Your site’s got 1/20th of a second to make a first impression Jason 17 Jan 2006

31 comments Latest by Rhet


In the study, researchers discovered that people could rate the visual appeal of sites after seeing them for just one-twentieth of a second. These judgments were not random, the researchers found — sites that were flashed up twice were given similar ratings both times.

The article didn’t talk much about what works and what doesn’t, or what customers consider a thumbs up or thumbs down, but the fundamental response is certainly worth noting.

I would suggest this is a big win for the big text camp. If people can quickly read something they’re probably more likely to “get it” and consider it worth their time. Just a hunch though.

31 comments so far (Jump to latest)

Joost van der Borg 17 Jan 06

I would suggest this is a big win for the big text camp.
I disagree. No matter how big your text is, do you really think someone can read any of it in 1/20th of a second?

John B 17 Jan 06

I agree with Joost. I think all you’re going to get in 50 milliseconds is a general impression of layout and color—content no matter how big and bold is going to take more time than that to process.

JF 17 Jan 06

The point isn’t that you can/can’t read it — the point is that people can recognize letters. They feel like there’s value there.

Imagine looking at a blob for 1/20th of a sec or a piece of big type for 1/20th of a sec. Which do you think more people will consider to be of more value?

It’s not the read, it’s the potential of the read that is valuable in a 1/20sec comparison.

Rick 17 Jan 06

You had me at 37sig

Joost van der Borg 17 Jan 06

The point isn�t that you can/can�t read it � the point is that people can recognize letters. They feel like there�s value there. […] It�s not the read, it�s the potential of the read that is valuable in a 1/20sec comparison.

Agreed, but really, big text or little text, a user can see it’s there either way, but can’t read it in that time. So I’d say the value is equal.

gwg 17 Jan 06

…rate the visual appeal…

All that this study examined is *visual appeal*. It did not study information absorbsion, information retention, or how visual appeal relates to anything else.

Any attempt to extrapolate additional conclusions from this study is pure speculation and argument. To suggest that it’s a win for any camp (big text, small text, red text, blue text) is silly.

Francis Wu 17 Jan 06

gwg’s right… this one’s about visual appeal. I hate to sound clich�ed, but beauty’s really in the eye of the beholder. While I see credibility and beauty in minimalistic 37s-like sites, my boss finds credibility and beauty in sites that are heavy with the graphics and overbearingly corporate-looking. Sigh. I’d post our corporate site here, but I don’t wanna get dooced again :P.

Francis Wu 17 Jan 06

Sorry, no… it’s no corporate site. It’s supposed to be a web application… but it looks very corporate 1998. The business blue, the image rollovers, the Nascar Effect, and the marketing pitch. It’s a veritable web time capsule, it seems. I’m just glad it doesn’t use frames.

Sorry, I’m done complaining :P.

Rimantas 17 Jan 06

I am with gwg on this one, too. All the study did is to show that people CAN decide visual appeal of the site in 50ms almost the same as in 500ms. So what? Do we only use 50ms?
No doubt, visual appeal is important, but site which has no info I am looking for right now will mean nothing to me, no matter how visually pleasing it may appear in 50ms.

Gary R Boodhoo 17 Jan 06

Proportion, visual rhythm and novelty are the criteria being selected for by this study. For example, if every website used big text, the ones using tiny text in a field of negative space would be the ones that got noticed in 1/20 sec. or less. Pattern recognition. Nothing more, nothing less. Our brains are really good at it though!

Martin 17 Jan 06

Well, speaking of first impressions… when I clicked the link to the article, I thought I ended up on the Dell site. The first thing you see is a big banner (from Dell in this case), with the Dell logo much bigger than the Reuters one in the corner.

MT Heart 17 Jan 06

Interesting. Malcolm Gladwell has other examples of this type of thing in his book, “Blink”. Like when doctors at emergency rooms go with their “gut feel” about a patient rather than wasting time running blood work, X-rays etc..

I guess you also have to have a PC/connection fast enough to render a page in that amount of time.

BTW, want to see what 50ms looks like…

I had to click the 50ms link a few times to get an overall impression though… Hmm, wonder what THAT means.

Sebastian Gr��l 18 Jan 06

You can also take it the other way. People come to your site see small text and stay longer to read it… ;-)

dusoft 18 Jan 06

Jason: The article I read was specific that people don’t read the text in 50 milliseconds, they just get the positive or negative feeling. I don’t really think this has to do anything with the big text or the presumed value in the big text.

brian 18 Jan 06

Allow me to testify for literate content. When you use construction such as “Your site’s got …” in the headline, you risk losing eyeballs immediately. The Internet is cluttered with sites featuring piss-poor grammar, lousy spelling, bloated narrative and half-formed thoughts of (it seems) mostly twenty-somethings fascinated by site design and technological tricks. But many have little of interest to say, and no ability to say it intelligently. I do not accuse you of any of these failings, but I do point out that we browsers develop certain mental tools to preserve our precious time. One of my tools is a red-alert when I see an ungrammatical contraction in a headline. Why not: “Your site has …”? Use apostrophes in possessive words. Do not use contractions in a written medium that has no real space constraints. Contractions are unnecessary. The reader needs more than your 1/20 of a second to decide that the contraction in the headline is not a possessive. If we could banish the word “got” from the language, we could effect a great improvement in grammar overnight.

Carsten Rose Lundberg 18 Jan 06

Having given it much thought over the day I’ve come to the conclusion that this test only tells us one thing worthy of interest - people tend to stick with their first impression.
No page can render fully in 50 ms. and the conclusion was that there was no significant difference in whether people was exposed 50 or 500 ms - so that leaves us with no clue as to how long it actually takes to generate a first impression. My bet is that it’s individual.

