10 ms·

The myth of the page fold: evidence from user testing

69 points by jmorin007 17y ago

req2 17y ago
I'm curious as to what their test protocol is that it has dispelled every other posting on this topic. A person told to buy shoes at Amazon.com isn't going to be stifled by a scrollbar, but someone who isn't being directly told to interact with the website in a given manner doesn't seem as likely to scroll down looking for content that isn't found on a sparsely populated top page. (Related: http://www.useit.com/alertbox/word-matching.html http://www.useit.com/alertbox/word-matching.html)
btn 17y ago
The lack of detail in what their users were actually doing on the sites they tested makes it hard to tell what they're trying to dispel. If you're giving users a specific task to perform on a specific website in a controlled environment, then of course scrolling isn't going to deter them. The argument against scrolling is that if you're trying to attract users to your site---to pick your site over other potential sites---then scrolling will be a factor when people want to make a quick decision if your site is useful or not.
- ams6110 17y ago
  They also didn't mention horizontal scrolling, which I personally cannot stand. Vertical scrolling is not nearly as objectionable, though in forms it's more of a bother than on pages where you are just reading.
- ubernostrum 17y ago
  Well, this is really the problem. The fold really only becomes a huge factor when you're trying to win someone's attention in that snap "does this site have what I need" judgment; once the user's mind is made up that, yes, this is the right place, it becomes largely irrelevant. But somewhere along the line, a useful guideline for getting people to stick around and explore a page or a site morphed into a hard rule of "never have anything at all below the fold, ever". Which is frankly stupid.
JDigital 17y ago
Jakob Nielsen helped to popularize the myth that web users don't scroll down. It led to an era of website design that crammed as much onto one page, leading to busy, crowded homepages. It wasn't until later that early adopters of CSS made websites simple again, in part to avoid compatibility issues with Internet Explorer.
- alabut 17y ago
  I'd rather blame people that took his advice to extremes and tried to shove everything above the fold, which isn't what he advocated, just to add the most important stuff up top: "Is all key information visible above the fold so users can see it without scrolling?" http://www.useit.com/alertbox/screen_resolution.html http://www.useit.com/alertbox/screen_resolution.html
  - ahoyhere 17y ago
    Of course you can blame the patient, so to speak, because they took the doctor's advice to the extreme. Except that everything Nielsen writes/promotes is couched in such a way that it looks black & white, and so naturally people take it to extremes. Because he encouraged them to, in the way that he presents his information. Consider the surrounding circumstances to the "pure" information he posts: 1. He says it is SCIENCE! 2. He does not allow for differing levels of user engagement. (If you're interested, you will not behave the same way as somebody who has no interest.) 3. He does not reveal the type of user / type of content being tested. (Are they bored to tears? Is it well-written, captivating content? Is it something relevant to their life?) Would this be hard? Yes. But you cannot evaluate the "facts" without it! 4. He may waffle a little now and again, and say something that's not entirely "my way or the highway" strident -- e.g. "Is all key information visible above the fold?" -- but... 5. He NEVER lists proven counter-examples. Examples for #5: many "sales letters" of products on the internet (esp. info products) rely on heavily engaging the user's interest FIRST, and then following a build-up-let-down pattern of excitement. Often their critical info -- price, call to action buttons -- are well below "the fold." Waay, way down there... they only reveal it to people who are biting the hook. And this works. Really, really well. It works because it engages the visitor, and builds suspense, rather than just laying it all in front of them. But has Nielsen ever addressed this? Ha! He uses the attachment that people have to "science," "statistics," and "facts," and exploits it to provide certainty in an an uncertain world. Too bad it's false certainty.
elblanco 17y ago
The heat maps of eye movement were fascinating...I noticed the almost complete lack of activity on the advertisements immediately. Users appear to do what I do, just simply ignore them or try and filter them out.
- ams6110 17y ago
  What I noticed was that most of the attention was between the left side and center of the page. The only attention point on the right side was the scroll bar. Made me think of the NEXTSTEP window manager where the scroll bars were on the left. The explanation at the time for this apparent deviation from the "standard" was that it kept the scroll bar closer to the user's primary attention areas on the page. In Mac OS X i guess Apple caved to popular convention and moved the scroll bars to the right.
  - elblanco 17y ago
    I wonder why only a particular part of the scroll bar appears to have attracted long gaze patterns? In my own case, I'm typically looking at the slider as a metric of page length, not a particular spot on the bar itself.
    - gambling8nt 17y ago
      I noticed that as well; I assume people basically all tended to check page length at the same point through the page--presumably at some boundary between content sections.
- ahoyhere 17y ago
  It's called "ad blindness," a well-known problem. That's why I don't even bother to put ads on my sites. People also ignore anything in colored blocks, or that looks like an ad. Thus ReCaptcha had a huge problem initially with people just "not seeing" it on the forms where it was placed. It looked like an ad.
  - elblanco 17y ago
patio11 17y ago
Somebody forgot to send my users the memo, apparently.
- req2 17y ago
  I'm sad to see that no one else has upvoted you; you seem to be the only one here that has had the experience to weigh in on the issue.
hussong 17y ago
Looks like the 'Holy Scrollers' have won, while the 'Sharks' (everything in one bite) are left to chew on their hypercard metaphor.
maudineormsby 17y ago
Worth noting that while you can have content below the fold and signal to users to find it, that's not true with branding and identity features - those should be prominent and above the fold. Further, as a commenter on the article pointed out, there's benefit in identifying your audience. Someone on NYT is obviously willing to scroll. Can the same really be said of someone shopping for a luxury car and looking to be wowed?
lurkinggrue 17y ago
Yeah, what about the scroll-wheel? I hardly ever touch the scroll bars anymore.
KarlGilis 17y ago
Whether users want to scroll or not, depends from situation to situation. Visitors will not always scroll. So it's not just a myth. It depends on the type of page and the type of website. A nice article illustrating when users want to scroll (and when not) can be found at http://webusability-blog.com/page-fold-fact-or-fiction/ http://webusability-blog.com/page-fold-fact-or-fiction/. It also gives some good examples of good use of the area above the page fold.

