Why Do All Websites Look the Same?

We can talk about millions of design examples in the world. Each one has its own rules — rules for animation, rules for websites, rules for art. But the most important thing to know is this: they all follow some shared, common rules. And these rules do not have to come one after another in a fixed order. In my opinion, these rules are used to catch the “visual eye” of different small groups of people.

Today I want to talk about exactly this: Why do some people follow a common design that everyone already knows, while others add a very special, niche design to their art or their website and go a completely different way? What I write here looks at the big picture, but our real topic is building websites. And the whole thing turns around one single idea: pattern recognition.

Pattern Recognition

Pattern recognition, in its simplest form, means that the same shape or the same situation repeats in more than one place. Today, almost everything an AI “sees” is built on this. But here is the interesting part: even when AI was not this advanced, pattern recognition was already one of the best ways to train a machine.

When an AI looks at a picture and decides if the animal is a cat or a dog, it can do this because it has seen thousands of photos before — together with the “right answer” for each one. It learns to notice the small differences that our human eyes cannot catch. This work is more mechanical than it looks. The technical books on pattern recognition split the process into four clear steps: data collection, pre-processing, feature extraction, and classification. In simple words: the raw image is first cleaned (noise is removed, edges are made sharper), then the numbers that describe the image are pulled out (the “features”), and finally these features are placed into a group — “cat” or “dog.”

The key point is here. In this field there are two kinds of learning: supervised and unsupervised. In supervised learning, there is a “teacher” that shows the system a set of already-labelled examples. The machine is first trained with known answers, and only then is it set free. So, before it can tell a cat from a dog, someone must first show it this is a cat, this is a dog. Sounds familiar, right? We learn in exactly the same way — the only difference is that our “teacher” is a training process thousands of years long.

We do not make any special effort to notice these differences, because we already have the information. It comes from a pattern that grew inside the human brain over thousands of years. We can see this most clearly in a few human examples:

Pareidolia (seeing familiar things in random shapes): Our brain tends to see clouds, power sockets, or car headlights as human faces or animals. When there is an unclear image, the brain still “sees” a shape it already knows; it fills the empty space with a familiar pattern.

Face Recognition: We can tell thousands of different human faces apart in a moment, just by reading tiny distances between the eyes, nose, and mouth. This is one of the hardest things for machines, but for us it is as automatic as breathing.

Weather Guessing: Feeling that rain is coming from the colour of the sky, the direction of the wind, and the wet air. This is one of humanity’s oldest “reading nature” skills.

Sensing Danger: Hearing a sudden sound in the bushes, or smelling something in the air, and reading it as “danger” — then running or fighting without even thinking.

Music and Rhythm: When we hear only the first few notes of a song, we match them with the musical shapes in our memory and remember the song at once.

All of these happen inside our head in a moment, without our will, and they come from the patterns of things we already know. We can even put them in groups: visual signals (seeing faces or shapes), survival feelings (guessing weather, sensing danger), and hearing skills (following a melody or a beat).

And this is not only true for us. At every point of evolution, pattern recognition shows itself in some way; animals, and even tiny living things, are shaped by it through natural selection. The one that cannot read a pattern, or reads it wrong, does not survive.

The Two Faces of Patterns: The Expert Eye and Cultural Belief

Let me open a small side note here, because patterns have two interesting faces — and both of them will connect to design in a minute.

The first is skill. Pattern recognition is not a fixed ability; it can be trained and made sharper. A 2009 report from the National Academies about pain in laboratory animals shows this very clearly. To know if an animal is in pain, you must know the normal behaviour of that animal type very well, so you can notice the small changes. Two people can look at the same animal: one sees nothing, while the trained eye reads a dozen signals. So a pattern becomes visible only to people who know the “baseline” (the normal, starting point). This is exactly why a web designer can look at a site for one second and say “this is a Shopify theme,” while a normal user notices nothing. The pattern is there — but you need training to see it.

The second is culture. Pattern recognition often does not show the truth; it only shows the brain’s need to fill in the blanks. This is why it also led to superstitions, religions, myths, and legends. The best example of this is the black cat.

In medieval Europe, people believed that black cats walking quietly at night were witches in a different shape. The colour black stood for darkness, the unknown, and death, so it was linked to evil. But here is the interesting thing: the same animal could mean the opposite in another place. In Scotland, a strange black cat that came into your house was a sign of money and good luck. In Japan, a black cat crossing your path was often a sign of a happy love life.

