How Simple Must You Make Your Design?

This past week showed me something very interesting. Mainstream users work with the internet in a completely different manner than early adopters. As an example, read the comments on this ReadWriteWeb post. By now, you have heard about this comment stream and how some users thought the Facebook Connect integration on ReadWriteWeb was the Facebook Login and that they were viewing a new Facebook design. I do not want to make fun of these people, but seeing a big, red, ReadWriteWeb banner should have been an obvious cue that something was wrong.

What was wrong was not that these people stumbled into the wrong site, they use the internet differently. First, these people searched for “Facebook login” on Google. That ReadWriteWeb post was first or second in the search results. Some people have blamed Google for this problem, including ReadWriteWeb themselves:

Google had completely failed its users. It put us, with a post about how an AOL partnership foreshadowed Facebook becoming the de facto user database, above the most logical search result possible – Facebook’s login page. While we mock those users, the simple fact is they haven’t necessarily failed, something failed them.

We can not blame Google for ReadWriteWeb having some excellent SEO. Obviously, they played the SEO game and won with this post. But, they do ask what failed these users?

Design has failed these users. Granted, the group of users that ended up commenting on that post are likely not the real target of any startup, but it should make you reconsider what the mainstream user really expects. I asked my wife what she thought of the situation. She thought it was comical at first until I asked her to login to Facebook. Like many mainstream users, she typed “Facebook” into the little Google search box in the browser, and the suggestions eventually listed and she clicked it. I asked her to type in “Facebook login” in the same box and she went to the Google results. In looking at the search results, she said she would not have clicked the ReadWriteWeb post because it did not say “” or “Facebook Home”. My wife is not an early adopter, but she is not a complete internet newbie either.

I realized that most people are completely dependent upon Google without really realizing it. They do not use URLs because they are too complicated for the mainstream user. They do not want to remember that Facebook is a “.com”, ProBlogger is a “.net” and BBC is a “”. Most of the time, they are not typing into the address bar, they search for their sites. If you tell them to “type the URL into the address bar of your browser”, you will likely get a confused look. If you remember that interesting experiment from a few months ago, most people do not even know what a browser is, so expecting them to remember what a URL is or the address bar is near impossible.

As engineers, early adopters and product designers, we have failed the mainstream user. How do we fix this? You may be thinking that people will just need to learn, but think about some hacker gaining some SEO juice and presenting a search result with the same title and same page layout as Facebook, just with a different URL? How many accounts will be compromised in that situation? Obviously, a lot of things would need to change in order to make all of this much simpler.

First, search results will need to be presented in a much friendlier manner. If you look at the search results for “facebook login”, you will see that it is not entirely obvious which link is really the Facebook site. At the time I wrote this, here are the top 5 search links:

  1. The actual login page, but you would need to look at the URL to determine that correctly and the title is “Login | Facebook” which is also not really intuitive.
  2. Some link on, obviously not the Facebook Login, but the title in the results is “Facebook Login”.
  3. A link to the private page for Facebook Lite, and the title is “Facebook | Login”.
  4. “News results for facebook login”, which links to the Google News search results for the same query.
  5. “Facebook Wants To Be Your One True Login”, which is the infamous ReadWriteWeb post.

Based on these, I would think that Google would need to almost completely rework their search algorithm to really fix this issue.

What does this mean for your website or startup? It means that your design must be ridiculously simple in order to really cater to the mainstream. If it does not, then you must have some significant feature, like Facebook’s connect with friends idea, to hook your users. Why do you think Facebook is continuously redesigning their site? Facebook started simple and basically had wall posts. As the user base grew, so did their feature set. As they added more features, they started to complicate the site. This is why groups are migrating to fan pages and wall posts are migrating to status updates. If “groups” are just like other users, that makes the user experience simpler. If status updates and Twitter conventions like @replies are simpler than wall posts, that increases engagement on the site.

For any website or startup, there is a balance that needs to be maintained between completeness of the feature set and the simplicity of the site design. How do you maintain that balance?

10 thoughts on “How Simple Must You Make Your Design?

  1. I found various desktop application UIs to be illustrative. Some start in a basic mode with only the most common (and obvious) features. If you need more, and you know enough about the topic to know you need more, there is a setting to enable an expert mode with a lot more features but which assumes a certain level of knowledge from the user.

    Facebook Lite seems like such a good idea, and the great majority of Facebook users would probably be quite happy with that feature set. If people want more they could enable one of several advanced UI presentations, perhaps focussing on photos or on management of a fan page.


  2. DGentry

    You are correct about some of the desktop application UIs being simpler. Facebook Lite could be a good idea as well. Simple features need to be easy to use, especially if you are targeting mainstream users. Even advanced features targeted to a much smaller set of users should still be easy to use.