69 points by jmorin007 17y ago

req2 17y ago

I'm curious as to what their test protocol is that it has dispelled every other posting on this topic. A person told to buy shoes at Amazon.com isn't going to be stifled by a scrollbar, but someone who isn't being directly told to interact with the website in a given manner doesn't seem as likely to scroll down looking for content that isn't found on a sparsely populated top page. (Related: http://www.useit.com/alertbox/word-matching.html http://www.useit.com/alertbox/word-matching.html)

btn 17y ago

The lack of detail in what their users were actually doing on the sites they tested makes it hard to tell what they're trying to dispel. If you're giving users a specific task to perform on a specific website in a controlled environment, then of course scrolling isn't going to deter them. The argument against scrolling is that if you're trying to attract users to your site---to pick your site over other potential sites---then scrolling will be a factor when people want to make a quick decision if your site is useful or not.

ams6110 17y ago
They also didn't mention horizontal scrolling, which I personally cannot stand. Vertical scrolling is not nearly as objectionable, though in forms it's more of a bother than on pages where you are just reading.
ubernostrum 17y ago
Well, this is really the problem. The fold really only becomes a huge factor when you're trying to win someone's attention in that snap "does this site have what I need" judgment; once the user's mind is made up that, yes, this is the right place, it becomes largely irrelevant. But somewhere along the line, a useful guideline for getting people to stick around and explore a page or a site morphed into a hard rule of "never have anything at all below the fold, ever". Which is frankly stupid.

