six little square tiles at the top.

Host 1

Yeah, they stand out as a distinct high contrast block. It guides the eye perfectly.

Host 2

Exactly. Because from a product perspective, Spotify knows that getting you to resume yesterday's playlist is the most frictionless way to keep you in the app.

Host 1

It deserved the brightest spotlight in the museum.

Host 2

But NMG notes a critical vulnerability here. If you take that same Spotify screen and push the blur further, up to 20 pixels, something dangerous can happen.

Host 1

What happens?

Host 2

Let's say one of the albums in your recently played list happens to have incredibly bright, high contrast cover art, like a heavy metal album with a stark white skull on a pitch black background.

Host 1

Oh, I see.

Host 2

At a 20-pixel blur, the sheer luminance contrast of that specific piece of album art can accidentally overpower the main navigation of the app itself.

Host 1

So the museum spotlight accidentally shifts away from the Mona Lisa and points directly at a random trash can in the corner just because the trash can is painted neon yellow.

Host 2

Exactly.

Host 1

Okay. And that brings us to the strict biological limits of human perception. Smashing magazine has some great research on visual dominance that points out a reality a lot of designers hate. Which is the human brain can realistically only discern three distinct levels of visual dominance in a single view before the hierarchy collapses into noise.

Host 2

Only three.

Host 1

I mean that feels incredibly restrictive. Think about a standard SAS dashboard or like an e-commerce product page. Right. You have the logo, the search bar, the navigation links, the hero image, the product title, the price, the reviews, the add to cart button, the secondary wish list button. I mean, we are talking about dozens of distinct elements on a single screen. How do you limit that to three levels?

Host 2

Well, you don't limit the number of elements to three. You have to strictly group the visual weight of all those dozens of elements into three distinct tiers of importance.

Host 1

Okay.

Host 2

If you try to create a fourth or fifth tier of visual weight, the user's brain simply cannot categorize the difference quickly enough. The design goes from feeling layered to just feeling cluttered.

Host 1

Okay. Let's define these three tiers. Okay. How do we map our dozens of elements into this system?

Host 2

Tier one is your dominant element. This is what shouts first.

Host 1

The brightest spotlight.

Host 2

Yes. It is the single entry point into the design for the user's eye. It must have the highest grayscale contrast, the heaviest visual weight, and it should stand completely unchallenged.

Host 1

So for a canopy, our tier one is our primary call to action, the main start free trial button in the hero section.

Host 2

Exactly. Or it might be the massive headline that explains the core value proposition, but there can only be one tier one zone.

Host 1

Got it.

Host 2

Then we move to tier two. These are the subdominant elements or focal points. This is what guides the user's scan second. These elements stand out from the background, but they clearly and immediately defer to tier one.

Host 1

This would be things like secondary buttons, the titles of our feature cards, maybe important subheadings, the stepping stones that lead the user through the narrative.

Host 2

Spot on. And finally we have tier three, the subordinate elements. This is what whispers third.

Host 1

The whispers, right.

Host 2

It is usually your body text, your tiny metadata, your footer links. It forms the foundational layer of the page. It has the lowest visual weight, the lowest contrast, and it requires the user to actively slow down and choose to engage with it.

Host 1

You know, if we borrow a concept from journalism, this sounds exactly like the inverted pyramid of writing.

Host 2

Oh, interesting. How so?

Host 1

Well, in news reporting, the absolute most critical information, the who, what, when, and where, has to go in the very first paragraph. That is your tier one.

Host 2

Right.

Host 1

Then you provide the supporting quotes and context in the middle paragraphs, your tier two. And you leave the deep background history, the fluff for the very end of the article, your tier three.

Host 2

That is a phenomenal way to conceptualize it. You are manipulating physical visual weight to perfectly match the informational weight of the content.

Host 1

And this is precisely where Canopy's baseline design is failing in our case study, right? We know that Canopy currently has a twin CTA problem in its hero section.

Host 2

Yes, the twin CTA trap.

Host 1

They have two buttons sitting right next to each other. One says start free trial, and the other says book a demo. They are identical in physical size, and more importantly, they are identical in grayscale contrast.

Host 2

They are both fighting to be the dominant element. They are both screaming at the user at the exact same volume.

Host 1

And because they are both screaming, they cancel each other out.

Host 2

Exactly. The user experiences an immediate spike in cognitive friction. Instead of their eye naturally resting on the primary action, they have to physically stop scanning, read the text inside both buttons, weigh the options logically, and make a conscious choice.

Host 1

You've forced the user to do the hard work.

