How to Create Great User Interfaces: Make It Invisible!

Paul Boag

There is no shortage of articles on how to create great user interfaces. However, the best user interfaces are the ones you don’t have to use.

We all want to know how to create great user interfaces that our customers love. However, in many ways, this is a faulty assumption. It presumes that people want to use your interface in the first place. That is rarely the case.

Think about the user interface you find the most enjoyable to use. Perhaps it is packed with design delighters and is ridiculously intuitive. Maybe every time you use it, it feels effortless. However, is it the interface that you want to use or is it the underlying functionality it provides?

Would you still use it if it worked the same way but failed to deliver the result? Would you use a gorgeous twitter app that didn’t tweet, or an easy to use ecommerce site where they didn’t take your order? Of course not.

We use a user interface to achieve an aim, not for the pleasure of using it, no matter how good it is. So is it time we started looking for ways to reduce the amount users have to interact with our user interfaces? I would argue it is, and that this is the natural evolution for many products.

The Journey Toward Great Invisible User Interfaces

When a new piece of technology comes along it passes through a series of predictable steps in its lifecycle.

To begin with, the company that introduces new technology have a significant advantage. People buy from them simply because there is nothing else like it on the market. However, that advantage doesn’t last long as competitors begin to crop up.

Then it becomes a competition over features. I remember shortly after Video Cassette Recorders came to market that competitors quickly started packing them with features to try and differentiate themselves in the marketplace.

Video cassette recorder
The VCR became increasingly hard to use as more and more features were added.

The problem with the features stage of a product’s lifecycle is that products quickly become incredibly challenging to use. That was certainly the case with the video recorder. However, it was also true with applications like Microsoft Word and even entire product line ups. For example, I remember Apple before Steve Jobs returned to the company. At the time, Apple had so many computers that it was impossible to know which one was right for you.

Old version of Word showing all toolbar options.
Microsoft word became overwhelming as more features were added.

The result of this feature bloat is that it opens up an opportunity for new companies to step in and introduce a more streamlined user-friendly experience. Sky Plus would be an example of an easy to use user interface that replaced the clunky VCR. At this stage of the product lifecycle, the effortless user interface reigns supreme and is the reason the iPod became dominant despite there being better MP3 players on the market. However, this isn’t the end of the journey.

The original iPod
The original iPod succeeded not because it was the best MP3 or had the most featured. It succeeded because it was easy to use.

As we have already established, users would prefer to see results; it is not the interface itself they want. So after the effortless user interface stage, comes an opportunity to reduce effort even further by removing the UI.

To follow through the VCR example, we now have Netflix. They are beginning to use machine learning to recommend the best shows for us to watch behind the scenes without us needing to search for them.

Netflix is reducing the time I spend using their interface by improving their recommendations, so I don’t need to spend as long browsing.

Of course, we still use the Netflix interface; but we use it less and less as it learns about what we like. The user interface takes more of a backseat as we don’t need to search for new shows as much.

I am conscious that this might all sound a little abstract, so let’s look at some practical examples.

Practical Examples of a Backseat User Interface

Probably my favourite example of an evolving backseat interface is how I do my finances using a tool called YNAB.

For many years I manually entered every transaction I made into YNAB by hand. I would then compare those transactions to my bank records to make sure they matched. It was painful, involving a lot of back and forth between YNAB and my bank.

Then YNAB introduced the ability to download my bank transactions and upload it manually to YNAB. Now I spent a lot less time entering data and so by extension using YNAB.

YNAB Bank Connection Facility
Because YNAB now connects automatically to my bank I spend much less time in their application.

Finally, more recently, YNAB added the ability to automatically connect to my bank and import the transactions keeping the two systems in sync. Suddenly I didn’t have to enter any data, and the whole app became mainly about budgeting.

With each step, I used the interface less and less but became more and more pleased with the service. It’s not about the interface.

CAPTCHA is another example of this. Initially, CAPTCHA involved entering squiggly text and numbers. Then, the interface was simplified somewhat to allow for the selection of pictures. After that, Google started only to ask you to complete a task in some circumstances, and now today, we have invisible CAPTCHA that rarely involves the user doing anything.


The problem is that this kind of approach requires a shift in thinking that some will find uncomfortable.

Challenging Existing Thinking

The idea of an invisible experience is something that makes both marketers and designers uncomfortable. After all, management often measures marketers on engagement levels, and designers fear they may lose their jobs if the interface goes away.

It is hardly surprising that Golden Kristina’s excellent book The Best Interface Is No Interface was greeted poorly by some.

However, many companies do well from making their product as invisible as possible. Zapier is an excellent example of this. Zapier exists to connect different software applications, and once you have set it up, you rarely look at its interface again. It just does its thing in the background.

Zapier Homepage
Although I ‘use’ Zapier multiple times a day, I cannot remember the last time I used their user interface. It is all automatic.

The same is true for my Tile tracker attached to my keys. I seldom open the Tile app, but I wouldn’t be without that notification that tells me I have left my keys behind.

Yes, these concepts might be challenging for some, but there is no doubt they are excellent for business.

So what does this mean for your business?

Putting Your User Interface in the Backseat

There is no one size fits all use of this kind of invisible interface. How it applies to your organisation will vary hugely. However, there are some places to start thinking about this stuff.

First, look for pain points that you could fix behind the scenes. CAPTCHA is an example of this. They now use technology to block spam, rather than making the user deal with it.

Second, look for opportunities to automate in much the same way as YNAB automated the importing of my bank transactions.

Third, make use of the advances in machine learning and recommendation engines to be more intelligent about what you put in front of users. Show them relevant content, rather than making them sift through everything you have.

The Next Big Differentiator

We already know that a well-designed user interface has been a big differentiator for many. I also believe that the invisible interface (alongside an outstanding customer experience) could be the way you start to stand out in the future.

That is especially true if you work in a sector where the user interfaces are already pretty robust. For example, I could see this as a big differentiator for software as a service where interface maturity is high.

However, the key takeaway is not to fall in love with the user interface. Instead, remember that ultimately, users are more interested in completing their task with as little effort as possible and preferably no interaction at all!

Stock Photos from pathdoc/Shutterstock