Want to know something that's costing you visitors, money and time, but is easily preventable?
It's the myths your marketing team (and maybe even people higher up the chain of command) have gobbled up about site design and the user experience, believing them simply because somebody with a little bit of fame said them, and everybody just went “Yeah, that sounds about right!”
User testing is not very scientific
The problem is, these claims aren't subjected to the most rigorous form of scientific testing, and the people doing the checking aren't normally real scientists. There can be exceptions, but even then it's not perfect.
There's this pesky thing called “response bias“, where a respondent will answer questions with what they think you want to hear instead of what they really feel, because deep down they want you to like them. That's also the reason my brother is taking diving lessons.
Even in those situations where the researchers take a strictly observational stance, there's still a problem, because they're making assumptions. Did the user skim past all that information because it was boring, or because it was something they were already familiar with, or perhaps it just wasn't relevant to them personally?
A random audience is useless for testing niche content
A farmer might read political content because it might be relevant to how he earns his living, but will an article about interior decorating hold his interest, however well it is written? Generally speaking, only if that farmer has a genuine passion for interior decoration.
An actual decorator or somebody who is in the process of buying a new home, however, could be riveted by the same content. KissMetrics says: “Aggregate data is kinda worthless,” and indeed they're right.
User testing must be targeted, and it hardly ever is
These problems and many others contribute to misunderstandings. The difficulty is that it's next to impossible for anyone to assemble sufficient numbers of people within a specific niche and demographic for usability testing. So they either use random sampling, or they cheat, and some of them just flat-out lie.
The bottom line is you can't rely on usability testing because it's too generalizing. It doesn't look at the typical user of your site, it just looks at the average man in the street, and that's a much bigger problem than it's given credit for.
Testing also doesn't take into account that individuals are variable. There may be times when I enjoy the company of others, but other times I'd prefer to be alone. If you ask me, the answer will depend to some extent on how I'm feeling at that exact moment, and whether or not we're in a bathroom.
So what are the main myths that are potentially costing us? Let's take a look at a few examples.
1. There's a perfect template or formula for a website
If that was true, the definition of perfection wouldn't keep changing. Carousels and infinite scroll were marvels to behold the first few times they were used. Now people are fed up to the back teeth with them.
Your infinite scroll may hold my interest for a while, but once the inevitable (yes, inevitable) browser memory leak kicks in, I'm going to hate you with a great deal of intensity, and I may never visit your site again. The more graphical your content is on an infinitely scrolling page, the faster the browser will max out. I'm looking at you, Tumblr.
So, anyway, this myth is one that clueless marketing desk jockeys and lazy web developers like to perpetuate (and some actually believe it themselves), but there's not a grain of truth to it.
They just want the boss to like their 1-3-2-3-1 design, which opens with a carousel wrapped inside a jumbotron div, followed by 3 dutiful columns with an image and a short blurb (complete with “more” link), and so on. There's nothing surprising there, and that's actually part of the problem.
If a user visits 10 different pages and they're all pretty much the same, it's not going to increase the chance that they'll buy from you. It will just switch them off. People thrive on variety and they don't want everything to be the same.
You don't want to be too different, just be different where it counts – give them great information and a bit of eye candy. They'll love you.
2. You need to write everything at a 4th grade level
This may be true if your primary demographic is either totally foreign, or mostly 4th graders. The assumption that people are basically not very smart is actually insulting. Nobody is going to give up on a site because they don't know a word or two.
Besides, if you believe the third myth, you're not expecting them to read it anyway. It's easy to understand where this myth originated from.
The pool of people available for usability testing is too diverse, except in one special way. It is composed of people who have time to participate in usability testing. Last time I checked, people with important jobs don't generally have time to waste on things like this, and the important jobs tend to go to the smartest people.
What this means, as you've no doubt already figured out, is that many of those who have time to participate in studies are not the sharpest tools in the shed. There are perhaps not a full quota of kangaroos bounding around in the top paddock. The lamp may be lit, but dimly. You can see where this is going, right?