Host 2

Right. So to fix Canopy in our pure grayscale environment, we have to ruthlessly demote button number two.

Host 1

We lower its contrast.

Host 2

We strip away its solid background fill and turn it into a delicate outline button, a ghost button.

Host 1

Instantly, button number one regains its status as the undisputed tier one champion. The hierarchy is restored.

Host 2

Just by making sure we strictly only have one solid primary CTA per viewport.

Host 1

Exactly. Okay, so we have stripped away all the color. We have forced our entire layout into a strict three-tier hierarchy of grayscale contrast using the squint test.

Host 2

Mm-hmm.

Host 1

We know what shouts, what guides, and what whispers.

Host 2

We have our skeleton.

Host 1

But a web page isn't just a vertical list of items floating in a void, right? It's a physical, two-dimensional space. How do we physically arrange these heavy, dark blocks and these light, delicate blocks on a screen so that the whole layout doesn't feel like a boat capsizing?

Host 2

This moves us into the mechanics of visual weight, scale, and balance. Now that we have our contrast tiers, we have to manipulate scale, the actual physical size of the elements relative to one another.

Host 1

And the sources outline a very similar rule for scale as they do for dominance tiers, don't they?

Host 2

They do. Just as you shouldn't have more than three levels of contrast, you really shouldn't use more than three distinct sizes within a given component.

Host 1

Small, medium, and large.

Host 2

Right. If you introduce a fourth, fifth, or sixth font size, you are inviting chaos. The Nielsen Norman group provides a stark contrast between two very different mobile experiences to illustrate those.

Host 1

Okay, let's hear them.

Host 2

First, they look at a camping and glamping booking site called HipCamp. When you land on their site, the text, find yourself outside, absolutely dominates the viewport. It is massive.

Host 1

It's unmistakably the tier one element based purely on its monumental scale.

Host 2

Right. It's acting like a billboard. The sheer size of the typography tells you the brand story in a fraction of a second before you even consciously read the words.

Host 1

What's the bad example?

Host 2

They contrast that with an older version of the Dropbox mobile app. In that app, there was a severe lack of scale variation in the file browsing interface. The text explaining the folder structure, the names of the files, the metadata, like the date modified. It was all rendered at almost the exact same small size.

Host 1

Oh, let me guess. Visual mud.

Host 2

Complete visual mud. Everything blurred together. Because there was no variation in scale to create stepping stones for the eye, the user could no longer intuitively scan the list.

Host 1

They were forced to stop and parse the interface line by line.

Host 2

Exactly. Actively reading every single word to figure out what was a clickable folder and what was just a text label.

Host 1

It's the exact same friction we saw with Canopy's twin buttons. And you fail to use scale to differentiate importance. You transfer the cognitive burden directly onto the user.

Host 2

Precisely. You are making them do the heavy lifting. Now once you have established your small, medium, and large elements, you have to place them on the canvas. You have to balance those heavy, dark, large elements against the light, delicate elements.

Host 1

Balance. Okay.

Host 2

Visual balance is fundamentally about distributing the total amount of visual signal evenly across an imaginary central axis.

Host 1

You know, this is where a lot of people misunderstand design. When we say balance, we usually picture a mirror image.

Host 2

Right. Symmetry.

Host 1

If I have a heavy box on the left, I need an identical heavy box on the right. But that is incredibly boring. Balance is much more like an asymmetrical seesaw on a playground.

Host 2

I love the seesaw analogy.

Host 1

Right. Because if you have a 200-pound adult sitting on one side of a seesaw, you do not need another 200-pound adult on the opposite side to balance it out. If that heavy adult sits very close to the center pivot point, the fulcrum, you can achieve perfect suspended balance by placing a 40-pound child very far out on the very edge of the opposite side.

Host 2

That maps perfectly to visual weight on a screen. A massive, dark, heavy element placed near the center of your layout can be perfectly counterbalanced by a tiny, brightly colored or high contrast element placed far out in the extreme margins.

Host 1

The tension between the heavy mass and the distant spark creates a sense of equilibrium.

Host 2

Exactly.

Host 1

And the sources categorize this into three distinct types of balance. We have symmetrical, asymmetrical, and radial. Symmetrical balance is that mirror image. It is the easiest to achieve, but it carries a very specific psychological weight.

Host 2

It's quiet.

Host 1

It's stable. It feels formal, almost static.

Host 2

The sources use a job hunting platform called The Hub as an example of symmetry. When you are looking for a job, you are looking for security. You want stability, you don't want chaos.

Host 1

Symmetrical design communicates that institutional reliability. It feels like standing in front of a bank with two identical marble pillars on either side of the door.

