“Grass is green and hyperlinks are blue.” — Elise Blanchard
If you’ve ever wondered why links on websites appear blue by default without anyone explicitly telling them ‘make me blue’, here’s a surprising story of technological evolution, perception research, and social conventions. It’s not an aesthetic whim or a mere coincidence: it’s the result of a combination of chance, usability studies, legacies from the digital past, and ultimately, the cultural ‘muscle memory’ of web users. All of this has turned blue into the color of links, to the point that it seems invisible to us due to how normalized it has become.
The background: link research before the WWW
Before the World Wide Web became ubiquitous, hypertext systems already existed that explored how to indicate links within texts. In the field of Human-Computer Interaction, Ben Shneiderman’s laboratory at the University of Maryland investigated various ways to highlight embedded links in text during the 1980s. Different colors and formats (underlines, bolding, etc.) were tested to see which options interfered least with reading and, at the same time, intuitively attracted user attention. From those experiments, a winning solution emerged: using a light blue by default for selectable links. It wasn’t for aesthetics, but for pragmatism.
“Highlighting in red made links more visible, but reduced the ability to read and retain content… blue was visible on black and white backgrounds and did not interfere with retention.” — Ben Shneiderman
These studies materialized in concrete projects such as the HyperTIES system (1983), a commercial hypertext environment where links already appeared with text highlighted in light blue on a dark background. In fact, HyperTIES is considered the first
Mosaic and the consolidation of blue as a standard
Despite these precedents in hypertext systems, the decisive moment for establishing blue as a standard arrived with the Mosaic browser in 1993. Created by Marc Andreessen and Eric Bina, it was one of the first popular graphical browsers, and in its initial versions, it introduced links as blue, underlined text. Before that, other interfaces had used different indicators: in Tim Berners-Lee’s first browser (WorldWideWeb, 1991) links were black or underlined on a white background, and in older text environments, underlines or default system colors were often used (for example, green text on black in the Gopher protocol). Mosaic, however, broke the mold by defaulting to blue for unvisited links (and purple for visited ones), against the characteristic pale gray background of web pages of that era.
Why did Mosaic choose blue specifically?
“Because I like blue. It’s a solid color, and I had to choose one.” — Marc Andreessen
This recent confession—half-joking—highlights the element of chance or personal preference in the decision. However, there were also important practical and contextual reasons: in the early 90s, color monitors were becoming popular, and a dark blue leaning towards ‘royal blue’ offered very good contrast on light backgrounds without being jarring. Furthermore,
Regardless, once Mosaic adopted underlined blue, subsequent browsers followed suit. Netscape Navigator (1994) inherited this convention from Mosaic, and Internet Explorer (1995), partly based on Mosaic’s licensed code, also maintained it. Thus, in a few years, millions of internet users in the 90s assimilated that link = blue underlined text. What had started as a decision by a team of developers became a de facto standard of the Web.
It should be noted that this ‘norm’ was not written in any official web manual. In fact, Sir Tim Berners-Lee (the inventor of the web) is often cited as the one who ‘made links blue,’ possibly because at some point he accepted this convention without opposing it. The historical reality is that the color blue gained popularity more than it was imposed: it was a proven practical solution and had spread thanks to Mosaic.
Perception, readability, and human factors
Choosing a visible color is only the first part of the challenge; the other part is ensuring that this highlight does not harm readability or the overall user experience. For years, some experts questioned whether blue was the best option from a strictly physiological point of view. A classic argument (mentioned by usability guru Jakob Nielsen) is that only ~2% of human retinal cones are sensitive to blue, which could make blue text less sharp or a ‘bad design’ in terms of pure utility.
However, modern practice and empirical studies disproved these fears. Experiments with eye-tracking (monitoring eye movements) have shown that underlined blue links do not cause significant differences in how we read a text compared to normal black text. During the first pass, the fact that a word is highlighted in blue does not make us skip it or hinder reading speed. However, the same study observed that, in total, readers spent a little more time re-reading highlighted words (i.e., they went back to look at them a bit more).
But this is not necessarily bad: it means that hyperlinks attract attention on a second pass, which can favor the retention of that information. In summary, blue is an almost perfect design decision: it doesn’t interfere with the first read, but it stands out enough for the brain to identify that there is something special to consider there.
Another advantage of blue, already hinted at, is its robustness against color vision problems. In the words of veteran designer Jeffrey Zeldman, it was ‘good luck’ that blue turned out to be a successful color: reds and greens are the most problematic colors for color blindness. Almost no one has a deficiency with blue. It was pure good luck that the default color for hyperlinks is blue and underlined.” That is, by chance or by trial and error, a universally recognizable color was chosen.