  3. Wow, just wow. You know what – Google hasn’t failed them, the people thinking RWW was Facebook just aren’t very bright. Let’s not dumb things down for the average, I actually think for the most part Google’s algo gives me exactly what I need from an information standpoint (esp when using search operators).


  4. Adam

    I totally agree that Google did not fail them, but there is some question that a mainstream tool like Google should be simpler to use the results of. For you, Google is fine, as it is for me, but when you target the mainstream you do need to “dumb things down for the average”.


  5. I’m not sure I agree with your line of reasoning. There are points at which making something very simple actually changes what that something is, and so in searching for ease of use you sacrifice power.

    For example: The fact is that there are many people out there who find an iPhone simply too complicated to understand and just a want a simple phone. They will never understand the iPhone, or if they do they will do so under their own steam. Are all the other users who do get it supposed to lose access to their power features because we must serve the broadest interest possible?

    Thus, while I’m a big fan of elegant design, my attitude is that part of our job as developers is to raise the bar that people meet over time. That is how we achieve progress. Some concepts may prove too difficult for many people (like URLs, or the Dewey Decimal system) but over time use cases and understanding emerge. Some people thought that phones would never take off because of all the phone numbers that people would have to remember, but it turns out that they can adapt to a system given enough time.

    This is all interesting by the way, but why can I not log in to Facebook???


  6. Tadhg

    I know I did not make myself clear on this point, so I will try to explain. Some applications/devices are not targeted at the “entire” mainstream. The targeting does make a huge difference however. If you look at the iPhone, it was meant as a much simpler smartphone that would target techies as well as some of the mainstream looking for something better. The iPhone is not meant to be the one phone for everyone.

    Facebook is a bad example to use, and that is why I focused on the browser itself and google search. These are targeted at every internet user and need to be dumb simple, with the ability to find advanced features as well.

    Your phone number example is interesting though. You are correct in assuming that people will learn things. I am not sure how that translates to “internet time” and I will need to think about it.


  7. Sure.

    I think where we’re getting crossed wires is that I don’t actually think of those people who use Google to find their Facebook login (say) as “the mainstream”.

    The Mainstream in that context is a very television-era idea of mass audience, in which the simpler and stupider you could be, the more your advertising would spread. The concept is basically that the mainstream is the middle, everything else is the fringes, and the fringes ultimately gravitate toward or orbit that middle.

    The internet-era doesn’t really work like that though. There are many streams, peaks and valleys and they have different and changing requirements. That a site or a service (like RWW) is built to serve a particular market means design choices that favour what that market wants. That’s their focus. Changing RWW drastically so that it is simpler isn’t making it “more mainstream”. What it’s doing is changing its audience to a different (and probably uninterested) stream.

    In the internet era, gravitational pull comes from the fringes, not the middle. The middle (or middles) drift over time toward the fringes. This is why the internet changes so much from year to year with new services and sites springing up all the time, while old school sites die if they don’t keep pushing boundaries.

    Gmail, the most techie of the big webmail providers, keep growing and growing, pushing out new features carefully but constantly, and trusting that its users will get most of them. And they do, and they blog about those changes, which gets more interest in Gmail, and so more users. The result: 176m users so far apparently.


  8. Tadhg

    This is an interesting perspective, regarding the different streams. I will agree that “simplifying” RWW will not guarantee that more mainstream users will visit. This is a problem of design targets and functional targets. Obviously, RWW will not be a mainstream site due to the fact that its content (or function) is targeted at early adopters. The design target should be similar to the functional target as well.

    You definitely have an interesting perspective on this though. Thanks for sharing it and thanks for reading.


  9. For me, the whole thing boils down to this – computers are supposed to take care of the routine work for us, so that we can get to the real item that we want to do, whether it’s to code an application or share jokes with our one hundred closest friends. In my mind, the act of logging into Facebook should be something that a computer takes care of for us, without us having to even think about URLs or authorizations or whatever. In fact, I would go one step further and say that if I tell my computer that I want to play Starfleet Commander, the computer should know that this is a Facebook application and act accordingly.

    Ten years ago, a comedian could get a laugh by suddenly saying “h t t p double dot forward slash forward slash,” but while a minority of users have taken this strange language as commonplace, for many people it’s still mysterious. The item that you recently shared via Google Reader, a post from entitled We’re the Stupid Ones: Facebook, Google, and Our Failure as Developers, put it best:

    There is LOADS of anecdotal evidence that most users simply use search engines as a sort of natural language CLI. Shouldn’t we be designing interfaces that work in the way most natural for the majority of users?

    Perhaps this is optimistic thinking, but for computers to truly be useful to billions of people, our ease of use goals must be increased by several orders of magnitude.


  10. John

    I totally agree that a lot of this should be simpler. “Shouldn’t we be designing interfaces that work in the way most natural for the majority of users?” Yes, absolutely.


Comments are closed.