It's no surprise then, that a fair few of these people will have complaints about high level language that is beyond the limits of their vocabulary.
High level language is efficient. It lets you communicate more meaning with less words and less ambiguity. It doesn't mean you slide all the way over into academic language, because that would be equally silly and disastrous. You need content with balanced language, like the content of this article.
If people don't know a word, there are plug-ins that can define the word for them in an instant. They've just got to be smart enough to figure out how to install those plug-ins.
Write your content in natural language, trying to be as clear as possible. You don't have to dumb it down, and you certainly shouldn't smarten it up. Just make it natural everyday language like you'd ordinarily speak.
3. People don't bother to read long form content
This just isn't true. You don't need to keep everything short. What you need to do is keep everything interesting. And even more importantly, provide all the information they need. Because that is the reason they arrived at your site, and that is the reason why they may choose to do business with you instead of somebody else. Who am I going to trust? The company that hides all the data and refuses to tell me the price unless I give up my email address? Or the guys who put all their cards on the table?
And while we're at it, let's stop hiding content behind “more” links, stop paginating things unnecessarily, and avoid turning everything into a slide show. None of these tactics create a good UX. They're bad for accessibility, bad for usability, and they're just downright annoying.
Show things. If you're hiding anything, it better be for a good reason. The bottom line is, if people aren't reading you're content, it's because you're boring, not because there's too much of it.
4. Video content rules (and video backgrounds are awesome)
Complete myth. Video content is useful where it's appropriate, but shouldn't be inserted just for its own sake. If your video adds something to the user experience by providing entertainment or information (preferably both), it's worth including. Otherwise it has no place on a web page.
Autoplay video is the most obnoxious thing you can do, and it's even worse when the autoplay content is an ad. Video backgrounds can look really cool too, but are you using one because it's gratifying to you or gratifying to me? Let's remember who is important in the website transaction. It's the user, right?
So you should use video backgrounds sparingly, in a genuinely awesome way, and with great consideration for the user's bandwidth. It should always be possible to disable or skip the video, or view a version of the site that doesn't include it.
Video content also doesn't do much to attract visitors at the moment unless you include captions (not auto-generated). There's a chance Google will read your subtitle file and add points to your index. That might still be a little way in the future. But you'll get points for accessibility anyhow.
5. It's OK to let Google translate your web pages for you
No, it's not OK. You can't let any automated service handle your translations if you're a professional business. Don't be cheap. Spring for a real live human being to translate your site for you. An unprofessional translation might be worse than no translation at all.
6. Everyone loves infinite scroll
Really, infinite scroll can keep somebody on your page for longer, and you might think that's a good thing. In a way it can be, but there's a big problem with it, especially if you have a lot of images.
Browsers like Chrome create a new CPU process and new memory space for every tab that's open simultaneously. As the user scrolls more and more content on your page, its tab process and memory space is going to grow larger, until it reaches a critical point where the user's system slows down.
At this point, if the user is lucky, they'll be quick enough to close your page that it won't crash the browser or the entire system. They won't always be so fortunate, and what they're going to remember is that it was your site that caused the trouble for them.
First impressions count, but so do last impressions. Don't let the last impression a visitor has of you be that you're annoying.
What you can take from all this
Users don't fit into neat little stereotypes the way marketing “experts” want to believe they do. People are all individuals, and when we design sites for them (and it's always for them, never for the people who hired you to design it), we need to respect them as individuals.
A good host doesn't get in the way of the visitor, doesn't interfere in what the visitor wants to do (as long as it's not going to cause any harm), and doesn't hide information from the visitor. Those are traits of a bad host.
As a good host, you'll want to be open, unobtrusive, and helpful. You'll guide the user without forcing their hand, and you'll make things easy for them without being condescending.
Do these things properly, and your website stands a good chance of being different enough to attract attention and convert that attention into action.
If you still haven't quite got it squared away, the tips on this article from Oprah's website apply just as much to hosting a website as they do to hosting guests in your home.
header image courtesy of