This is not a coincidence. The historian of religion Ingvild Sælid Gilhus, in her study of how people saw animals in ancient Greek, Roman, and early Christian thought, points to one key idea: animals wake up mixed feelings in us, so they easily become “natural symbols,” and the same animal can be read in completely different ways in different cultures. So the black cat is neither lucky nor unlucky; the meaning is put on it by the shared pattern map of that society. When these patterns sink into the identity of a society, they guide our behaviour without our knowing. Even today, some of us will walk down a different street instead of the one where a black cat is walking.

So this is what I want you to keep in mind: pattern recognition is both a skill that lets us read the truth, and something that can make us “see” meanings that are not really there. Design lives exactly where these two things meet.

Our Main Topic: Design Choices

Today, designs are already split and shaped into clear types, and most of us can tell them apart with our eyes without even noticing. When you look at a piece of paper, the font, the pictures, and the other details help you know if it comes from a book or a newspaper. But what if newspapers had always been small books, printed just like books? Then no one could tell them apart, and we would have to read the content and look for other clues.

It is the same with websites. A website’s design is just a first tool that tells you what the site is for. By “first” I mean this: the design shouts, in the very first second, who should use this site and who it is made for. Picture a blog, a news site, and a forum in your mind — three different design shapes appear at once. This is because your mind has built a template from the thousands of sites you have seen. And this template is a pattern.

When you enter a site, you understand what is there by looking, not by reading. The site itself shouts what it is; and you decide “do I need this or not?” This is where the importance of design shows up: design is a signal that works before you read the content. And research shows how fast this is — a very large part of people’s judgment about a site is formed in less than a second, from the visual design alone, before they even look at the content.

Why Are They All the Same? The Comfort of the Brain

Now I come to my main point. In my eyes, there is one reason why most websites look the same: comfort. People’s taste moved in this direction because the brain gets tired less when it works with something familiar. And this is not just my own opinion; there is a rule in design psychology that stands firmly behind it.

This rule is called Jakob’s Law. Jakob Nielsen, the co-founder of the Nielsen Norman Group, wrote it in 2000: Users spend most of their time on other sites; so they expect your site to work like all the other sites they already know. Think of an online shop — you probably pictured something like Amazon in your mind: a search bar on top, a cart icon in the corner, product cards. This is not a coincidence; because everyone is used to this order, a new user can use the site without learning anything. Nielsen takes this so seriously that in one study, just moving the logo to the centre (breaking the usual left-side habit) made many more users fail to reach the home page with a single click. One unusual move can break how easy a site is to use.

But why is the familiar so comfortable? The answer is in a idea from psychology called processing fluency (how easily the brain “processes” something). The 2004 study by Reber, Schwarz, and Winkielman says this: if something is easy for the brain to process, the brain finds it more beautiful. Simple, balanced, high-contrast, and prototypical designs (designs that are close to the most typical example of their group) need less mental work; and this low effort turns straight into a feeling of “I like this.” One famous paper by Winkielman says it in one line: prototypes look attractive because they are easy on the mind.

Now think about it the other way. The more you see a pattern, the more easily your brain processes it; and because it processes it easily, it finds it more “right” and more “beautiful.” So websites looking the same shapes people’s taste; and this shaped taste then asks for even more similar sites. It is a loop that feeds itself. This is why a normal-looking site is often not a mistake, but a quiet comfort given to the user.

The “Rules” of This Job Are Also Patterns

If you are a web designer, you know one very simple rule: the 60-30-10 rule. 60% of the screen goes to one main background colour, 30% goes to a second, helper colour (for cards and other surfaces), and 10% goes to a strong, opposite colour that works as a mark and catches the eye.

So let me ask you: is this rule really important? Actually, no. And the funny part is this — the rule itself is just another pattern. The first websites did not have “simplicity”; too many colours and too much noise made a page that was hard to read. Designers needed a more simple structure, so they started to keep one colour heavy in the background to control where the user’s eye would go. This choice worked, it was repeated, it spread — and at some point it became a “rule.” So 60-30-10 is not a law of nature; it is the name of a pattern that is repeated because it works, and is called “right” because it is repeated. Just like the black cat.

The thing that made this loop huge today is ready templates and design systems. Tools like Bootstrap, Squarespace, Shopify, and Material Design made it possible to build sites that “fit the mould and look good” with almost one click. The price of this ease was sameness: more and more sites share the same grid, the same round corners, the same font order. Add UX data on top of this — because everyone leans on the same “proven” best practices, we are all pulled toward the same result. So the sameness of the web is not an accident; it is the most natural output of a system built by patterns, tools, and data together.

What About Niche Sites? The Thorn Inside the Crowd

Now the other side of the coin. If everyone follows the same pattern, why do some people go outside it on purpose?

