The Psychology of First Impressions: What Your Homepage Says in 50 Milliseconds

Your homepage does two separate jobs in quick succession, and almost everything written about first impressions collapses them into one. The first job is over before anyone is conscious of it. The second takes a few seconds and is decided entirely by the words. They need different fixes, and confusing them is why so many redesigns change how a site looks without changing what it does.

Job one: the glance

Lindgaard and colleagues at Carleton University, writing in Behaviour & Information Technology in 2006, found that visual appeal ratings formed after 50 milliseconds matched those formed after 500 milliseconds. Show someone a page for a twentieth of a second and their verdict already agrees with the one they give after a much longer look.

That is faster than a deliberate thought and far faster than reading. Whatever your homepage is saying at that point, it is not saying it in sentences. It is saying it in shape, density and contrast.

Most homepage arguments happen about the third stage, which is the one fewest visitors reach.

What actually wins the glance

This is where the honest answer irritates people who commission design. Tuch and colleagues, in the International Journal of Human-Computer Studies in 2012, tested 119 real websites and found that low visual complexity and high prototypicality — how closely a page resembles what people expect that kind of page to look like — produced the most positive first impressions, with the effect detectable at 17 milliseconds.

Simple and familiar. Not striking, not distinctive, not memorable. A conventional layout with generous space and one clear focal point is working with the grain of the finding; a full-bleed video with three overlapping panels and an unusual navigation pattern is working against it.

That does not mean every site should look the same, and it is not an argument for a default template. It does mean that if you want to be visually unusual, you are spending something at the very start of the visit, and you should know that you are spending it.

What this research does not say

Findings this striking get stretched, and this one gets stretched more than most. It is worth being precise, because a reader who checks will find the same limits.

The Lindgaard study measured visual appeal. It did not measure trust, credibility, conversion or revenue, and it did not follow anyone to a purchase. “Visitors decide whether to trust you in 50 milliseconds” is a sentence you will see everywhere, and it is not what was tested. What was tested is whether a page looks pleasant, judged very fast.

That still matters, because a page that looks unpleasant has to overcome that impression to get anything else across. It just is not the same as trust, and any article that tells you otherwise — including the earlier version of this one — has gone further than the evidence.

Job two: saying what you do

The second job is where most homepages actually fail, and it is not a design problem at all.

Jakob Nielsen of Nielsen Norman Group, writing in 2011 on Microsoft’s analysis of more than 205,000 pages, found that visitors often leave within 10-20 seconds, and that holding attention beyond that means communicating your value proposition within about the first 10 seconds. The mechanism is scepticism rather than impatience: people arrive assuming the page will not be useful and give it a brief chance to prove otherwise.

They also read far less than you would hope. Nielsen Norman Group analysis of 45,237 page views, published in 2008, concluded that on an average page a visitor has time to read at most 28% of the words, and that 20% is likelier. Both studies are old and should be treated as directional. The direction has not changed: a short window, and a fifth of your words.

Which puts a hard constraint on the top of your homepage. One sentence, doing the work. Not a mission statement, not a welcome, and not “Delivering excellence since 1998” — a plain statement of what you do and who you do it for, in the language your customer would use to describe their own problem.

What to do about it

The fixes divide neatly along the two jobs, and they are unglamorous in both cases.

• Reduce the number of things competing at the top of the page to one
• Remove anything that moves on its own sliders, auto-playing video, animated counters
• Give the page one dominant next action rather than three equal ones
• Replace the positioning sentence with a plain description of what you do and who for
• Put the specific detail early — the sector you serve, the area you cover, the price band
• Make body text properly readable rather than pale grey, and check it in daylight on a phone

A useful test costs nothing. Show your homepage to someone outside the business for a couple of seconds, take it away, and ask two questions: how did that feel, and what does the company do? The first answer tells you about job one. The second tells you about job two. In most cases the first is fine and the second is a shrug, which is a copywriting brief rather than a design one.

Where design still earns its fee

None of this makes design decorative. The glance is a design outcome, and getting it right means someone deciding what to leave out, which is harder than it sounds and rarely done well by committee. Beyond the first impression, design is what makes the specific parts findable — the price, the proof, the phone number — and what stops a page collapsing into an undifferentiated wall on a phone.

The shift is in what you ask design to do. Not “make it look impressive”, which pushes towards complexity that the evidence says costs you. Something closer to: make the first glance calm, and make the important sentence impossible to miss.