Host 2

But then you look at a brand like Nike, specifically their product pages, they completely abandon symmetry. They embrace severe asymmetry.

Host 1

Oh, for sure. You might land on a page and see a massive, highly detailed photograph of a basketball shoe dominating the entire left side of the screen, bleeding off the edge.

Host 2

And on the right side, there is just a stark, simple block of text and a tiny add-to-cart button floating in empty space.

Host 1

Asymmetry is inherently dynamic.

Host 2

Because the eye is constantly moving back and forth across the fulcrum, trying to resolve the visual tension between the heavy shoe and the isolated text. The layout feels like it possesses kinetic energy.

Host 1

It feels alive.

Host 2

In motion. It is significantly harder for a designer to pull off an asymmetrical layout without it looking broken, but when it works, it is incredibly engaging.

Host 1

Well, let's talk about the third type: radial balance.

Host 2

Radial balance is where everything on the canvas radiates outward from a single central point or spirals inward toward it. The eye is inevitably drawn directly into the middle of the screen.

Host 1

The sources point to the Brathwaite Watch Company's website as a prime example of this. The entire layout circles around the face of the watch in the center of the viewport. It's hypnotic.

Host 2

It really is. So when we are trying to achieve the Seesaw balance on a modern web page, we have to realize that we are actually juggling multiple Seesaws at the same time.

Host 1

Wait, multiple Seesaws.

Host 2

Yeah, the engineering insights from web.dev introduce a crucial distinction here. We must understand the difference between macro layouts and micro layouts.

Host 1

Okay, let's break that down. Conceptually without getting into the code. The macro layout is the grand architecture of the page. It's the bird's eye view. Where does the header go? Where is the main content column? Where's the sidebar?

Host 2

Right. Right. And this macro layout is entirely governed by the physical constraints of the user's device. Does this design have a giant ultra-wide desktop monitor to spread out on, or is it trapped inside a tiny narrow phone screen?

Host 1

That is the macro viewport. Right. But nestled inside that grand architecture, you have dozens of micro layouts.

Host 2

These are the individual self-contained components. Think of a product card on an e-commerce site. It has an image, a title, a price, and a button. Or a user profile badge with an avatar and a name.

Host 1

What is revolutionary about modern web engineering, specifically the concept of container queries, is that it liberates these micro layouts from the constraints of the macro page.

Host 2

Yes. How does that change your playground seesaw analogy?

Host 1

Think of it like a set of Russian nesting dolls where each doll has its own internal gravity. A microcomponent like our product card can now be programmed to manage its own internal asymmetrical balance completely independent of what the entire web page is doing.

Host 2

It doesn't care if the user is on a phone or a monitor.

Host 1

Right. It only cares about the physical dimensions of the specific invisible box it has been placed inside. If that container box gets squeezed and becomes too narrow, the micro layout detects the squeeze and automatically rearranges its own internal elements.

Host 2

Like stacking the image on top of the text instead of placing them side by side.

Host 1

Exactly. To maintain its own perfect visual balance. We essentially have dozens of tiny self-regulating Seesaws operating independently inside the giant page-wide Seesaw.

Host 2

Which is incredible. And this entirely solves the problem we diagnosed with canopy's heavy bordered feature cards. Yes. Their cards were breaking and looking awful on tablet screens because they were relying on the massive macro layout to tell them how to balance. The hierarchy survives the responsive shift.

Host 1

Okay. So we have engineered our grayscale contrast. We have established our small, medium, and large typography scales. We have perfectly balanced our macro and micro seesaws. But if you look closely at any interface, it is still just a chaotic collection of dozens, sometimes hundreds of disparate tiny elements. Little icons, lines of text, dividing lines, form fields.

Host 2

Right.

Host 1

How does the human brain instinctively know that this specific text label belongs to this specific input box while ignoring the box sitting right next to it?

Host 2

To truly understand the mechanics of visual grouping, we have to leave the modern era of web design and travel back to Germany in the year 1910.

Host 1

Oh, I love this story. We're going to talk about the gestalt psychologists.

Host 2

Yes. So it's 1910. A psychologist named Max Wertheimer is sitting at a railroad crossing waiting for a train to pass. And Wertheimer realizes something profound about the human visual system. Our brains do not process the world as a stream of isolated individual data points.

Host 1

We don't see light A turns on, light A turns off, light B turns on.

Host 2

Exactly. What Wurtheimer perceived, and what all of us perceive is a single red light physically moving back and forth between the two housings.

Host 1

The five phenomena.