Because niche and different sites are not made for everyone; they are made for a certain area, for single-style projects, or for people who are looking for that style. In short, these sites are built for their buyer, or for the person who is searching for them. And the mechanism here is the second half of the same fluency theory that people often skip: how “easy” a thing feels depends not only on the thing, but also on the person looking — on their past experience and their skill. So a design that looks “messy” to one group can be a smooth, familiar, and correct signal for someone inside that world.

The clearest example of this is the brutalist / anti-design movement of recent years. In a sea of round corners, soft colours, and perfect animations, brutalist sites show raw HTML, system fonts (like Courier New or Times), broken grids, and rough, plain surfaces — on purpose. This is not bad skill; it is a stand. The design articles about this movement come back to the same sentence again and again: a brutalist site is a raised head against the Shopify and Squarespace templates, a signal that says “we are not like them.” For groups that care about being real, honest, and creative, this “breaking the mould” is exactly what pulls them in. In an age when AI-made content is everywhere, even looking rough and imperfect becomes a signal: a human made this.

But here is the secret: a brutalist site does not work for everyone. In a place where sameness is so common, the unusual is also risky — it can lose a confused visitor, or miss a sale. This is why everyone who defends this movement gives the same warning: before you break the rule, know your audience. So niche design does not come from not knowing the rules; on the opposite, it comes from breaking the rules on purpose and for the right people. Whatever the familiar comfort is for a normal user, that “breaking the mould” gives the same comfort to a niche user — just through a different pattern map.

Conclusion: A Site Shouts to Its Buyer

Let me put everything I have tried to say into one sentence: a design is a pattern that speaks before its content, and this pattern speaks straight to the mind of its target group.

Most sites looking the same is not laziness or a lack of imagination; it is the result of the comfort the brain feels with the familiar — a pattern built up in the taste of millions of people. Jakob’s Law says this through usability, and processing fluency says it through beauty; both point to the same place. Being normal is, most of the time, a quiet ease given to the user.

Different and niche sites reject this comfort on purpose — but not into empty space, into another pattern. They do not speak to everyone; they speak to a group that sees itself inside that world. Just as a black cat is unlucky on one street and lucky on another, a design can be “messy and cold” for one group and “real and correct” for another.

In the end, whether you choose the normal or the unusual — both are a choice of pattern. The real power of design is not in which pattern you choose, but in knowing who that pattern speaks to. Because a site shouts to the one who looks, before they read; and a well-designed site shouts to exactly the person who needs to hear it.

Sources

Books

Yuan Yan Tang, Wavelet Theory Approach to Pattern Recognition (2nd edition), Series in Machine Perception and Artificial Intelligence, Vol. 74, World Scientific, 2009. — Used for the technical side of pattern recognition: the four steps (data collection, pre-processing, feature extraction, classification) and the difference between supervised and unsupervised learning.
Ingvild Sælid Gilhus, Animals, Gods and Humans: Changing Attitudes to Animals in Greek, Roman and Early Christian Thought, Routledge, 2006. — Used for the idea that animals become “natural symbols” and that the same animal can be read differently in different cultures (the black cat example).
Committee on Recognition and Alleviation of Pain in Laboratory Animals, Recognition and Alleviation of Pain in Laboratory Animals, National Academies Press, 2009. — Used for the “expert eye” idea: reading pain needs someone who knows the normal baseline behaviour of a species and can notice small changes.

Research / Web sources

Jakob Nielsen, “Jakob’s Law of the Internet User Experience,” Nielsen Norman Group (first stated in 2000). — The rule that users expect your site to work like the other sites they already know, and the study on the centred-logo problem.
Rolf Reber, Norbert Schwarz & Piotr Winkielman, “Processing Fluency and Aesthetic Pleasure: Is Beauty in the Perceiver’s Processing Experience?”, Personality and Social Psychology Review, 2004. — The idea that things which are easy to process (simple, balanced, prototypical, high-contrast) are found more beautiful, and that ease depends on the viewer’s experience and skill.
Piotr Winkielman, Jamin Halberstadt, Tedra Fazendeiro & Steve Catty, “Prototypes Are Attractive Because They Are Easy on the Mind,” Psychological Science, 17(9), 2006. — The point that prototypical (most typical) designs are preferred because they need less mental effort.
Web writing on the brutalist / anti-design movement (e.g. Nielsen Norman Group, Laws of UX, and design blogs such as Shopify Partners, Designer Daily, and others). — Used for the counter-movement: raw, “un-designed” sites as a signal of “we are not like them,” and the advice to know your audience before breaking the rules.
Cover Photo https://wallhaven.cc/w/mpx3jk