fbpx Skip to main content
Tag

AI

AI in UX: The Next Frontier in Design Innovation cover

AI in UX: The Next Frontier in Design Innovation

By AI, Digital Transformation, Experience Design, Innovation, Personalization, UX

Photograph of AI robots
Remember how just a couple of years ago we were frequently discouraged by SIRI not being able to understand what we were saying and constantly answering the wrong question? Our cars had trouble with voice recognition and more often than not were calling the wrong person. Those of us who have accents had an even harder time talking to our devices and hoping to be understood. How many jokes have you heard about autocorrect? It is amazing to see how quickly technology improved in just a few years. Clearly, AI is already being used in a variety of ways to improve UX today.

AI applications in our daily lives

Every one of us experiences the effects of artificial intelligence in our everyday life.

  • Companies like Netflix use AI to recommend movies and TV shows based on your viewing history.
  • Amazon Alexa, Siri, and Google Assistant are using AI to control smart devices, send messages, set reminders, and provide information, creating a more seamless user experience.
  • A lot of companies use AI-powered chatbots to provide customer support, answer questions, and provide personalized recommendations, enhancing the user experience and reducing the workload on human support teams.
  • UserTesting uses AI to analyze user feedback and provide insights to UX designers.

We all can agree that as technology continues to advance, AI will revolutionize UX design in a number of ways. Here are some of the most obvious changes we can expect to see:

  1. Improved personalization.
  2. Better automation
  3. Enhanced predictive analytics
  4. Increased variety and quality of conversational interfaces
  5. Enhanced accessibility
  6. Improved testing and feedback

But like with everything in life, it is important to know not only the strengths and benefits of the new technology but also the potential dangers.

Let’s take a deeper look at each one of those items.

Personalization

Advanced data analysis can help UX researchers to analyze user data more effectively and accurately. It can provide insights that can lead to more personalized experiences. However, just like in market research, collecting more data doesn’t necessarily help with understanding its meaning. Designers and researchers may be tempted to rely too heavily on data instead of considering user feedback and intuition.

AI can help navigate the complexity of real-time personalization by quickly analyzing user behavior and preferences. The latter will allow to better adapt to the needs of individual users. Personalization, in its turn, can help to further improve chatbots and voice assistants. However, some users may be uncomfortable with the level of personalization that AI can provide. Hence, designers should be ready to face reluctance to accept the new technology solutions on the customer side.

Automation

There is a number of daunting and tedious tasks that UX designers do on an everyday basis. AI can help speed up design processes. By generating design variations such as layout, typography, and color selection, designers will be able to quickly test different assumptions. On top of that, design areas that can benefit from AI extend to design systems and much beyond.

Image of colored gear wheels

At the same time, designers relying too much on automation can cause a loss of creativity. It can also shift expectations on how long the design process should take and promote the so-common “anyone can do it” attitude.

We also should not forget that automated systems require upkeep and maintenance. So designers must invest time and resources into maintaining and improving their automation tools. Always keep in mind, automation should be something that complements, rather than replaces creativity and thought.

Enhanced accessibility

One of the interesting areas where AI can make a significant difference is accessibility. Accessibility refers to the practice of designing products and services that can be used by people with disabilities, such as visual or hearing impairments. Companies should start looking into AI to help them with 508 compliance.

AI-powered voice assistants, automatic captioning, image recognition, and natural language processing (NLP) are critical elements of enabling people with disabilities to work seamlessly and productively. Voice commands can provide an accessible experience for people with visual or motor impairments. Automatically generated captions for videos and other media can make them accessible for people with hearing impairments. Besides, image analysis and description can provide users with visual impairments with the much-needed ability to understand visual content. And finally, chatbots and other conversational interfaces help people who need assistance with cognitive or language processing.

The use of AI can help developers of those systems generate code to provide much-needed features. In this case, we can view developers as users of intelligence-powered tools that increase their productivity.