JDigital 17y ago

Jakob Nielsen helped to popularize the myth that web users don't scroll down. It led to an era of website design that crammed as much onto one page, leading to busy, crowded homepages. It wasn't until later that early adopters of CSS made websites simple again, in part to avoid compatibility issues with Internet Explorer.

alabut 17y ago
I'd rather blame people that took his advice to extremes and tried to shove everything above the fold, which isn't what he advocated, just to add the most important stuff up top: "Is all key information visible above the fold so users can see it without scrolling?" http://www.useit.com/alertbox/screen_resolution.html http://www.useit.com/alertbox/screen_resolution.html
- ahoyhere 17y ago
  Of course you can blame the patient, so to speak, because they took the doctor's advice to the extreme. Except that everything Nielsen writes/promotes is couched in such a way that it looks black & white, and so naturally people take it to extremes. Because he encouraged them to, in the way that he presents his information. Consider the surrounding circumstances to the "pure" information he posts: 1. He says it is SCIENCE! 2. He does not allow for differing levels of user engagement. (If you're interested, you will not behave the same way as somebody who has no interest.) 3. He does not reveal the type of user / type of content being tested. (Are they bored to tears? Is it well-written, captivating content? Is it something relevant to their life?) Would this be hard? Yes. But you cannot evaluate the "facts" without it! 4. He may waffle a little now and again, and say something that's not entirely "my way or the highway" strident -- e.g. "Is all key information visible above the fold?" -- but... 5. He NEVER lists proven counter-examples. Examples for #5: many "sales letters" of products on the internet (esp. info products) rely on heavily engaging the user's interest FIRST, and then following a build-up-let-down pattern of excitement. Often their critical info -- price, call to action buttons -- are well below "the fold." Waay, way down there... they only reveal it to people who are biting the hook. And this works. Really, really well. It works because it engages the visitor, and builds suspense, rather than just laying it all in front of them. But has Nielsen ever addressed this? Ha! He uses the attachment that people have to "science," "statistics," and "facts," and exploits it to provide certainty in an an uncertain world. Too bad it's false certainty.

elblanco 17y ago

The heat maps of eye movement were fascinating...I noticed the almost complete lack of activity on the advertisements immediately. Users appear to do what I do, just simply ignore them or try and filter them out.

ams6110 17y ago
What I noticed was that most of the attention was between the left side and center of the page. The only attention point on the right side was the scroll bar. Made me think of the NEXTSTEP window manager where the scroll bars were on the left. The explanation at the time for this apparent deviation from the "standard" was that it kept the scroll bar closer to the user's primary attention areas on the page. In Mac OS X i guess Apple caved to popular convention and moved the scroll bars to the right.
- elblanco 17y ago
  I wonder why only a particular part of the scroll bar appears to have attracted long gaze patterns? In my own case, I'm typically looking at the slider as a metric of page length, not a particular spot on the bar itself.
  - gambling8nt 17y ago
    I noticed that as well; I assume people basically all tended to check page length at the same point through the page--presumably at some boundary between content sections.
ahoyhere 17y ago
It's called "ad blindness," a well-known problem. That's why I don't even bother to put ads on my sites. People also ignore anything in colored blocks, or that looks like an ad. Thus ReCaptcha had a huge problem initially with people just "not seeing" it on the forms where it was placed. It looked like an ad.
- elblanco 17y ago

patio11 17y ago

Somebody forgot to send my users the memo, apparently.

req2 17y ago
I'm sad to see that no one else has upvoted you; you seem to be the only one here that has had the experience to weigh in on the issue.

hussong 17y ago

Looks like the 'Holy Scrollers' have won, while the 'Sharks' (everything in one bite) are left to chew on their hypercard metaphor.

maudineormsby 17y ago

Worth noting that while you can have content below the fold and signal to users to find it, that's not true with branding and identity features - those should be prominent and above the fold. Further, as a commenter on the article pointed out, there's benefit in identifying your audience. Someone on NYT is obviously willing to scroll. Can the same really be said of someone shopping for a luxury car and looking to be wowed?

lurkinggrue 17y ago

Yeah, what about the scroll-wheel? I hardly ever touch the scroll bars anymore.

KarlGilis 17y ago

Whether users want to scroll or not, depends from situation to situation. Visitors will not always scroll. So it's not just a myth. It depends on the type of page and the type of website. A nice article illustrating when users want to scroll (and when not) can be found at http://webusability-blog.com/page-fold-fact-or-fiction/ http://webusability-blog.com/page-fold-fact-or-fiction/. It also gives some good examples of good use of the area above the page fold.