Snow Fail: Do Readers Really Prefer Parallax Web Design?

The first careful study of the popular new style suggests they don't; some even find it disorienting.

You may not know the term "parallax scrolling," but you've probably seen it in action. In the past couple years, parallax has become perhaps the most popular site design tool out there, embraced by commercial products and (largely thanks to "Snow Fall" from the New York Times) mainstream media alike. The effect occurs when various page elements move at different speeds, creating a sense of animation and a heightened interactive experience (examples below). It's a step away from pragmatism and functionality toward novelty and visual appeal.

Whether that step takes web design in the right or wrong direction has become a topic of considerable debate. The parallax style has excited web developers and inspired any number of hype lists. It's also triggered a backlash among critics who feel its bells-and-whistles approach detracts from actual content. Pitchfork creative director Michael Renaud recently told the Atlantic Wire he expects people to "tire" of the trend within a year or two.


What the discussion has lacked so far is much evidence from web users themselves. That's the first thing Dede Frederick noticed when he decided to study parallax scrolling for a graduate project at Purdue University. Frederick, who's currently getting a doctorate in web design, scoured the literature for studies of parallax--and found none. "It looks nice, and I like it," says Frederick of parallax, "but I couldn't find much [research] to start me off."

So Frederick set up a study of his own. He designed two hotel websites similar in every way, from content to color scheme, except that one featured parallax scrolling and one did not. Frederick then intercepted 86 people in the lobby of the Stewart Center, a main gathering point on Purdue's campus, and brought them to a computer where they engaged with either the standard site or the parallax site (screenshot below).

The test itself didn't last very long. Participants spent a few minutes getting familiar with whichever site they'd been assigned. They entered demographic information into a web form. (They all experienced a site error, too, because Frederick wanted to incorporate the unruly nature of web use.) They made a hotel reservation. Finally they rated the site on a questionnaire crafted from prior web research.

Frederick's survey focused on five areas of the user experience: usability, enjoyment, fun, satisfaction, and visual appeal. (Enjoyment and fun, while ostensibly similar metrics, differed in that a site like IMDb can be enjoyable without being fun in the way a video game site might be.) Before he tallied the results, Frederick was convinced the parallax site would blow its opponent away.

"I've read from many blogs how people say it's going to attract users and create so much of a better user experience," Frederick tells Co.Design. "I thought it was going to be superior to a typical website in every aspect."

As it happens, the parallax site was only superior in one sense--fun. None of the other survey measures indicated a significant difference in user experience between the two sites. Parallax didn't even edge the standard site in questions about visual appeal (although participants did think it looked slightly more "professional"). Frederick also discovered one critical disadvantage of parallax: test participants who suffered from motion sickness found the style disorienting.

The results left Frederick at a bit of a loss to explain the perception of superiority that surrounds parallax design. "I guess, just like me before my research, they're going from what they see," he says. (Frederick plans to submit his research to a journal for publication soon.) One reason he may have found virtually no difference in user experience between the sites is that the standard site he created was also very pleasant. Then again, if parallax were truly a game-changing tool, it should be able to elevate a site from good to great.

Future tests might go beyond the general experience of users to consider the ways they engage with a parallax site. Eye-tracking research, for instance, could help designers recognize which elements appeal to readers and which merely distract them. Content-related questions, rather than general surveys, might let researchers determine whether parallax interferes with a user's ability to process the story the design is supposed to enhance.

Sobering as this first careful study of parallax might be to web designers, Frederick still believes it's a fad with a future. He cautions developers to think more carefully about the context in which parallax is applied. Text-heavy sites that employ parallax scrolling seem more likely to disorient users, he says. Sites that emphasize visual elements--images, infographics, or data visualizations, in particular--are probably a better fit for the style.

"Right now, people have not understood much about parallax scrolling, and they're using it in all different contexts," says Frederick. "So I think, for example, in a content-based website, you're probably going to see less of it, and it's going to move to the more suitable context."

Image courtesy of Dede Frederick

[Illustration: Topform via Shutterstock]

Add New Comment

38 Comments

  • Jon Frum

    He failed to test it on a smartphone and tablet. Part of the advantage of scrolling is the multi-screen functionality.

  • I think that Frederick should have not selected hotel websites to start with. And the Snow Fall is not even just a parallax web design.

  • Dee Step

    this is not very scientific. i don't want a parallax scroll site when booking a hotel room, or buying anything, or doing any other things not 'fun' or utilitarian. Discovering art, stories, etc, parallax is great for that. Different tools for different purposes. THE END

  • Jonathan William Cutrell

    First, let me go ahead and say that Snowfall (and many other op-eds and covers) have taken on a new shape with collaborative efforts from front-end devs. We've seen a beautiful explosion over the past two or three years of interactive story-telling and narrative support that goes far and beyond the static pages of yesteryear.