Improved testing and feedback

It is hard to overestimate the importance of testing and feedback in any design process. AI can improve designers’ efficiency and effectiveness by automating testing processes, analyzing user feedback, identifying patterns and trends, and A/B testing.

Predictive analytics and NLP can be used to anticipate user behavior, extract important insights, and better understand user needs.

However, it is critical for designers to understand the limitations of AI and incorporate human feedback and intuition into the design process, balancing the insights provided by AI with the creativity and empathy of human designers.

Predictive analytics

Predictive analytics uses machine learning algorithms to analyze user data and predict future user behavior. Everything we’ve talked about in this article heavily relies on predictive analytics. AI helps with the analysis of large volumes of user data, making real-time predictions, optimization of user experiences, and testing multiple design variations.

Image of a screen that shows data charts

Integration of AI tools into existing applications

Yes, AI can help designers do their work, but let’s not forget about the enormous help that AI can provide for our users in their day-to-day work.

Incorporating generative AI into email applications and text editors can help people to communicate more effectively by providing them with initial structures of emails, PowerPoint presentations, and documentation.

Bringing AI into internal corporate portals can save time for thousands of employees during the onboarding process, or when seeking for right people or procedures within the organization.

Thousands of existing applications currently heavily rely on people using other applications or search engines (Google, YouTube, etc.) to provide additional assistance with their products. Imagine how powerful and productive those products could become if by using AI they could seamlessly bring this external content into their products.

Dangers of AI

We spoke a lot about the multiple positive effects that AI can have on design processes, but let’s not forget that there are multiple areas in which using AI (especially in its current state) can be rather dangerous.

AI algorithms are not perfect, and designers as well as their managers must be aware of the limitations of the technology they are using. The correct interpretation of user feedback is pivotal for a designer. However, the use of AI does not always guarantee accuracy.

The accuracy and reliability of predictive analytics depend not only on the data quantity used to train the AI model but also on the data quality. Designers may start relying too heavily on data from AI-powered testing and feedback tools, potentially overlooking important subjective experiences and feedback.

In addition, AI algorithms may reinforce existing biases in data, leading to biased feedback and potentially misleading design decisions. AI also doesn’t have empathy which is necessary to truly understand user needs and preferences.

When using AI tools, designers need to verify that data is collected in an ethical and responsible way, protecting user privacy and ensuring that the system doesn’t have negative impacts on users or society.

Final thoughts on AI in UX

Overall, there is a lot of complexity associated with the use of AI in the design process, and it is important to find the right balance between possibilities and practicality. Designers should also beware of the limitations of emerging tools and take precautions to prevent misleading design outcomes.

Voice Interfaces: New Era Of Human-Computer-Interactions

By AI, Interaction Design, User Experience, Voice

Voice interaction is the ability to speak to your devices, have them proceed your request and act upon whatever you’re asking them. Today voice user interfaces are everywhere: we can them in smartphones, TVs, smart homes and a range of other products. The rapid development of voice interaction capabilities in our daily lives makes it clear that this technology will soon become an expected offering as either an alternative or even a full replacement to, traditional graphical user interfaces.

According to Gartner, by 2018, 30 percent of our interactions with technology will happen through conversations with voice-based systems.

Amazon Echo
Apple’s Siri, Amazon Echo and Google Now, which have been available for a few years, prove that this technology is no longer in its infancy.

Voice interaction is the next great leap forward in UX design.

In this post, we’re going to explain why voice interfaces will be the next big thing and what does this trend actually mean for designers of the user experience.

What Are Driving Forces Behind Voice Interaction

Before we dive into the specific implications of voice interaction systems or design aspects for them, it’s important to understand what’s lead to rapid adoption of this new interaction medium:

Technology is Ready

It’s clear that improvements in natural language processing have set the stage for a revolution. In 2016 we saw a significant breakthrough in natural language processing, and we’ve reached a point where advances in computer processing power can make speech recognition and interaction a viable alternative to visual interfaces. Another important thing is a number of devices that support voice interaction – today almost a 1/3 of the global population is carrying powerful computers that can be used for voice interaction in their pocket, and it’s easy to predict that a majority of them are ready to adopt voice interfaces as their input method of choice.

