Is your web site left or right handed?

Here is a question that nobody has ever asked in a website redesign meeting: is your website left-handed or right-handed?

Before you start wondering whether I have finally run out of things to complain about, bear with me. We spend an astonishing amount of time designing websites around assumptions about where people look, where they click, where they expect things to be and, apparently, which hand they are going to use to do it.

We just don't call them assumptions. We call them "best practice", which is generally where the trouble starts.

Because that's where navigation goes

Take navigation. We put it at the top because that's where navigation goes. Or we put it down the left because that's where navigation goes. Or we put a hamburger menu in the top right because, apparently, that's where hamburger menus go now.

Nobody remembers who decided this. There was no international standards committee that sat around a large table and voted unanimously that the little three-line menu icon should live in the top-right corner. It just happened, and then we started calling it intuitive.

The web is full of this sort of thing. Your logo goes in the top left, your search box goes in the top right, your shopping basket goes in the top right, your login link goes in the top right, your language selector goes in the top right and, for reasons that escape me, the accessibility widget usually ends up floating over the bottom right of the screen.

We've essentially developed a system where every corner of the screen has already been allocated before anyone has bothered to ask what the user actually wants to do.

Which hand are you using?

But here's the thing: which hand are you using?

If you're holding your phone in your right hand, the bottom-right corner is conveniently underneath your thumb. If you're holding it in your left hand, that same button might as well be in another postcode. Of course, you can use your other hand. Left-handed people have been doing that for years because the entire bloody world was designed by people who apparently never met a left-handed person.

Scissors are a particularly good example. Someone hands you a pair and says, "Here, have these." You use them and discover they don't work properly. "You're holding them in the wrong hand," they tell you. "Which happens to be my actual hand," you reply. At least with scissors the problem is obvious. With a website, the designer can simply assume that if you can't comfortably reach something, then you are the one using it incorrectly.

Websites are different from scissors because we don't actually know which hand someone is using. They might be holding their phone in their left hand because they're right-handed but carrying a coffee in their right. They might be holding it in their right hand because they're left-handed but carrying a toddler in their left. They might be using both hands, neither hand, a mouse, a keyboard, voice control or some combination of all of them. The important thing isn't whether somebody is officially left- or right-handed. It's what they are doing at that particular moment.

And on a desktop computer the whole thing gets even sillier. We tend to assume that the mouse is on the right because most people are right-handed. And why are most people accustomed to having the mouse on the right? Because most desks, computers and instructions have been designed around right-handed users. It's a magnificent little self-fulfilling prophecy. "We know most people use the mouse with their right hand." "Why?" "Because most people use the mouse with their right hand." Brilliant.

Meanwhile, the web industry can spend an enormous amount of time arguing about whether a button should be eight pixels or twelve pixels from something else while giving considerably less thought to whether the person clicking it can comfortably reach it in the first place.

We love our conventions

We love conventions because conventions are comforting. They mean we don't have to think too hard. The logo goes there, the navigation goes there, the call to action is a large colourful button and the footer contains four columns of links that nobody has ever clicked. Somewhere there is a newsletter signup, probably accompanied by a promise that we will "never spam you", which is interesting because nobody who intends to spam you ever puts "we intend to spam you" in the signup form.

And then there are the other sacred rules of web design. We still talk about putting things "above the fold", despite the fact that there is no fold. It's a web page. You scroll. Nobody at the newspaper stand is unfolding your website from the middle. Yet we continue to use terminology inherited from a completely different medium as if a giant printing press is going to arrive and cut our screen in half.

Then there is the equally sacred belief that important things must be on the left because people who read English start on the left. There is some genuine logic behind that convention, particularly for left-to-right languages, but it isn't a universal law of physics. People read right-to-left languages. People use different scripts. People scan rather than read. People jump directly to headings. People ignore your carefully crafted visual hierarchy because they are looking for one particular sentence that answers the question they typed into Google.

The user doesn't care about your grid. They care about finding the thing.

Users have their own ideas

Search is another good example of this disconnect between how we think people use websites and how they actually use them. 

There was an interesting discussion on the WebAIM accessibility mailing list about a site where usability testing had shown that most people preferred using search to navigate. 

Rather than forcing users to find the search field, the developers experimented with allowing people to simply start typing, with whatever they typed automatically being sent to the search field.

It turned out not to be such a simple solution from an accessibility perspective, because screen readers and keyboard users can use those keystrokes for other things, but the underlying observation is fascinating: people don't necessarily interact with the interface in the way we expect them to.

They have their own mental model of what they want to do, and sometimes the fastest way to search is simply to start typing. The whole WebAIM discussion is worth reading.

Design for the person, not the website

And that, really, is the point.

We have become very good at designing websites for what a website is supposed to look like. We are not always quite so good at designing them for what people are actually trying to do.

A website should not require somebody to use a mouse simply because that's how the designer tested it. It shouldn't require them to understand that three horizontal lines mean "menu", that a magnifying glass means "search" or that some mysterious little person-shaped icon means "your account". It shouldn't assume that everybody has perfect vision, can comfortably reach every corner of a touchscreen or will remember where you decided to put something the last time they visited.

It certainly shouldn't require the visitor to be the same sort of person as the person who designed it.

That sounds obvious, but we repeatedly design for an imaginary average human being who doesn't actually exist. There is no single typical user who arrives at your website under ideal conditions, sits upright at a desk, uses a mouse with their dominant hand, has perfect eyesight, understands every icon and patiently follows the journey you have carefully constructed for them.

There is just the person who happens to be using your website at that particular moment.

They might be left-handed. They might be right-handed. They might be using their phone one-handed because the other hand is holding a cup of tea. They might be using a mouse with their "wrong" hand. They might be using a keyboard. They might be using voice control. They might have zoomed the page to 200 percent. They might have turned the device sideways. They might have disabled JavaScript. They might even have deliberately ignored every carefully researched UX recommendation ever published on LinkedIn.

And your website still needs to work.

That's the bit we sometimes forget.

Perhaps your website should be ambidextrous

Good design isn't about finding the one perfect position for a button. It is about making the task possible regardless of the person, device or method of interaction. That doesn't mean every website needs a "Left-Handed Mode" hidden somewhere in the settings, because that would be wonderfully ironic: another setting the user has to discover, understand and configure before they can use the website comfortably.

Instead, perhaps we should design interfaces that don't care.

If something is important, make it easy to find and easy to operate. If an action can be performed in more than one way, don't arbitrarily prevent one of those ways. If a control only works comfortably when the user is holding their phone in one particular hand, perhaps question whether that is really necessary. And if the only reason something is positioned somewhere is because "that's where websites put it", then perhaps that isn't a design decision at all.

It's a habit.

And habits are very difficult to distinguish from best practice once nobody remembers why they started.

So go and look at your website. Not as the person who built it, but as someone who has never seen it before. Look at where the important things are and how much effort it takes to reach them. Look at what disappears into a hamburger menu, what is hidden behind a tiny icon and what requires a precise mouse movement or an uncomfortable thumb stretch. Then ask yourself the wonderfully awkward question: is this actually the best place for it, or is it simply the place where we've always put it?

Because maybe your website isn't left-handed. Maybe it isn't right-handed. Maybe it has simply been doing everything with the hand the web designer told it to use.

And frankly, after thirty years of websites telling users where to click, perhaps it's time we let the users decide.

Just don't put the cookie settings in a modal over the cookie settings.

That's where I draw the line.