Co.Design

Why The Pull-To-Refresh Gesture Must Die

Last week, at a company event in New York City, Instagram unveiled Direct, a private-messaging feature for its popular photo-sharing platform. The press was quick to compare it to competing social services like SnapChat and WhatsApp (consensus: it’s different though trending in the same direction), but there was one subtle update to the app that many overlooked: Instagram introduced pull-to-refresh functionality, a decision not even cofounder Kevin Systrom is satisfied with. Systrom feels the gesture, which enables mobile users to refresh their photo feeds with a simple tug of the thumb, is a superfluous addition to his app, a relic of another smartphone era. “I don’t believe there should be refresh buttons,” he says.


A relic of a bygone era

The pull-to-refresh gesture is far from new. An endless number of apps use the interaction for in-app content updates, including Apple's native email client, social apps like Twitter and Foursquare, and most news applications such as The New York Times. But years after designer Loren Brichter first dreamed up the novel interaction to get rid of UI clutter, Systrom and others now feel it’s outmoded--an unnecessary extra step. After all, smartphones are fast and strong enough to auto-refresh. And Systrom says moving beyond refresh buttons and actions will give mobile services a more “real-time” feel.

In earlier versions of Instagram, the app featured a button that allowed users to refresh the images displayed in their feeds. Now, the button is gone--replaced by an Instagram Direct inbox icon--and the Instagram team moved to the pull-to-refresh paradigm. “We introduced pull-to-refresh, so now when you pull on your feed, it just refreshes,” Systrom says. “[But] I’d like [to get to] a day when you didn’t have a refresh button--where it just updates [automatically].”

The issue is that the gesture is so universal now that it's hard for developers to put it out to pasture. Users expect it to be part of the app experience--expect that the app will essentially be paused until they choose to click or pull to play. But Systrom believes we'll get past that paradigm--eventually. “We’re moving in that direction, but [with] baby steps,” he says.

He’s not alone in that thinking. Even Loren Brichter, the former Apple and Twitter designer, now agrees. When I connected with him recently, he indicated that the pull-to-refresh concept was designed for a different time. “The whole idea of manually refreshing anything is kind of a stupid idea,” Brichter told me--why even create that superficial layer between users and their content?


Simply put, it's an arbitrary feature. When you head to Gmail in the browser on your laptop, for example, the service will show you the latest emails automatically--and keep your inbox up-to-date in the background. So why would email clients on your mobile devices act any differently? Why would they wait for you to manually refresh them for new content? It’s an unnecessary step, especially in the age of fast LTE connections. Imagine if you had to pull-to-refresh your SMS service in order to receive new texts. Users would revolt!


Pull-to-action

Admittedly, there are benefits to the pull-to-fresh gesture. For one, some users like landing in a familiar place. When you open Twitter, for example, the app brings you to the last read tweet, providing context and continuity; if users want to load new tweets manually by pull-to-refresh, they’ll know to scroll up for new tweets or down for older ones. The app won’t rocket you to an unfamiliar spot in your Twitter timeline, in other words. It also saves on bandwidth for data-conscious customers.

Brichter, however, feels that it's high time his gesture evolves. “The fact that people still call it ‘pull-to-refresh’ bothers me--using it just for refreshing is limiting and makes it obsolete,” he says. “I like the idea of ‘pull-to-do-action.’”

It's a promising idea, and we are already seeing some impressive and modern implementations of pull-to-action gestures. When users open Jawbone’s Up app, for example, it automatically syncs and updates with the company’s companion activity tracking wristband, UP24. So now, when users pull-to-refresh, Jawbone instead displays a bite-size, digestible summary of what data was just transferred when the program synced. Call it pull-to-review.

And on Apple's home screen in iOS 7, pulling down now gives now users access to a search box--a tweak that allowed Apple to get rid of the entire search screen from previous versions of iOS. It's basically turned pull-to-refresh into a pull-to-search gesture.

The idea here is that innovation is finally coming to vertical swipes. We've already seen apps take advantage of horizontal swipes to introduce invisible interactions--you can swipe sideways in Mailbox, for example, to seamlessly archive emails. We need that same evolution to come to vertical interactions (beyond complicated edge gestures).

Of course, there is the risk of usability fragmentation here. If apps move to the pull-to-action paradigm, it might make it difficult for users to predict what action pulling on an app screen will perform--say, if Instagram featured a pull-to-camera gesture and Foursquare had pull-to-check-in.

But it's certainly worth experimenting with and moving past boring pull-to-refresh gestures, a massively untapped resource that could give way to a new style of app interactions.

Add New Comment

45 Comments

  • Not every user trusts that the app has the most up to date content. Providing a way to refresh can give a user a means to confirm this -- aka giving them control.

    Auto-refresh can fail and if it does the user must wait until the next refresh to get new content. Manual refresh can overcome this.

    Gmail includes a Refresh button in their web client.

    Automatic and a manual refresh should be combined for the best experience.

  • Usain Xeno

    Just because something is possible doesn't make it a good idea. People like to feel in control. Pick your battles and contribute to cross-app consistency.

  • Ironically I have to press the "Load More Comments" button on this page multiple times to read all the comments instead of the page automatically loading more comments once I reach the bottom.

  • Wes Jones

    I can't get behind this article's argument.