Mobile app and voice input
Image credit: Samsung

Natural Means Of Interaction

People associate voice with communication with other people rather than with technology. This means that voice interaction systems can be a more natural way of interaction for the majority of users.

Voice interaction chart
Image credit: Google Mobile Voice Survey 2014

People Want a Frictionless Experience

To interact with a voice interaction system all users need to do is to simply speak to the devices and be understood. In comparison with graphic user interfaces (GUI) where users have to learn how to interact with a system, voice interaction systems can significantly reduce the learning curve.

Windows 10
Even the most advanced graphical user interface still requires humans to learn a computer’s language.

Opportunities For Business

Adding Personality To Branded Content

Companies can leverage the medium of voice interaction as an extension of their personalities. Gender, tone, accent and pace of speech can be used by experience designers to craft a particular customer experience with their brand. For example, kids may finally get to talk directly to their favourite cartoon characters.
Branding

Make Experience More Personalization

Using voice-based system it’s possible to create a deeper personal connection to the system. Even today if you look at the online reviews for Amazon’s Echo speaker, it’s clear that some people establish a close bond with their device in a way that more resembles a pet than a product.

Samantha from Her
Samantha from Her

Voice Interfaces Aren’t a New Direction, They Just a New Step In UX Design

If you are new to designing voice user interfaces, you may quickly find yourself unsure of how to create great user experiences because voice interaction represents the biggest UX challenge since the birth of the iPhone. They are very different from graphical user interfaces and designers cannot apply the same design guidelines and paradigms. But while designing for voice differs from traditional UX, classic usability principles are still critical to the quality of the user experience.

Understand The Basics Of Human Communication

To design great voice user interfaces, you must handle the expectations users have from their experience with everyday conversations. And for that, we must understand the principles that govern human communication: how people naturally communicate with their voices.

Understand User’s Intent

Voice-based interactions between a user and a machine can lead (potentially) to infinite possibilities of commands from a user. While designers may not be able to predict every possible user command, they need to at least design an infrastructure that is contextually driven. For that, it’s important to start with a use case (a reason for interacting in the first place) and try to anticipate users intent at each point in the conversation (to shape the appropriate response).

User Intent
The processing flow of a comprehensive speech interface. Image credit: API

Provide Users With Information About What They Can Do

While on a graphical user interface, a designer can clearly show users what options they can choose from, it’s impossible to do this on a voice interface. In voice user interfaces, it’s almost impossible to create visual affordances. Consequently, looking at a device that supports voice interaction, users will have no clear indications of what the interface can do or what their options are. Therefore, it’s still possible to provide the user with the options for interaction. For example, if you design a weather app you can make it say: “You can ask for today’s weather or a forecast on this weekend.”

Limit the Amount Of Information

While with graphic user interfaces you can present a lot of different options, with verbal content, you need to keep the information brief so that the user does not become confused or overwhelmed. It’s recommended that you don’t list more than 3 different options for an interaction.

Craft Meaningful Error Messaging

Error handling is an essential component of designing thoughtful voice interactions. The wide variation in potential responses places much more emphasis on the importance of crafting meaningful error messaging that can steer the conversation with the user back on track if something goes wrong.

Use Visual Feedback

It’s recommended to use some form of visual feedback to let the user know that the system is ready and listening. Amazon’s Echo is a good example of this: on hearing a user say ‘Alexa’, the bluish light swirls around the top rim of the device, signalling that Alexa’s ‘all ears.’

User Input and Amazon Echo
Image credit: thewirecutter

Conclusion

Voice is the next big platform – it represents the new pinnacle of intuitive interfaces that make the use of technology more natural for people. Properly designed voice interfaces lead users to accomplish tasks with as little confusion and barriers as possible. And the good news is that UX designers already possess the skills they need to design effectively for voice.