Host 2

Right. Our brains aggressively hallucinate motion and connection where none actually exists. We are biologically hardwired to perceive the whole rather than the sum of the parts.

Host 1

Our brain is essentially a narrative machine. It takes chaotic, isolated sensory inputs and instantly weaves them into a simple cohesive story.

Host 2

And it does this for survival. If you are walking through a dense jungle and you see a patch of orange fur behind a tree and a black stripe behind a bush, your brain doesn't catalog them as separate abstract colors. It instantly groups them together and screams, Tiger, run.

Host 1

This survival mechanism forms the absolute bedrock of gestalt psychology, encapsulated in the law of pregnancy.

Host 2

The law of pregnance. The overarching idea that human beings will relentlessly perceive and interpret ambiguous, complex or chaotic images in the simplest, most ordered form possible.

Host 1

Because simple forms require less cognitive energy to process. The brain is always looking for the shortcut to understanding.

Host 2

And the Gestalt psychologist identified the specific subconscious mechanisms the brain uses to create these shortcuts. The first and most fundamental is figure ground organization.

Host 1

Figure ground.

Host 2

Yes. The brain must instantly separate the subject the figure from the background it sits on.

Host 1

We've all seen the classic optical illusions that play with this the famous drawing that either looks like an ornate vase in the center or it looks like two human profiles facing each other on the edges.

Host 2

You can see vase or you can see the faces but your brain absolutely refuses to let you see both simultaneously the figure and the ground are fighting for dominance.

Host 1

Or consider the NBC television network logo. It looks like a colorful fan of six teardrop shapes but the actual subject the body of the peacock is formed entirely by the negative white space, the background carved out between the colored feathers.

Host 2

Right. And in web design, if your figure and your background did not have sufficient grayscale contrast the brain's figure ground separation engine stalls out. The user literally cannot determine what they are supposed to be looking at.

Host 1

Once the brain separates the subject from the background it starts looking for patterns among the subjects that is where the law of similarity comes in.

Host 2

Oh, yes.

Host 1

Similarity. If multiple elements look the same, if they share the same shape, the same color, the same size, the brain instantly assumes they perform the same function.

Host 2

This is a rule that modern designers violate constantly, usually by accident. Think about a blog page where the designer has placed little rounded blue pill-shaped badges above the article titles to indicate the category, like technology or design.

Host 1

Oh, those look exactly like buttons.

Host 2

Yes, they share the exact visual language of a clickable CTA. The brain applies the law of similarity, groups the category badges with the actual buttons, and the user frantically clicks the badge expecting something to happen.

Host 1

When nothing happens, the user feels stupid, and then they feel frustrated. You broke the gestalt rule.

Host 2

But the most powerful tool in the entire Gestalt arsenal, especially for user experience design, is the law of proximity.

Host 1

Proximity.

Host 2

Elements that are physically placed close to one another are perceived as a related group, completely overriding similarity or shape. The Nielsen Norman group illustrates this by comparing a beautifully executed Uber sign-up form with the absolute disaster that was the 2017 United States tax form.

Host 1

Oh my gosh. The US tax form is a master class in how to induce panic through bad spatial grouping.

Host 2

It really is.

Host 1

Let's put ourselves in the shoes of users staring at that tax form. You are already anxious because it is the IRS. The financial stakes are high. You are looking at a grid of tiny text and little white boxes. You are trying to fill in the last name field for your spouse. But the spacing, the vertical distance between the rows, is completely identical across the entire page.

Host 2

The empty space between the label spouse's last name and the input box above it is physically the exact same distance as the space between the label and the input box below it.

Host 1

So your eye darts up, then darts down. The brain cannot use the law of proximity to group the text with the correct box. It is entirely ambiguous.

Host 2

The cognitive load skyrockets because you have to rely entirely on reading the tiny bureaucratic instructions rather than just trusting your spatial intuition.

Host 1

You are terrified of putting your spouse's name in the wrong box and triggering an audit. It is physically exhausting.

Host 2

Now look at the Uber sign-up form. It is the exact same type of data collection. First name, last name, phone number. But the text labels are placed incredibly tight, nestled right up against the top edge of the input boxes they describe.

Host 1

Meanwhile, there is a massive gulf of empty space between one group and the next. So if proximity is so powerful and simply moving things closer together solves the grouping problem, why do we see so many cluttered noisy interfaces? What happens when designers don't trust proximity?

Host 2

What happens when they panic?

Host 1

Yeah, what happens when they panic?

Host 2

When designers lack confidence in their spatial arrangements, they fall back on another gestalt principle, the law of common region.

Host 1

Okay, what is that?