What stands is that first impressions do last so you have to make a good first impression. We all like to be right so we tend to debate ourselves into continuing to like a site even when we find flaws merely to prevent us from proving ourselves wrong. This means that it’s steep uphill for sites that generate a bad first impression.

HS 18 Jan 06

It seemed to me also that it was about visual appeal, and I understood that to mean nicely designed. Usability or “IA” (or text size analysis) help for sure but I don’t think that’s what makes users form their snap judgements - I think design does.

Per Melin 18 Jan 06

Here is a longer article on the study:

Caudron suggests that the amount of graphics on the page should be strictly limited, perhaps to a single eye-catching image. “It’s not about getting as much stuff on the page as possible,” he says.
These days, enlightened web users want to see a “puritan” approach, Caudron adds. It’s about getting information across in the quickest, simplest way possible.

Darrel 18 Jan 06

So…they are saying that first impressions happen upon first viewing?

For example, will you buy anything from ? I wouldn�t.

Absolutely. In fact, the less ‘shiny and marketing-fluffy’ the site is, the more likely I’m going to think it’s a sincere business.

And my experience seems to back that up. The best online shopping for me has often been from ma and pa shopping sites. Ugly, but usable, and a real person behind the ‘counter’.

Really no different that dining at your local corner dive cafe. The place may be ugly, but the foods good. And no matter how shiny a Perkins might be, the food is always crap. ;o)

Dan Boland 18 Jan 06

Darrel: Interesting that you see it that way — it’s certainly a valid point and one that I hadn’t considered. Kinda reminds me of the riddle with the two barbers — you go to the one with the crappy haircut because he doesn’t cut his own hair. Okay, maybe that doesn’t really apply… but it made me think of it nonetheless.

My thought is that if they don’t take the time to make their site look good, it doesn’t say much about their products. (And any use of Comic Sans qualifies.)

Gary R Boodhoo 18 Jan 06

Brian, great point re: grammar as validation. I feel the same way whenever I see quotation marks used inappropriately. For example, outside a Burger King drive through window, I am presented with the text:

- This is a “drug free” workplace.

What do the quotes mean? Irony? Grudging conformance? Comic Sans (and usually Mistral) generate the same ironic distance in less than 1/50 sec. (for me anyway)

Javier Cabrera(ClearYourMind) 18 Jan 06

I think the article is right. It may take a little longer than 1/20th of a second, but first impressions are everything in the web; of course, if you’re browsing for something specific you will be more likely to read for at least 10 seconds before you run away to keep looking, but I might be guessing, like those guys on the article ;-)

Just kidding on that last one.

Like how SVN gets updated so often! keep working!

Darren James Harkness 19 Jan 06

Honestly, when I see big text on a webpage, I think to myself, “God, they must think I’m slow,” shortly before thinking, “and they must not have much to say.”

Splashman 19 Jan 06

Any given design feature (big text, Photoshop bevels, 3-D ray-traced soft-shadowed rollover nav) can be used to positive effect by someone who knows what they’re doing, or used to negative effect by someone who can afford the software but doesn’t know what the hell he’s doing. The point of the referenced study is that most people will be able to conclude the former or the latter in 1/20 of a second.

Show me five sites that have use [feature] well, and I’ll show you 100 that don’t.

Geez, Jason.

Splashman 19 Jan 06

Whoops — not sure what happened to that last sentence. How about this:

Show me five sites that make good use of [feature], and I’ll show you 100 that don’t.

And on the basis of Jason’s blog entry, I’ll just betcha we start seeing WAY more of the “don’t” variety.

nextquestion 20 Jan 06

I read this story earlier this week and I was kind of shocked. The number used to be around 5 seconds.

Just goes to show that you need to 1) have the content 2) present it in a logical way to users.

Anyway, I have more to say here:

Nooberman 26 Jan 06

Scientists say “blah-blah-blah” after some study …
And … ?
Anyone who surfs the web loads, like most of us here I guess, should already be fairly aware of this sort of thing.
If a webpage takes ages to load - do you sit and wait? I know I don’t bother.

As for first impressions … what can you do? (Rhetorical!)

Maybe you don’t like the wallpaper in my house …
Maybe I don’t like the carpet in yours …
Design is something that can be really bad from one person’s perspective and not so bad from another. Now, really bad design gets noticed by all, as does really good design, but if you land somewhere in the middle I don’t think you’ll offend everyone that comes along.

FFS, we don’t all see colour the same way. There is no guarantee with design. When I talk of design here, I’m referring to colours and fonts and graphics specifically.

Massive (or uber tiny) pink text on bright yellow background in an italic font with lots of flashing images - yuk!

Layout and structure are more serious matters though, and probably do fall into design, but if you get these right and choose the wrong colours, fonts and graphics then was there any point in getting your structure and layout right in the first place?

theCreator 27 Jan 06

Hmph! I’m in the Big Text camp.

Rhet 30 Jan 06

I COMPLETELY AGREE. I KNOW HOW I FEEL ABOUT SOMETHING IN A MILLASECOND. The spirit of a thing conveys, communicates with my spirit in a moment. I can tell whether the truth is being told or not. I feel the energy that was infused in the work. Was the creator coming from the heart or was the head the primary tool used? Were they ‘trying’ to create results or were they actually doing it? These are all spiritual subtleties that affect us all.

It’s what happens when someone sees to a movie then says ‘it was ok’ or ‘I loved it’ Point proven when you ask them why? They have no real idea, just an impression a feeling. The truth, we are spiritual beings, we are about feelings not ideas. Tangible results - and the truth can’t lie. We know the truth in a second.

The problem, most do listen to their heart, intuition or first impression - the ones that do - succeed. The others, make mistake after mistake.

What is is what is - period. The truth hurts but you can’t argue with it!