Many types of color blindness affect the perception of reds and greens, but it is extremely rare to have deficiencies in blue. This makes default blue a safe choice in environments where the palette cannot be highly customized or when a universal solution is desired. Furthermore, the convention of underlining links offers additional reinforcement for accessibility: even if someone couldn’t distinguish the color well, the constant underline indicates that it is clickable. It’s interesting to note that underlining had already been used in contexts like Windows 1.0 (1985) to indicate clickable text, and it is still a recommended accessibility feature today. Ultimately, underlined blue stands the test of time very well because it combines contrast, familiarity, and visual redundancy for different types of users.
Cultural memory: the habit of clicking blue text
After decades of seeing blue links, we have all developed a kind of conditioned reflex: if we see blue underlined text, we automatically assume we can click on it. It’s like a universal visual language of the internet. This cultural inertia—this ‘cultural muscle’ we’ve trained without realizing it—has powerful consequences in UX design. For example, when you see normal black text within a paragraph, you don’t think it might be a link (even if sometimes there is one!). In contrast, if you see a blue and underlined fragment, you barely need to think: you know it’s a hyperlink and you instinctively move your cursor over it to click.
A famous case illustrating the strength of this habit is the experiment Google conducted in 2016, when it decided to try displaying search result links in black instead of the traditional blue. The reaction from users (and the tech press) was
Interestingly, at the other extreme, Google also experimented extensively with which shade of blue was optimal, going so far as to test 41 different shades of blue to determine which generated the most interaction in sponsored links. Never underestimate the power of a good blue!
This cultural memory of blue is so powerful that even designers of new applications, who have total freedom to choose colors according to brand identity, often keep blue for important links or clickable elements. They know it’s a safe bet. After all, if the user already expects something to work in a certain way, following the convention improves usability. A 2019 study showed that users found a target word on a web page more quickly when that word was blue and underlined, compared to if it was black and underlined. That is, our brain detects what looks like a link sooner, because we have been ‘training’ it for years.
We can consider this phenomenon a case of ‘invisible design’: when a design element works so well and is so taken for granted by users, it becomes invisible, it goes unnoticed. As guru Don Norman said, ‘when everything works, design disappears.’ The blue color of links is a piece of invisible design because we no longer consciously think about it, we just use it. Only when it’s changed (as happened in that Google experiment) do we realize how much we miss it. Human adaptability also plays a role: we have adapted so much to this convention that a change disconcerts us. Users adapt to daily routines, and problems become invisible through repetition, not resolution. In the case at hand, blue is no longer perceived as a design decision, but as an intrinsic property of links – just as no one questions why normal text is standard black, we also don’t question the blue of links.
Alternatives and ‘playing with fire’: what happens if we don’t use blue?
Nowadays, obviously, there is no law that forces the use of blue for links. Web designers have a whole rainbow of colors to choose from, and many brands opt to apply their corporate color for brand consistency. It is common to find websites with red, orange, green, or any imaginable color links. However, this freedom comes with responsibilities: it is necessary to ensure that the chosen color offers enough contrast and is clearly perceived as a link. If we break the blue convention, we must be sure to build a new, equally clear convention within our environment. This implies, for example, maintaining the underline or other visual indicators (e.g., a change in hue when hovering, an arrow icon, etc.) that make it clear that the text is clickable. Accessibility must also be monitored: an uncommon color can be problematic for users with color blindness or black and white screens, etc., if not properly checked. As the standard states, there must be enough contrast (4.5:1 ratio against the background) for un-underlined link text.
In design circles, the debate occasionally arises: ‘should we abandon blue because it’s too boring or too old-fashioned?’ It’s an understandable temptation to want to innovate. But many experts warn that changing link colors without a compelling reason is ‘playing with fire’ in terms of usability, especially if you have a massive or less expert audience. Why? Because you risk making your site’s basic functionalities less obvious. Imagine a less web-savvy user who visits your page and finds nothing blue or underlined: how will they know where to click? Perhaps they’ll deduce it from context or by hovering, but it will be harder for them, and that’s added friction. A Nielsen Norman Group study concludes: ‘shades of blue give the strongest signal of being links, but other colors can also work almost as well, provided you offer redundant cues’. That is, you can innovate, but it must be done carefully. And under no circumstances should you use blue for texts that are not links — that would indeed be a recipe for confusion , as users would interpret it as clickable when it is not.
In summary, the default blue color for hyperlinks is a product of practice and science: it emerged from usability experiments, spread thanks to a pioneering browser, took root in digital culture, and has proven to be a wise choice from multiple perspectives (perception, accessibility, user habituation). Could we have ended up with red or green links? Perhaps, if history had been different. But technological-cultural destiny wanted them to be blue. And now it’s part of the web’s DNA. So the next time you click on a piece of blue, underlined text, remember: behind that color is a small story of evolution and shared conventions. And it works so well that almost no one stops to think about it . If that’s not well-resolved design, what is?