Host 2

This dictates that elements enclosed within a shared boundary are perceived as a group. So the designer panics and draws a harsh physical border around the items.

Host 1

This is the trap of boxitis.

Host 2

Yes, boxitis.

Host 1

If you have spent any time playing around with the generic, unguided AI website builders that are flooding the market right now, you have seen a terminal case of boxitis.

Host 2

Oh, absolutely.

Host 1

They love the bento box aesthetic. They take every single text block, center the alignment, and then slap a heavy gray one-pixel border around it. They take a cluster of three icons and they draw a fence around them. The screen ends up looking like a spreadsheet.

Host 2

The inherent problem with boxitis is that every single border you draw is a new visual element that the user's brain has to process. It introduces massive visual noise. Right. And the expert sources diagnose exactly why designers fall into this trap. They draw borders because their internal spacing and their external spacing are mathematically identical.

Host 1

Let's unpack that. If the empty space inside your group of elements is 20 pixels, and the empty space separating that group from the next section is also 20 pixels, the grouping fails. Proximity is broken.

Host 2

So instead of adjusting the spacing, the designer just draws a box around it to force the grouping.

Host 1

Yes. I love the analogy you use for this, the grocery aisle.

Host 2

Oh, right.

Host 1

It is exactly like walking into a massive supermarket. When you walk in, you know the serial is an aisle four, because there is one large clear sign hanging from the ceiling above the entire aisle. You look up, you see the sign, and you walk down the aisle.

Host 2

You do not need the supermarket manager to install a flashing neon arrow pointing to every individual box of court and flakes. Exactly. Exactly.

Host 1

If every single box of cereal on the shelf had its own flashing sign, you would be blinded by the sheer volume of visual noise. You wouldn't be able to find anything because everything is demanding attention simultaneously. How do we group the serial or the UI elements without building fences around every single box?

Host 2

The cure for boxitis is the masterful application of white space. But to use it correctly, we have to understand the distinction between passive white space and active white space.

Host 1

Mark Poulton wrote a foundational piece on this for a list apart, didn't he?

Host 2

He did. Passive white space is the standard structural space that occurs naturally. The margins between paragraphs, the line height between sentences that makes text readable. It is the background hum of the page.

Host 1

And active white space?

Host 2

Active white space is space that is intentionally strategically injected into the layout to draw the eye, to create emphasis, or to force a gestalt grouping. It is asymmetric, it is deliberate emptiness.

Host 1

Bolton brings up the legendary typographer Eric Speakerman and his famous redesign of the Economist magazine to illustrate the power of space.

Host 2

It is a brilliant case study in micro white space. Think about a magazine like The Economist. It is incredibly dense, text-heavy, complex financial reporting. The pages are packed.

Host 1

When Speakerman was tasked with making the magazine easier to read, he couldn't just add massive, luxurious white margins around the articles. The publisher couldn't afford to waste the paper.

Host 2

Right. He had to solve the density problem without actually removing any content. So he went down to the atomic level. He redesigned the custom typeface itself.

Host 1

He opened up the micro white space inside the individual letters, right? Like the empty space inside the loop of a lowercase E or the bowl of a P.

Host 2

Yeah, and he slightly increased the space between the letters themselves. He lightened the visual load at a microscopic level.

Host 1

The result was that the same amount of dense, terrifying financial information suddenly felt lighter, airier, and significantly easier to read.

Host 2

So whitespace physically groups elements and improves legibility, but there is a massive, often overlooked psychological component to white space as well, isn't there?

Host 1

Oh, absolutely. Whitespace isn't just about grouping, it is an economic signal.

Host 2

It absolutely is. And this is where design intersects with pure marketing psychology. I once consulted for a client who ran a massive direct mail marketing business. They produced the junk mail catalogs that get shoved into your physical mailbox.

Host 1

Oh boy.

Host 2

They used to constantly argue with me over my designs. They would look at a layout and scream, white space is empty space. We are paying for the paper. Fill it up with bigger text and more product photos.

Host 1

Why were they so terrified of empty space? Why did they actively want the catalog to look cluttered and chaotic?

Host 2

Because they intuitively understood the psychology of brand positioning for their specific target demographic. Dense, cluttered, noisy layouts. Think of the branding for a discount supermarket like Tesco Value or a late night infomercial or a car dealership flyer. They subconsciously communicate cheap discount and bargain.

Host 1

So if a flyer looks too clean, too organized, the bargain hunting consumer assumes the product is expensive and they throw it away.

Host 2

Exactly. The clutter is the signal that a deal is to be had.

Host 1

That is fascinating.

