We hear a lot of people tell us what their dream site looks like. “Clean,” the perennial marker of a good website that is easy to understand and beautiful. “Like Apple,” for their austere and well-loved visual and interface design. But then, there's a frequent and specific inclusion that muddies the water between metrics and emotions:Everything should be three clicks from the homepage.
What is so enamoring about three?
We hear a lot of people tell us what their dream site looks like. “Clean,” the perennial marker of a good website that is easy to understand and beautiful. “Like Apple,” for their austere and well-loved visual and interface design. But then, there's a frequent and specific inclusion that muddies the water between metrics and emotions:
Everything should be three clicks from the homepage.
Like so many common ideas about websites, this one shatters as soon as it hits reality. It's an overly simplified bellwether that hides the real complexity. Making your entire site live only three clicks from the homepage almost never happens, because there's too much stuff and it breaks the pacing of the site. The worst problem is, it's not even the number of clicks that is frustrating.
There’s too much stuff
The easiest way to have everything three clicks from the homepage is to have only a few dozen pages on your site. So simple, right? Most of the time, the sites that Fastspot works with have built up vast libraries of information that have to be categorized, labeled, and grouped into areas that are both logical and emotionally resonant. Sure, we want to throw some of it out, but all of it? Not really.
When you slap an arbitrary depth restriction on your site without drastically cutting down the content, the site flattens out and becomes broad and shallow. Everything is three clicks from the homepage, but suddenly every user has to parse huge amounts of information at every step.
We have to consider cognitive load—the difficulty of a task. Is it harder to make four small decisions or one massive one? Is it easier to climb a flight of stairs with 20 steps, or a cliff with three ledges? The challenge is in the journey, not necessarily the steps that it takes to get there.
We've got to nail the pacing
Information architecture isn't just about limiting choices, it's also about giving rewards. When you’re using a website, it gives out rewards in the sense of satisfaction that you’re finding what you’re looking for, or learning something new. If we thin out the information too much, each page becomes inconsequential. Users aren’t getting what they need, so they flit between pages getting more and more frustrated that they aren’t finding what they’re looking for. On the other hand, if we create pages that are too dense, slogging through each page starts to feel like chewing an overcooked steak. The feeling of reward evaporates as the user lingers on each page and parses reams of information. The reward of finding what you're looking for disappears into the chore of sifting through the text.
Lopping off the page depth at “three clicks” becomes an arbitrary rule that distracts from the real challenge of determining what matters, and how much you dish out at each step. It distracts from elevating specific tasks and helpful goals by acting like we don't need to prioritize.
It’s not the number of clicks that’s frustrating you
When you are looking for something specific, how do you find it? In our work with higher education clients, we hear a lot from current students about how they find information on the website, and the usual answer is, "Well, I just type what I'm looking for into Google." This isn't an obscure pattern for website users—when they know the precise thing they're looking for, they default to "Thing that returns what I'm looking for" not "Thing that shows me all of my options."
Now this doesn't mean that we're recommending that you throw out your IA entirely. (It can be done, but requires specific consideration to ensure that you aren't leaving your browsing users out in the cold.) When a user browses through a site’s navigation, they use it to find the things that they don't already know about, to explore the space, and to understand the vocabulary they share with you.
Many clicks don't make frustrating journeys, but frustrating journeys have many clicks.
The challenge is in optimizing that exploratory process, in showing the state of the site accurately to the user, and informing them that they're moving in the right direction. Fewer clicks should instead be prioritized by task frequency or urgency, exposing pages and resources that are needed quickly or often.
This isn’t just a guess though. A usability study by UIE showed that click path and task satisfaction have a very weak correlation—longer paths didn’t frustrate users more. NNGroup in a study published in Prioritizing Web Usability found “users’ ability to find products on an e-commerce site increased by 600 percent after the design was changed so that products were four clicks from the homepage instead of three.” (This doesn’t mean that everything should be four clicks from the homepage, though!)
So what should we do instead?
Number of clicks isn't a helpful metric to ensure that users are able to get through your site quickly or easily—so how can we replace it? How can we ensure that we are creating something that is usable and frictionless for our site's users?
There are very few metrics that are a universal good, outside of maybe "reduced bounce rate" or "the site accurately represents its institution." In order to figure out if your site is improving a user's experience, you first have to know which experiences they're looking for. Here are a few questions you could ask instead:
- What is the primary reason users will come to this site? Can they do that quickly? If you have a physical location, usually that means get them there. Show the user how to get involved. Make it simple for them to take that next step.
- Where are people going currently? Can they get there quickly and simply? Your analytics can clue you into this a little bit by telling you the pages that receive the most traffic and the paths that users take to get there. Take it with a grain of salt, this sort of web analytics is good at showing what, not why. (Example: Is everyone going to this page because they need to, or because we have tons of links that drive to it?)
- When you ask a user to navigate somewhere, can they get there? Do they feel like they're moving in the right direction? This is where user testing really shines over analytics, because you can hear what users are thinking about a particular data structure. They shouldn't just guess correctly, they should also feel like they guessed correctly. Success here isn't just a “yes” or “no,” the context around the “yes” is important.
On the big flat sites
At Fastspot, one of our most commonly referenced sites is the redesign for Bucknell University, which dispensed with a traditional information architecture in favor of a navigational trio of a directory, a persona/task based navigation, and a history utility. The site heavily uses on-page navigation and visual features to move users between pages.
Ostensibly, this shortens the click paths, right? A user can quickly get into the directory and then to their relevant page. Sounds like a win for the three-clickers.
However, if this was truly the case, then Bucknell would have never needed the persona and task navigations. We didn't just flatten out the site, we also had to give thoughtful tools to re-elevate the users’ thought processes and tasks, and ensure that they could still find what they were looking for. The answer wasn’t just providing a search box.
Sorry about the silver bullets
The bad news is that there’s no simple way to ensure your site is usable. Sad trombone. The good news is that means your site doesn’t need to obey a bunch of arcane rules just because, but instead can adapt to your—and your user’s—needs.
Start by understanding what users are looking for, and make it easier to get there. Follow it up by structuring the other information sensibly, and giving clear paths through the complexity. Verify it by testing it with real people, and by checking it out yourself.
Don’t design for clicks, design for people.