    But let's get one thing clear: parallax is a horrible label for a very interesting change in user interaction patterns.

    Parallax is a term used to describe a real-life phenomenon that occurs as a result of the way we see the world - not metaphysically, but quite literally. When we have a tree in front of us and a mountain in the background, when we move to the left, the tree appears to move to the right faster than the mountain. This is because of the perspective our eyes give us.

    However, scroll-triggered animations and fixed elements in the midst of moving elements doesn't constitute "parallax".

  • branded07

    The New York Times Snow Fall site is not parallax. Designers poor application and misunderstanding of what parallax is, is what gives this effect it's bad name.

  • kylepfeeley

    The entire premise of this experiment is flawed. Surveys aren't a good predictor of whether a site is effective or not. You need to AB test these sites to see whether important metrics of user engagement on one or the other site is higher – be it conversion rate, bind rate, value, etc.

  • Walt Haas

    First, a disclosure: I'm a backcountry skier. So are many of my friends. Avalanches are a regular danger for us, and one of the things that makes them dangerous is the way groups make decisions about avalanche risk.

    The New York Times Snow Fall article was absolutely outstanding in its analysis of the process that led to the tragedy. I've read it carefully, several times, and recommended it to many people. One of the things that makes the article so valuable is the quality of the journalism; the care and thoroughness that the Times staff showed in their interviews of those involved and their reconstruction of the avalanche.

    Today as I took a group backcountry skiing in the Wasatch Mountains outside Salt Lake City, I tried to apply what I learned from Snow Fall. I asked the least experienced member of our group to fill me in on what signs of avalanches he could see in the mountains around us. He made a good effort to observe, and we talked about what he noticed and what he didn't notice. One of the clear lessons learned from Snow Fall was the importance of involving every skier in the process of observing and discussing avalanche risk.

    As a web developer myself, I paid careful attention to the effects that the New York Times used in the Snow Fall article. In general I felt that they did a good job of illuminating the chain of events and clarifying how the group dynamics and the resulting decisions turned a joyful ski day into a mass tragedy. One use of the technique you are referring to was an illustration of where each skier went, and why they went there, in terms of their own perception of the mountain and the acts of the people around them. I followed each skier's movement carefully and learned a lot from what they did and why they did it.

    The Internet is changing the way we talk about these things. For another recent example, on December 9 there was a skier-triggered avalanche just outside Alta, Utah. There is an excellent report by the rescuer here: http://utahavalanchecenter.org... That report links to a blog post by the victim and an interview with her companion, which illuminate what happened and most importantly WHY it happened. Needless to say, this represents a revolution in how we understand accidents. I'm really excited to see this development, from both the skiing and the Internet side of my life.

  • Anthony Reardon

    Snow Fall was brilliant. If you're going to study whether or not parallax is better for "readers" the case is made. However, if you are trying to answer that question by doing a study of the design applied to a hotel site, you're really comparing apples to oranges. What does the conclusion of the study have to do with how good it is for reading?

    Parallax is interesting. Draw a few pictures on a piece of paper. Cut a window on another piece of paper and superimpose. Just slide the window sheet up while pulling the picture sheet down. Now you got a fun little activity for making holiday cards you can use to teach your kids Parallax web design!

    Best, Anthony

  • Brock Gratton

    I think it just sets you a part from your competition... I've had so many clients mention how they LOVE the rockets on our site and also the lightbulb movment.. We've been implementing a lot of parallax into our client sites.

  • bfredit

    I'm sure that the clients love pictures of cats and dancing babies on their websites, that doesn't make it a good website — well, unless all you want is *your* client's immediate satisfaction rather than *your client's clients*

  • Mathieu Gosselin

    I think it depends on a lot of factor. How the parallax is implemented, how much real estate it takes etc... One study on one particular website cannot prove much.
    Depends how this serves the content indeed.
    There might be some other cases where it serves content more than others and applied more intelligently than others. I'm thinking the latest medium update for instance.

  • raykanani

    Content is definitely king. However, I wouldn't discount the point that participants in the study found the parallax site more fun. If parallax has the ability to deliver a positive emotional experience to the user, there is still significant branding value to the technique.

  • Johnny

    This just reinforces the fact that content is king.
    Period. Always has been, always will be. Although I think the
    parallax effect complimented the Snow Fall piece, it was ultimately the story that kept me scrolling.

    IMO, design should be used like the seasonings a chef uses to
    complement a recipe. So, if the fundamental recipe sucks (i.e. bad
    content) you need to pile on the seasonings (i.e. more design tricks) to make
    it more palatable.

    …So in a way, I guess you could call Parallax the Siracha of bad
    content.