Host 2

The noise is the feature, not the bug. Conversely, look at the other end of the spectrum. Look at the websites for luxury cosmetics, high-end fashion houses, or Apple products. They utilize massive, almost uncomfortable amounts of macro white space. You will see a tiny, elegant bottle of perfume floating in an absolute sea of stark white screen.

Host 1

That deliberate emptiness communicates exclusivity. We don't need to cram 12 features into this layout. The product speaks for itself.

Host 2

Exactly. So when we apply this to fixing canopies UI, we aren't just deleting the heavy gray borders around their feature cards to cure their boxitis. We are introducing strategic active white space to fundamentally alter the user's perception of the product's value.

Host 1

We ensure that the empty space between the feature cards is at least twice as large as the internal padding inside the cards.

Host 2

The gestalt grouping happens naturally through proximity. The interface is finally allowed to breathe.

Host 1

And instantly without changing a single word of copy or a single meature, the canopy app feels like a more premium expensive and trustworthy tool just by taking things away.

Host 2

It's powerful stuff.

Host 1

Okay. So at this point, we have built a pristine, grayscale, perfectly balanced, gestalt-approved layout. It is mathematically sound. But here is the rub. Websites are not viewed by mathematical algorithms.

Host 2

They are viewed by human beings.

Host 1

And humans are messy, distracted, and impatient. They bring their own psychological biases and physical motor limitations to the screen. When a real human hand touches the mouse or a real thumb hovers over the glass, how do they actually physically interact with this space?

Host 2

To understand the physical reality of user interaction, we move to the laws of UX. These are the documented scientific limits of human cognition and motor function. The most famous of these is Fitz's Law.

Host 1

Fitz's Law. It sounds so academic, right? But the concept is beautifully simple.

Host 2

It was originally developed in the 1950s to study human movement, but it applies perfectly to a mouse cursor or a thumb on a touch screen. Fitz's law states that the amount of time it takes a human to acquire a physical target depends entirely on two factors: the distance to the target and the size of the target.

Host 1

So the further away a button is from where my mouse currently is, and the smaller that button is, the longer it will take me to click it. And the higher the chance I will miss it entirely. Exactly.

Host 2

The practical application of Fitz's law in UI design is why your primary call to action, your tier one element, must be physically large. It needs a massive hit area.

Host 1

This is especially critical on mobile devices, where the user is relying on an imprecise, clumsy instrument, their thumb, while walking down the street. If your submit payment button is a tiny text link, you are punishing the user.

Host 2

And closely related to the physical friction of Fitz's Law is the cognitive friction described by Hicks Law.

Host 1

Hicks Law tells us that the time it takes for a human to make a decision increases logarithmically with the number and the complexity of the choices presented to them.

Host 2

The paradox of choice.

Host 1

Yes. If you give a user a navigation drop-down menu with 20 different links, their brain freezes. The cognitive load required to read, evaluate, and discard 19 wrong options, just to find the one right option is exhausting.

Host 2

But if you give them three broad categories instead, they will click instantly.

Host 1

And then we have a psychological quirk called the von Restorf effect, which is also known as the isolation effect. This principle predicts that when a human is presented with multiple similar objects, the one object that differs significantly from the rest is the one that will be remembered.

Host 2

Okay, so wait. Let's pause and synthesize this. If the von Restorff effect says make the important thing look drastically different, and Fitz's law says make the important thing physically massive, shouldn't the logical conclusion be to make our primary buy now button a colossal flashing neon green starburst that looks completely alien compared to the rest of the page?

Host 1

On paper, yes, that seems like the logical extreme. But human psychology is rarely a straight line. If you do that, you trigger a massive severe psychological backlash known as banner blindness.

Host 2

Banner blindness. This is fascinating.

Host 1

Oh right. If a used interface element is too isolated, if it is too brightly colored, or if its shape too closely mimics the visual language of a cheap advertisement, the brain's spam filter kicks in.

Host 2

The brain literally scrubs the element from the user's perception.

Host 1

The user will look directly at your massive neon green starburst button, and their optic nerve will simply refuse to process it as actionable information. They will not see it.

Host 2

So the art of design is finding the exact sweet spot. We must balance distinction with uniformity. The CTA has to be different enough to leverage the von Rustdorff effect and secure its status as the tier one dominant element, but it still has to look like it belongs to the native, trustworthy design system of the application.

Host 1

It has to look important, but it cannot look like an ad. While we are discussing how the brain filters options, we also have to factor in how human memory works when scanning lists or navigation menus. The sources highlight the serial position effect.