    As a person who lives in a foreign country with slow and unreliable internet, refresh buttons are crucial to my perception of control. I hate seeing the spinning pinwheel and not knowing what's happening.

  • “Boring” is never a reason to get rid of something. “Boring” is related to predictable behavior—which is beloved by users yet is anathema to “designers” everywhere. We need less designing for the media, gatekeepers, and other designers, and more designing for users.

  • Victor Kzam

    It's a great idea! I think it should be improved like Tweetbot 3, which has the pull to refresh button as an alternative to users, but it streams new content via Background App Refresh on iOS 7 and it stills on the same tweet, the new ones are displayed with a marker.
    That's the best solution I've seen yet

  • I actually thought this brought my irritation with the pull to refresh gesture into clearer focus, and I hope it was deliberate on the part of the authors.

  • Jason Burns

    I don't hear anything about data usage. While I can appreciate that most apps don't use a great deal of data to update, not all users understand this. And if a user with a limited data plan taps on an app unintentionally. Meaning to open the camera, or set an alarm, and instead hitting an app that suddenly begins downloading. This could quickly become frustrating. This really could be setting that users can choose, "Check the box to auto refresh". It really doesn't have to be an all-or-nothing situation.

  • Jeremy Stewart

    "When you head to Gmail in the browser on your laptop, for example, the service will show you the latest emails automatically--and keep your inbox up-to-date in the background. So why would email clients on your mobile devices act any differently?"

    First thing that comes to mind is battery life. How is something that important overlooked here?

  • Chris Kelly

    When I worked as a designer with with data hungry apps, there was huge pressure not to auto refresh, particularly if the user wasnt actually looking at the app. Pull/refresh gives me the user control, it makes me feel like Im in the driving seat. Its also a deeply satisfying movement where motion has lead the way. So no, in the rush tp be the newest of the new, stuff that actually works shouldnt be thrown to one side because its 45 seconds old. Its everywhere because its great and it works so step back and leave it alone

  • Alexander Gee

    The idea that devices will have the network connectivity, and spare battery power to sit refreshing data in real time is nice. It isn't terribly practical though.

  • Scott Jenson

    I strongly agree with this article. While I'm sympathetic to the 'dont use up all my data' argument, that can't be the ONLY driving force going forward can it? DO we really think that saving 10k of data is going still be a critical question in 2-3 years? That is thinking like an engineer, not a designer. While I never want to encourage wasteful data usage, this issue is clearly a subtle one, laid across a spectrum: it *is* possible to auto refresh without insisting that the user do it every single time.

    Take my new twitter app, Tweedle. It requires that I pull to refresh even when I first open the app, a refresh when I open the app should just happen automatically. Making the user do it manually every single time is masochistic.

    For those in this thread that are worried about auto updating content screwing up your reading experience, that is confusing updating with presentation. The official twitter app does this quite well: it auto updates but doesn't scroll so you don't even know. When you scroll up, the data is already there. That seems like a much more elegant solution.

    This should't be a flame war discussion of yes/no but more nuanced, trying to find out how as can we make things better for the user. Clinging to the past is never a useful (or insightful) approach.

  • Chris Kelly

    No one is looking at the design of the car steering wheel to be changed…because it works. I bet your car has one, and no, it doesnt mean you are clinging to the past. The same applies here. While we still live in a mobile world of feeds, pull/refresh is the way to update it. If you remove it, every drop in data connection will leave a dirty red alert in the UI that will make your app look like its broken.

  • Scott Jenson

    I'm surprised you are willing to standardize on something that is, in your own words, 45 seconds old. I won't deny your main points however, pull to refresh is satisfying (when done well) and it does save bandwidth so we are, for the most part in agreement with the benefits. It just feels horribly fatalistic to say 'that's it, we're done, this can NEVER be improved'. This is especially evident when I see apps use this gesture poorly, refusing to *ever* auto refresh making the user use the gesture frequently. That isn't elegant, that's lazy and certainly not good UX design.

  • timbearcub

    Very strange to call for the death of a feature that not only users like, but also accepts that unlike those in the bubble of HappyFunNewMediaValley and not on Art Director or even Creative salaries or in parts of the world with crap connectivity might not want background refresh?

    For those of us who either don't want to pay £20-30 a month unlimited contract or can't get the credit or afford that, it's all to easy to go over your allowances on PAYG etc.

    Also a suprising amount of apps don't have 'refresh only on Wifi' settings either, so you have to set the frequency to low and hope, or manual and then you need a way of refreshing...sure try different ways but pull to refresh is popular, at least among app developers partly cos the patent owner has said they won't enforce it. But I'm guessing if the app users positively HATED it then it would have died out years ago.

    But don't let that stop you from enforcing your 'vision' on users and take functionality away from users, I mean it will further help with the perception that designers (of which I am one, old skool since '96) just want to mess users around so it looks clean. I mean it worked so well for Google, Apple and Sony recently, no backlash at all *cough*

    No I was taught that the best design is the union of form vs function...sadly I see very little of that nowadays, it's one way or the other, never a happy union...and why I see pull to refresh as one of the genuinely new and cross-app developments to come along, one that *shock horror* unifies the user experience. But I guess the idea of Apple's old HCMI guidelines and the idea of not making the user relearn the interface is old fashioned, eh?