Exploring AI Role in UX Design and Research
In this article, I will share my experience using AI in UX Design and Research.
While some AI tools offer the potential for efficiency, not all deliver on their promises. To use AI, you need to consider business value and ethical considerations and ensure it solves real problems rather than creating new ones.
Let's review examples of how AI is used in the customer experience, internal UX design, and research processes.
AI and Customer-Facing Solutions
Personalization and product recommendations give customers a relevant experience based on previous browsing or purchase history. If done correctly, they increase customer satisfaction. Examples would be serving more relevant articles, offering product recommendations, or changing page layout.
AI Chatbot that helps customers get answers to their questions when they don't have time or don't find what they are looking for on the page. It is often used before connecting to the agent or when a live agent isn't available.
AI technology enhances customer experience and helps the bottom line if implemented well. But when it isn't well thought through or tested, it causes a bad customer experience and a drop in customer trust and conversion, lowering the bottom line.
AI and UX design and research processes
Many AI features appear within research and design tools, trying to help with data analysis and efficiency. Some vendors deliver value, but many are still not delivering on their promises.
Process-wise, we save time when we start automating routine tasks. AI could be helpful when we:
Conduct basic initial research to help understand the audience, needs, and pain points of a customer group with which you need to familiarize yourself. In this case, you usually take the output of generative AI as the assumption that you need to validate, but it provides a significant step forward.
Write the first draft of the usability study, interview guide, customer communication messages, or when you need a reminder on criteria for heuristic evaluation if you are early in your research or design career.
Need help with ideas for rewording content, checking grammar, etc.
Try to improve the quality of design work by conducting quick checks with an AI eye-tracking tool. This tool helps the design team zoom out and see their design through a different lens, do another rapid iteration, and catch the issues we missed.
Need help with identifying trends in large amounts of data that we haven't noticed during manual analysis or when we don't have time to process.
The future of AI in design is promising, particularly in advancing prototyping automation. Imagine a workflow where AI generates responsive mockups across various breakpoints, seamlessly fills in data within prototypes, and even translate prototypes. By offloading these tactical tasks, designers could focus on creative problem-solving and strategic decision-making, unlocking new levels of efficiency and innovation in the design process.
Best Practices for Implementing AI in UX
Keep an open mind and try new AI technology as it comes up. If a vendor is offering an AI functionality option, try it. It may improve your process or experience.
Ensure AI solves a real problem and adds value. If someone comes to you to implement an AI functionality proposal, ask why. Don't mindlessly implement it because it's popular or you feel under pressure. AI should solve the problem instead of creating more work. Measure the impact and share with the rest of the team.
Be a champion for AI solutions that truly add value because it could move your team forward and significantly improve the results. Help other people adopt when you see they are struggling by explaining the value of the change.
Use critical thinking and evaluate the output of AI when you get it. It could be getting feedback on the quality of output of the people using it. If you are serving recommendations to customers, it could be asking them to rate the usefulness of the result. If you use it in your work, validate questionable results with another research method.
Keep in mind ethical considerations like customer privacy, transparency, and keeping your results unbiased.
Challenges of AI Adoption
The biggest challenges I ran into were:
Push for AI tools and processes when there is no clear business value.
The misconception is that AI is further along than it is in my industry and can replace the actual research we conduct.
Lack of trust in AI technology from the workforce and resistance to changing existing processes
The biggest misconception about AI is that it would replace many jobs. It's super important to make it clear what it can and cannot do. AI is good at processing large amounts of data and performing repetitive tasks that take a long time. It isn't good at critical thinking or complex problem-solving. So, it's a great tool to help teams be more effective.
AI excels at processing large amounts of data and handling repetitive tasks, making teams more efficient. However, it falls short in critical thinking and solving complex problems. By offloading routine work to AI, teams can focus their time and energy on strategic, high-impact challenges.