Host 2

The serial position effect proves that humans have a strong propensity to best remember the very first item in a series and the very last item in a series.

Host 1

The stuff in the middle gets completely lost in the limitations of our working memory.

Host 2

So, practically speaking, if I'm designing the top navigation bar for a website, I absolutely cannot put my most important actions in the center of the screen.

Host 1

It would be a fatal mistake. Your most critical anchor points, typically your company logo, which acts as the home button, and your sign-in or get started buttons must be anchored to the extreme far left and the extreme far right of the navigation bar.

Host 2

You take the boring administrative stuff, the about us, the careers page, the privacy policy, and you hide it in the middle of the list where the brain naturally glazes over.

Host 1

Precisely. So we have navigated the laws of interaction. We have optimized the sizes, the choices, and the positions. But we still need to answer the core question of human movement. We established in the very beginning that the F pattern is a symptom of a broken page.

Host 2

Right.

Host 1

When a page is actually designed well, when the structural hierarchy is sound. How do users actually read it? What do healthy scanning patterns look like?

Host 2

Well, the Nielsen Norman group, eye tracking studies identified several real, highly efficient scanning behaviors that emerge when a page is properly formatted. The gold standard is what they call the layer cake pattern.

Host 1

The layer cake pattern.

Host 2

It is exactly what you want to see. When a page has clear, distinct headings and subheadings, the user's eyes simply scan horizontally across those headings, quickly skipping over the dense paragraphs of body text below them.

Host 1

So if you look at the heap map for this behavior, it looks like a series of horizontal stripes, like a layer cake.

Host 2

It proves that your tier two focal points, your subheadings, are successfully doing their job. They are acting as stepping stones, efficiently guiding the user down the page until they find the specific section they actually care to read.

Host 1

Then there is the spotted pattern.

Host 2

A spotted pattern is much more predatory. It is when a user is actively hunting. They are skipping massive chunks of content, scanning quickly for a very specific visual shape.

Host 1

Like they might be looking for a specific set of digits like a price or a phone number.

Host 2

Or they might be hunting for capitalized words like the name of a city. Or they might be looking for the distinct underline and color shift of a hyperlink. Their eye is bouncing around the page like a pinball, locking onto those distinct shapes.

Host 1

And on mobile devices, the scanning behavior becomes distinctly physical. We see the marking pattern.

Host 2

I realize I do the marking pattern constantly on my phone without ever thinking about it.

Host 1

Right. Marking is fascinating because it integrates motor function with visual focus. It happens when a user keeps their eyes locked on one specific stationary physical location on the glass screen, usually somewhere near the upper middle.

Host 2

Rather than moving their eyes down to read the text, they use their thumb to continuously swipe and scroll the content upward, pushing the text past their stationary focal point.

Host 1

It is exactly like a ballet dancer doing pirouettes. The dancer spots a single point on the wall and whips their head around to lock onto that point during the spin. It keeps them from getting dizzy and losing their balance.

Host 2

The mobile user is spotting the screen so they can consume massive amounts of scrolling text without getting motion sickness.

Host 1

Now, while the layer cake and the spotted patterns are signs of a healthy design, the Nielsen Norman group research issues a massive warning about a very specific modern layout trend that completely destroys scamming efficiency. It is called the zigzag layout. And it is a trap that SAS companies fall into constantly.

Host 2

Oh, the zigzag layout. I see this everywhere.

Host 1

This is the classic tech landing page aesthetic. You scroll down below the hero section, and you see a block of explanatory text on the left paired with a nice illustration or photograph on the right. Then in the very next row, the layout flips.

Host 2

The illustration is on the left and the text is on the right.

Host 1

It zigzags back and forth as you scroll down the page. Visually, from a macro perspective, it looks incredibly balanced. The seesaw is perfectly level. So why is it so damaging?

Host 2

It is damaging because it completely misunderstands how the human eye gathers information. NNG ran a rigorous eye tracking study on this specific layout using the HelloFresh website. When a designer forces the text and the images to alternate left to right on every single row, they are forcing the user to constantly violently shift their starting focal anchor.

Host 1

But wait, isn't the image there to help tell the story? Doesn't the image have value?

Host 2

That is the crux of the problem. In the vast majority of these zigzag tech layouts, the images do not carry any actual informational weight. They are purely decorative filler.

Host 1

They are generic stock photos of abstract geometric shapes or a smiling person pointing at a laptop or a bowl of soup.

Host 2

Right. They don't actually explain how the software works.

Host 1

So the user looks at the first image, realizes in a millisecond that it is completely useless to them, and then what happens?

Host 2

They develop a deeply inefficient reading pattern that NNG calls the hot potato.

