Scrolling: Do They or Don’t They? A Data-Driven Analysis
April 8th, 2008 by Grace PreyapongpisanTags: , Analytics, best practices, eye tracking, fold, jakob nielsen, page design, page views, scan, scrolling, statistics, useit, veritical scrolling
We’ve all heard it: “Web users don’t scroll.”
But through our analytics work, our team has found very little actual data to support that common web myth. In fact, one of our biggest clients knows that users can be motivated to scroll down even the longest page (89% of users scrolled a long story page, and fully 60% saw 90% of the page).
On the other hand, another client’s homepage initially inspired little scrolling, with just under four out of ten users scrolling at all, and only three in ten viewing 90% of the page. So the answers to the questions of when and how much users scroll is more complex than a simple “yes they do” or “no they don’t.”
It’s a truism that important content, the content you want users to focus on, should appear “above the fold.” Does this mean that all content should be crammed into this valuable real estate? Should all content be designed to be consumed in one-screen chunks? The key to answering these design questions lies in understanding user scrolling behavior. Unfortunately, scrolling is one metric that is rarely reported on in the growing Web analytics industry and consequently, understanding how users explore long pages of content can be a mystery.
Within our Site Optimization practice, scrolling data is one of many behaviors we collect and interpret to inform the design recommendations we develop for our clients. We use scrolling data to answer three fundamentals questions:
1) Do users scroll
2) Do they scroll before they do anything else
3) How much do they scroll
These three questions provide a wealth of insight into content exposure and whether or not one-screen chunks are the best way to parse information on a given website.
Looking at a variety of “long” pages from publishers and eCommerce sites (pages in which more than half of the content is below the fold) we found that indeed, users do scroll. In fact, on average more than two-thirds of users on these pages scroll. Furthermore, nearly 75% of these users scroll before they do anything else on the page. What does this mean? Users can’t help but scan a page. They aren’t necessarily reading every word, but they are skimming, looking for that bit of content that piques their interest — a hot headline or that fancy phone the person next to them on the train this morning was busy texting away on.
But this is only part of the story. Yes, users are scrolling and skimming content and the fold isn’t a cliff which stops users in their tracks. Rather, the fold marks the top of a gradual hill. This gradual drop-off becomes readily apparent when we look at how much of the screen users view by scrolling. We find that most users scroll to about 50% of the page’s length. Interestingly, we’ve found there are some differences in how far down the page these users scroll on publisher pages versus product pages. For example, on average, users on publisher pages scroll to view about 40% of a page, while users on product pages scroll to view about 60% of a page (see chart).

While placing valuable content above the fold is a best practice, most users will be exposed to content halfway down the page. And users who are motivated to scroll farther are often more qualified users.
By understanding users’ scanning behavior one can make better decisions about content placement. And the answer to the question of “do they or don’t they” on a given site may be that only an advanced analytics team knows for sure.
This post on the data behind web page scrolling is taken from the Avenue A | Razorfish Global Solutions Newsletter and was written by Grace Preyapongpisan, a lead analyst, and Randy Barney who leads our advanced analytics practice — Site Optimization.










8 Responses to “Scrolling: Do They or Don’t They? A Data-Driven Analysis”
I would like to highlight another aspect of the scrolling with an example. Let\’s say I decide to buy some type of product in a webshop. Say book. Then, this is what happens:
- I will research several products.
- I will make initial selection (or rather exclusion) based on the information provided on the top, that is the most relevant/important.
- On the one/few products I will seriously consider I will look through all of the info on the page and in doing so scroll until the bottom.
So. If you analyse how many times it was scrolled down and based on this data decide to provide less info or rearrange it into different pages - you would make on incorrect decision as I benefited from the info at the bottom even though I scrolled down only one/several times.
So, the point I want to make is, I think that the most relevant measure of customer behaviour is what has he done on successful instances of interaction, if these can be identified, say bought, subscribed, replied etc.
What are the stats on users scrolling articles on scrolling?
Nice article - very interesting. Another theory on why users scroll without reading is page framing — scrolling down to cut out unwanted content like banner ads, and putting a frame around they content they are interested in. This often happens unconsciously and immediately upon arriving on the page — we see it in the lab all the time. I guess people have a natural tendency to put a border/window around the things they want to see…
Blasting the Myth of the Fold - http://www.boxesandarrows.com/view/blasting-the-myth-of
good read on the same topic.
I personally hate when i have to scroll to see everything on a page. I think simple sites are much more effective.
I think this changed when all mice started coming with a scroll wheel - around the year 2000?
We now all scroll because it is easy.
Most people hover the mouse over what they are reading, in case they want to click. Having to go to the scrollbar was a pain, you might even lose your flow.
With the scrollwheel you dont have to move to scroll, and therefore we all scroll.
I scroll vertically all the time, but that doesn\’t mean I read the content - often I scroll just to see how much I\’m going to be required to read before I both investing the time. If it\’s more than one page of scroll (below the fold), I don\’t read it unless it\’s REALLY interesting. At best, I skim, read the last paragraph, then move on.
I watch other people do the same thing all the time.
What’s this all about? I just scrolled down to this box and started typing. Joking! Well written.
I find myself scrolling downwards with the wheel for my “skimming” for the full length of the page/article to see if anything grabs my attention then I roll back up. If it’s really long I’ll use the side-bar. =)