November 26, 2019

What Makes New Features Unpopular

When Software Breaks Your Brain

by @maguay

Writer. Amateur photographer. Information architect. Frm. founding editor @racket and @capiche, senior writer @zapier, Mac and Web editor @AppStorm. Personal blog: @techinch.
maguay's avatar

When your car has an odd noise at a specific speed in inclement weather, it’s frustrating yet forgettable. When the car won’t start in the morning, though—or, perhaps worse, when it randomly won’t start even after repairs for seemingly no reason—it’s maddening. Something that used to work doesn’t, and your formerly forgettable routine now requires more thought.

Software does this routinely without actually being broken. Things that worked yesterday break overnight with updates that promise the new, the better, the future. We’re unlikely to complain that loudly over a new icon, different colors, or genuinely new features. But break our routine, make something that formerly worked without fail suddenly fail, and we’ll be frustrated at best, ready to switch products at worst.

Just when something became second nature and we quit having to think about how to use a product, updates force us to learn how to use something again. We think our computer broke, that we’re suddenly suffering from keyboard shortcut amnesia, then realize no, it’s the software that changed without telling us. That’s when we complain.

When Slack broke markdown.

Perhaps the most ingrained computing workflow is copy/paste, with Ctrl/Cmd+C then Ctrl/Cmd+V being one of the first keyboard shortcuts we learned. And when you copy something and paste it, you expect your data to be the same in its new location. That’s how it’s always worked.

Then Slack broke that assumption, and the pitchforks came out. Where before you could copy code or text and paste them into Slack without problem, or format text with Textile/Markdown-style characters, the new editor replaced formatting characters with actual formatting, breaking code when pasted it into Slack.

You’d paste text, hit enter to send the message, then frustratedly see that sample code was broken and your formatting was all messed up. What was a reliable task that didn’t take thought turned into something that might break your work. Perhaps worst of all, when you do a task often enough, it becomes muscle memory. You’re likely to accidentally mess up Slack formatting again and again until, painfully, you re-learn how the new version works.

That’s what makes people dread updates. It’s not that they’re luddites afraid of change. It’s that they don’t want to waste time and energy relearning how things work.

Updates used to be optional, to a degree at least, especially when they changed things. Windows XP and Office 2003 each got updates, but typically to fix bugs. The major changes came with Windows Vista and Office 2007, each with entirely new designs—shipped as new products that you could choose to skip.

Today’s rapid development sees web and mobile apps get incremental updates without notice—until they break something. That’s when we lose our minds.

Slack, to their credit, has begun walking back the changes on public demand (though not before someone made a Chrome extension just to disable Slack formatting). If you paste text with Markdown formatting into Slack today, it now asks if you want to apply formatting—with a new Cmd/Ctrl+Shift+F shortcut to add formatting if you want. It helps—but for a while, it'll be hard not to double-check every time you add something to Slack.

Breaking changes.

The worst changes break things we already know how to do.

Slack did it with its formatting changes. Microsoft Office did it with the Ribbon design in Office 2007, and Windows 8 took it to the extreme by turning the traditional Windows desktop into a tablet interface. Photoshop did it by changing keyboard shortcuts for scaling images proportionally and undoing previous changes. WordPress it with its new Gutenberg editor that breaks articles into sections and makes you click multiple times to add new “blocks” to insert quotes, lists, images, and more-something that used to take a single click or a keyboard shortcut.

It’s not that you can’t do what you could do before, but that the old method doesn’t work, and the new methods require learning new tricks. Worse, the new methods often take more time, as with WordPress where tasks that formerly took no clicks now require two or more.

Different, but better.

The frustrating-but-toloratable changes break things, but for a good reason.

Photoshop’s new Undo shortcut finally works the same as every other app—but after decades of it working a different way, the changes were an unwelcome break in many designers’ workflows. Mac OS X Lion’s swap of scrolling direction was similar, where you push the virtual page up instead of scrolling down. It makes more sense especially on trackpads and touch devices, but broke decades of muscle memory.

These changes make software more approachable for new users, and may simplify existing users’ work in the long term. But we all lose productivity as we learn the new way—and might be tempted to check out the competition, as half the trouble in switching software is in re-learning new habits.

Wait, where was that again?

Then there are changes that simply add annoyance and friction.

Twitter and Instagram’s switch to algorithmic-driven feeds lost our sense of direction and place, made us not trust that we could click the Back button to see something we’d seen seconds ago. Such changes are surprisingly common for social networks, even though they often bring disastrous results. Digg’s redesign cost it 26% of its traffic in 2010, while Snapchat’s 2018 redesign led 3 million people to quit using the app and 1.2 million more to sign a petition to bring back the old design.

Websites that break the back button, web apps where links don’t load the page you think they would, and menus that move things around for no apparent reason all fall in this category. They’re like Google Maps continuously saying it’s recalculating your route. While they’re changes that might not break your work, they at best make you wonder why the older design wasn’t enough, and at worst add frustration and make you feel like you’re fighting your computer.

Losing our trust in tech.

Annoyances add up, and over time break our trust in software.

When Dropbox redesigned its app and auto-launched window to browse through files, it broke the trust many had with the company. We expect Dropbox to work inside Explorer and Finder silently, with a reassuring Dropbox icon in our taskbar keeping everything synced. We don’t expect to be greeted with a Dropbox window that feels like some new, potentially malicious, app got installed on our computers without our consent. It’s the app version of pop-up ads from the web of years gone by.

To a far larger degree, Facebook’s original launch of their news feed felt like a breach of trust. We put stuff on our Facebook Walls expecting it to be our place on the internet, something people could see only if they chose to look at our page. Then Facebook put our status updates together, and we never felt like we had the same control over Facebook again. When a few years later Facebook decided to show things we’d bought-and quickly walked that back after backlash-it only exacerbated the feeling.

We trust software to work the way we have learned, to do what we asked it to do and no more. When apps send notifications and open windows that we don’t expect, they break that trust. When they use our data in new and unexpected ways, we’re less likely to trust them again next time. And little by little, software feels more annoying than helpful.

First do no harm.

The old solution was to stick with older versions of software until you had time to learn the new version. That’s a rare luxury in today’s fast-paced world, where changes come without option or warning. Instead, we need software that doesn’t assume we want changes.

Software at a minimum should try to maintain backwards compatibility with workflows—keep keyboard shortcuts and core functionality the same while adding new, better features.

When that’s not possible, at least provide a fallback option to the older version. Give us a way to turn off the new features and revert to the older way of working.

Either way, explain. Don’t show us a new window we wouldn’t expect to see normally, or move around buttons to force us to accidentally discover a new feature. Tell us about it, give us the option to use the new, and make the change as painless as possible if they’re necessary.

Tools should be reliable, dependable, trustworthy. They should work when you use them, without fail. When things change without our permission, it slowly erodes our trust in tech.

Updates aren’t bad. Breaking our workflow is.

What was the worst software update? Share yours.

Click here to Tweet this essay.

Image Credit: Header photo by Adri Tormo via Unsplash.

What's the best way to get feedback about a design or product?

Do you have a favorite survey or poll tool to get feedback, or a unique process to get actionable insights from your users about your design, product features, pricing, and more? There was an inte...

What's the best tool for building a customized API documentation?

Let's say you have an API you want to document and you want to: - Let your users test their calls to the API thanks to a "playground" like the one on the Figma API - Be able to customize it to matc...

The community for power users.