Whenever I design something, I keep in mind that it should be easy to access, but I feel that I am not creating something new. So is this normal or should I always design something which is new or which changes user behavior completely?

Like Tinder came up with swiping cards, Facebook with likes, etc. All made a huge impact.

So should I design by keeping in mind that it should be easy for the user or should I create something new which changes user behavior?

Example: I've had a few people tell me that LIKE is being used in all apps and websites....why not try something new like hi-five and/or other stuff?

So what is the main thing that I should keep in mind while designing?

  • Everything should be very easy for the users. It should be efficient - no unnecessary actions. Commented Oct 25, 2016 at 8:08
  • What is "hifi ant"? Commented Oct 26, 2016 at 13:37
  • LOUD AND RESOUNDING YES. Otherwise we'd still painting a bison in a cave. It's the sheer essence of advance and improvement. However, if you fail at making something new, at least make sure to master the established conventions
    – Devin
    Commented Oct 26, 2016 at 16:57
  • @devin - Does it always need to be new though? You're saying that only one bison painting should have ever existed, after which something new should have been done; only one impressionist style painting should have been; Jackson Pollock should have only made one painting, because the rest weren't "new". I don't think that is what you intended. Commented Oct 26, 2016 at 17:25
  • no. what I mean is that after the bison you could add a hunter or two. Maybe a moon. Or at least try to make that bison the best possible bison ever. What I mean is that you should ALWAYS strive for new. If not possible, then strive to do your best using existing conventions. All the answers below are more or less on the "stay put" side, I simply don't believe in that, hence why I made this a comment. Either way, this question should be closed since it's primarily opinion based.
    – Devin
    Commented Oct 26, 2016 at 17:39

6 Answers 6


Innovation for the sake of innovation alone is not good design.

Tinder came up with the swipe gesture as a solution to their unique interaction problem, Facebook too added the "Like" button to solve the problem of users wanting to give an immediate but brief reaction to a friend's post.

Incidentally, 'Like' is now a paradigm that most internet/app users understand so it makes sense to utilise that in opposition to anything new that your users will have to learn.

Novelty in design doesn't necessarily need to mean new paradigms - it could just mean new combinations of existing paradigms.

While number 1 of Deiter Rams 10 Principals for Good Design states that 'Good design is innovative', 4 states that 'Good design makes a product understandable', 5 states that 'Good design is unobtrusive', and 10 states that 'Good design is as little design as possible' - in other words; keep it simple and don't get in the way of the user.

Changing 'Like' for 'Hi Five' would mean your users would need to learn the function of a 'Hi Five' before they became comfortable with it. They already know and understand the function of 'Like' so why would you want to make that difficult for them?

Following on from that, if 'Like' doesn't satisfy the kind of functionality that you require, then you may need to change it to something else. At this point it's wise to find out as much about your users as you can so that you can engineer a solution that would be readily understood by them. Build up persona profiles of your key user types, find some real people who match those personae and get them to do task completion tests with a few different designs to see which one performs best.

I have to repeat my first statement again here: Innovation for the sake of innovation alone is not good design.


To the contrary, you should actively avoid making new paradigms for user interaction.

You want users to easily understand how your design works, and this means using the familiar whenever it is practicable.

Only do something new if you have a clear, compelling reason. Such as:

  • Existing paradigms do not meet your needs.
  • You have come up with a better way of doing something (and it is "better" enough to outweigh the additional learning curve).

This will vary based on context, such as how committed your users are to your software, what their level of technical expertise is, and so on.


Design is a very personal process. Use of the word design often confers something new has taken place. If you have it in you to make something new, and you believe it to be better than alternatives, then yes, do NEW!

If that turns out to be something you're constantly capable of, then BRAVO!

The entire purpose of designing something is to do it better than it's been done before, within whatever limitations you perceive and/or are subjected. This means newness is often coincidental to (and resultant of) design.

If it hasn't been done before, then what you do is new, too. So it might just be incidental to your creative and productive processes. Don't fear newness, fear wrongness.

New is not always wrong and old is not always right.

And therein lays the only caveat; if you're going to make something new, make sure it's better than what's gone before (and definitely not worse) in accordance with the project's (and your own) perceptions, intentions, beliefs and limitations.

It might just happen that your designs are always new. And that's not necessarily a bad thing, even if it's newness for the sake of newness that discovers a better way to do things.

  • completely agree +1
    – Devin
    Commented Oct 26, 2016 at 18:36

I will try to simplify the answer

In some cases, as @Andrew Martin said, you are able to create custom patterns that better fulfill the problem you are facing.

The top criteria, that tells whether you should or should not design a new pattern:

1. Do you really need it?

Have you done your homework? research about existing patterns that could solve your problem, if you find it, stick to it.

2. Is your user willing to learn?

If your users need your product enough to make them learn something new, then you have a better chance.

However, in most cases, we recommend sticking to existing and common Design Patterns that are already in use out there.

Creating a new Design Pattern requires user learning, no matter how simple the pattern is, users tend to be passive and avoid learning.

Good Luck


No, design does not need to always be original.

Many designers would rather attempt reinventing the wheel than to adapt conventional user interface design patterns. It should be considered, however, that such design conventions are well-working because they’ve already been introduced and tested for usability. Since the users know them well, you don’t need any explanation or instruction manual. As users appreciate usability over novelties, standard patterns will eventually benefit your audience.

It might occur that a new approach is needed, but you must be 100% positive that your solution is better than the existing pattern.

Source: UX Myths, Myth #9: Design has to be original

The above article cites several sources for further reading that go into further detail:

  • Steve Krug claims in Don’t make me think that “conventions are your friends” and are essential for users when going from one site to the next.
  • In How to create a good enough website Seth Godin advises that “there are more than a billion pages on the web” to get inspiration.
  • According to Web Design from Scratch, not forcing originality at any price makes the design process “quicker, easier, and more profitable” and “benefits your user”.
  • Carsonified’s Think Vitamin blog points out that “the great design solution you seek is probably already out there”.

well, since everybody is giving their opinions, here is mine: always strive for new. Then test it. This is basic UX, you should test everything.

As for the new vs conventional part, let me explain this with a very common example: music.

In music, you have composers/creators, you have interpreters and you have a mix of both.

A composer may or may not be a good interpreter, but he creates, therefore you could become The Beatles.

An interpreter strives to be great at the interpretation of its instrument, but she doesn't create. Hence, she might be regarded as a great player, but she will never be the Beatles.

Most of us will never be The Beatles, neither amazing interpreters. So as you can see, both are needed, and both are amazing at its own field.

In design (whatever kind of design you want to mention), if you create something new, you might lead the pack. If you take the philosophy of ALWAYS strive for new, then you'll probably create something meaningful down the road, maybe revolutionize the world, who knows?

But if you can't or don't want to be innovative... the world always needs good interpreters, there's not shame in it, being an amazing designer that never creates anything but it's spectacular at working with convention and rules is something great. OK, you won't create the new Facebook or the new iPhone. But you'll be amazing at implementing things.

However... testing is everything. This is something the musician and the designers know, no matter if creators or interpreters: there are ways to measure the success of an idea. So make sure to test everything to avoid failure.

Not the answer you're looking for? Browse other questions tagged or ask your own question.