Host 1

The hot potato.

Host 2

The user's eye lands on the big heavy decorative image on the left side of the screen, their brain processes it, realizes it contains no data, drops it like a hot potato, and scrambles aggressively over to the right side of the screen to find the actual text.

Host 1

Then they scroll down to the next row, their eye naturally returns to the left margin to start reading, but now because of the zigzag, there is text there, so they start reading.

Host 2

Then they scroll down to the third row, their eye returns to the left margin, and boom, it's another useless image.

Host 1

They drop it like hot potato and scramble to the right again.

Host 2

It creates a stumbling, chaotic, deeply inefficient sequence of eye movements. You are exhausting the user. You are placing visual speed bumps directly in their scanning path just because you, the designer, were worried the page was looking a little too boring.

Host 1

The takeaway from that research is brutal but necessary. You must only use images if they carry genuine, heavy informational weight. If it is a detailed graph or a screenshot of the actual UI that explains a feature, keep it. But if the image is just decorative filler designed to balance the seesaw, get rid of it.

Host 2

Align your text to one side, create a crisp, predictable Swiss style tabular rail on the left, and let the user's eyes scan cleanly and effortlessly down the page in that beautiful layer kick pattern. Don't punish the user for your aesthetic insecurities.

Host 1

Exactly. When we apply this to the final piece of canopy, their wall of text editorial section, we don't try to fix it by injecting random stock photos of people in offices to create a zigzag. We fix it by breaking that massive wall of text into smaller paragraphs. We inject bold tier two subheadings to encourage layer kick scanning. We use active white space to give the text room to breathe.

Host 2

We have taken a deeply flawed interface, and without using a single drop of color, we have transformed it into something that feels intuitive, premium, and effortless to use.

Host 1

This has been an incredibly deep revelatory journey into the unseen mechanics of the digital world. We started by completely debunking the myth of the F pattern, realizing that it is not a biological destiny, but a desperate symptom of a broken, unformatted page.

Host 2

We learned that to truly fix a layout, we have to strip away the seductive crutch of color. We have to design in the dark using the squint test to build a strict, uncompromising three-tier grayscale hierarchy of what shouts, what guides, and what whispers.

Host 1

We learned how to balance that hierarchy on an asymmetrical seesaw, understanding how container queries allow micro layouts to manage their own visual weight independent of the macro screen.

Host 2

We traveled back to a railroad crossing in 1910 to learn how the brain desperately seeks simplicity through Gestalt's psychology. We cured boxitis by throwing away noisy borders and embracing the economic signaling power of active white space.

Host 1

And finally, we accounted for the messy reality of human friction. We respected Fitz's law by making our targets reachable. We respected Hicks' law by reducing choices. We threaded the needle between the Von Restroff effect and banner blindness, and we eliminated the exhausting zigzag layout to let our users scan smoothly.

Host 2

We have learned that a 50 millisecond first impression is not a happy accident. It is not magic. It is a meticulously engineered sequence of psychological triggers and spatial mathematics.

Host 1

By doing the incredibly hard, invisible work in the layout structure, we save the user from having to do the hard work of thinking. We organize the chaos so their brain doesn't have to.

Host 2

Which perfectly sets us up for next time. SO9EO2 color and contrast, where we get to finally bring hue and luminance back into the mix without breaking our brand new grayscale skeleton.

Host 1

I can't wait. But before we go, I want to leave you with a final provocative thought. We talked extensively about the law of pregnants, our brain's desperate, instinctual, evolutionary desire to simplify complex information just to conserve energy and reduce cognitive load.

Host 2

Yeah.

Host 1

In a modern world defined by a relentless, crushing avalanche of digital information overload, perhaps the ultimate luxury isn't just seeing active visual white space on a premium website. Perhaps the ultimate luxury is finding mental white space.

Host 2

That is a profound question to leave with.

Host 1

Take a hard look at your own personal digital environment today. Look at the home screen of your phone, look at your computer desktop, your browser tabs, your email inbox. Are you accidentally forcing yourself to scan through a cluttered, anxiety inducing F pattern every single time you open your laptop?

Host 2

Yeah, how can you apply that strict three level grayscale rule to your own digital life?

Host 1

What in your life truly needs to shout as a tier one priority? What needs to just guide you? And what can you finally mercifully allow to just whisper so you can reduce your own daily cognitive load and find a little bit of peace?

Host 2

That's definitely something to mull over.

Host 1

Thank you so much for joining us on this custom deep dive. Until next time, keep looking past the surface.

Save to playlist

Manage playlists →

Share