Conversational Interfaces

By AI, Onboarding, User Experience, User Interface

You may have heard that “conversational interfaces” (interfaces that mimics chatting with a real human) are the new hot trend in digital product design. Several factors are contributing to this phenomenon:

  • With the advent of WhatsApp, Slack or Facebook Messenger the way we exchange information changed irreversibly. According to Business Insider, we are now spending more time in messaging apps than on social networks.
  • Artificial intelligence and natural language processing technology are progressing rapidly. Major technology players including Apple, Google, Microsoft and Amazon placed huge bets on this type of interfaces, leveraging big data and machine learning to get as close to human intelligence as possible.

This represents an interesting shift in how we think about user experiences and interactions, more as a text/voice based ‘conversation’ that helps us to achieve our goals. In this article, we’ll examine all major aspects of conversational interfaces in the context of chatbots.

5 Basic Principles of Conversational Interfaces

1. Be specific about chatbot purpose

Unless you develop a bot like Facebook M, it’s always better to deploy a specialized, purpose-driven bot to engage your target audience. Don’t try to design your chatbot to do everything all at once. Instead, identify the core use cases for your chatbot based on user’s goals and focus on achieving domain mastery.
Faceboom M

2. Mimic natural conversation

Keep in mind that when the conversation is the interface, experience design is all about crafting the right words: bots must use and understand natural language. A vocabulary that’s limited to only a handful of generic answers will immediately destroy an illusion of real conversation and leave users feeling frustrated. Nobody want to participate in chats muffed by pre-determined answers.
Conversation flow with Hi Poncho chatboat
An early version of the weather chatbot Hi Poncho struggled to provide any meaningful information due to a limited understanding of natural language. Image credit: Gizmodo

3. Make it clear what options are available for user

In traditional GUIs, what you see is what you get. However, with conversational interfaces, the paths that the user can take are virtually infinite. For conversational interfaces, users should know what paths are available for them. If you app is complex and has a few main routes, you can use an onboarding process to show the users what’s available.
Onboarding experience in Kia Niro
Kia Niro using the carousel to explain how to use a chatbot. Image credit: Sabre Labs

4. Avoid lengthy messages

Lengthy messages look like text paragraphs. People don’t speak in paragraphs, we speak using single sentences. You should plan for no more than 90 characters per message (around three lines on mobile). Anything more than three lines of text seemed to activate the tl;dr (too long; didn’t read) response in users.
Too long sentences in KAYAK
Kayak chatbot hits the users with 4 opening messages, totaling nearly 750 characters. Most users glazed right over when they saw the wall of opening messages.

5. Animating the conversation

Animation can take the chatbot user experience to the next level, making the interactions more natural and pleasurable for user. Simple typing indicators can be used as an equivalent to phatic expression in speaking, making the conversation flow smooth.
Chatbot and animation
Typing animation via Buzzfeed

Best of the Best

Conversational interfaces open lots of new possibilities how you can interact with users. Below are two popular apps that successfully embraced the new paradigm of conversational UX:

Domino’s Pizza

Domino’s pizza allows “conversational ordering” via Facebook Messenger. Customers add Domino’s pizza as a friend via Facebook, set up the basics of their account, and can then “reorder their favorites” or ask for the latest deals.
Domino's pizza chatbot
Domino’s Pizza via Techcrunch

Duolingo

Duolingo is a language learning platform which uses gamification and personalization to make learning a new language effective. Last year Duolingo introduced Bots. This feature allows users to practice language skills by texting with a ‘Bot,’ which takes on different topics as a way to explore a range of conversations, such as going to a restaurant, going through border checks, or ordering a taxi.
Domino's pizza chatbot

Conclusion

Whether you love them or hate them, conversational interfaces have started making a significant impact in communication. Of course, most of them today have certain limitations and they don’t have human-like conversations perfectly that’s why it’s so important to follow basic principles of conversational interfaces  mentioned above. But in the near future, continuous advancement in machine learning and artificial intelligence technologies will fill this gap and we will see AI-powered chatbots which will have human-like conversation.