Welcome to a deep dive into the influence of AI in the field of UX research. Let’s dig in to truly understand how AI technologies will shape the way future researchers and designers work.
In 2023, global AI market revenue was $196.6 billion. Over the next five to ten years, that number is expected to grow at 37% annually. Or 31%. Or 19%, depending on your source.(1)(2)(3)
Our point is, AI’s influence is already BIG. And it’s only going to get BIGGER.
This disruptive technology is quickly changing the UX landscape. It’s forced a rethink of research and design roles, and the organizational design and dynamics that surround them.
In this article, we’ll explore fundamental concepts of AI and how it’s permeating the UX Research and Design fields.
Here’s a tl;dr summary of what we’ll cover:
- Understanding AI in UX Research
- Enhanced Data Collection and Analysis
- Personalization and Tailored Experiences
- Using Advanced Research in Research Analytics and Design
- Chatbots, Virtual Assistants and Conversational Interfaces
- Extracting Actionable Insights
- Future Trends and Innovations
- AI’s Transformative Impact on UX Research
Understanding AI in UX Research
Ever notice how colleagues at a new job use abbreviations completely alien to you?
Worry not. When it comes to fundamental AI, we’ve got you covered. Use the table below to stay well-versed with basic AI concepts and terminology:
[Forewarning: “Abr” stands for abbreviation. We did it again. Sigh.]
Marvin’s Guide to AI Terminology
| Phrase | Abr. | Description |
| Artificial Intelligence | AI | Umbrella term for the simulation of human intelligence by a machine. Intelligence = perceiving, synthesizing and inferring information. |
| Machine Learning | ML | Subset of AI, ML uses algorithms that enable computers to learn patterns or make predictions based on a dataset. |
| Natural Language Processing | NLP | Type of AI that conducts interactions between humans and computers. Machines understand, interpret and generate human-esque responses. Initially trained with real-life data, they use pattern recognition to come up with responses. |
| Large Language Models | LLM | AI algorithm that uses language as their input and output. Trained on extremely large datasets to understand, summarize, generate and predict new content. |
| Generative AI | GenAI | Type of AI that can create images, video, audio, text and 3-D models. |
| Application Programming Interface | API | Software that facilitates communication between different applications. |
| Neural Network Model | NNM | AI model that teaches computers to process data like a human brain, so it can make decisions on its own. Unlike ML (which makes decisions based on what it learned from data). |
| Generative Pre-Trained Transformers | GPT | Series of neural network models that learn context and meaning by tracking relationships in sequential data. Think about how auto-complete knows what to write next in messaging apps. |
Learn about the capabilities and limitations of AI in UX research.
Now, let’s examine how these various technologies fit into a researcher’s or designer’s workflow.
Enhanced Data Collection and Analysis
Two approaches to data collection here — passive and active. Active studies are conducted to answer a specific research question. Passive data collection involves collecting data on an ongoing basis.
To collect data from active research projects, Marvin’s AI uses NLP and LLM to transcribe your user interviews. Get a verbatim transcript within minutes. No more frantic note taking.
Learn more about the differences between passive and active research.
Let’s talk about passive data collection. The prevalence of APIs means that apps (Google Analytics and the like) supply a continuous data stream to companies. Once in a research repository, ML helps researchers unearth trends and patterns from the dataset. AI tools for data analysis ensure that critical insights don’t go overlooked.
Researchers can use AI to conduct rapid analysis, processing volumes of data from complex datasets in real time. The benefits are threefold:
- Aggregating and interpreting user data in real time brings adaptability. In an ever evolving digital landscape, it helps researchers unearth “in-the-moment” insights. These insights provide companies with feedback for on-the-fly decision making, eliminating guesswork.
- AI’s ability to handle large volumes of data enables researchers to take on more extensive projects without increasing the time and resources spent.
- Another upside is scalability. AI allows researchers to meet increased data processing needs (previously impossible if conducted by humans). AI allows companies to ramp up their research efforts rapidly and efficiently.
AI can collect and analyze data round the clock for you — 24 hours a day, 7 days a week…you get the idea.
Learn more tips and tricks to incorporate the use of AI in your research.
Personalization & Tailored Experiences
Ever scrolled through a friend’s YouTube feed? You likely noticed it’s entirely different from yours. A prime example of a customized experience.
UX will always be centered around designing an experience that delights each and every user. Designers and researchers’ primary aim is to create tailored experiences that improve customer:
- Adoption
- Engagement
- Satisfaction
- Retention
Using data about how users interact with an application, AI can decipher individual user patterns and preferences. To craft well rounded user personas, AI uses NLP to understand peoples’ attitudes and emotions from social media posts, forums and customer reviews.
AI is capable of dynamically adapting interfaces and suggestions based on these user personas.
This opens up the door for more inclusivity and accessibility. Using AI-driven tools, researchers and designers can create applications and processes that cater to a more diverse user base, considering their abilities, backgrounds and general product use. UX researchers can then study the effectiveness of personalized features and iterate on design choices to improve user satisfaction.
AI will eventually tailor experiences to each individual user. This results in highly customized user experiences based on an individual’s preferences.
Imagine a day when ALL your apps look and feel completely different from someone else’s. With AI, that level of personalization is right round the corner.
Using Advanced AI in Research Analytics and Design
Ever wondered how Amazon knows exactly when you’re out of dishwasher tablets or detergent?
It’s down to predictive analytics.
Amazon’s AI examines your consumption and purchase history, and it uses ML to estimate when you’ll run out. That’s when it automatically sends you the eerie notification.
Predictive analytics uses statistical algorithms and ML techniques to enable a deeper understanding of user behavior. Designers can identify patterns in data and troubleshoot user issues or pain points to improve UX design. It enables UX professionals to make better-informed design decisions.
AI analytics can alert professionals to anomalies and issues. However, it still requires a human being to interpret results and understand its implications.
AI has forged its way into visual design as well. AI features expedite design tasks. It can help create a plethora of prototypes or wireframes that weren’t humanly possible before.
Companies such as Adobe have introduced AI features to help designers build visually appealing and effective user interfaces. To name a few:
- Visual asset analysis — comparing attention across various creative elements
- Text-to-image generation
- Color & font pairing
- Cropping or resizing
- Removal of items/objects
Implementing advanced AI in research analytics and design results in well designed user interfaces. AI can also help gather and analyze usability metrics to understand what’s working and what isn’t. It makes the feedback cycle a lot shorter.
Chatbots, Virtual Assistants and Conversational Interfaces
ChatGPT’s release was revolutionary. It changed the way college students worked (or avoided work), how we access information and brought efficiency to the workplace.
Our friends at User Interviews conducted a study with over 1,000 UX professionals and found that 77% of respondents already use some form of AI in their work.
NLP is the underlying technology behind applications like ChatGPT. It’s essential for chatbots, voice assistants (such as Siri or Alexa) and sentiment analysis. UX researchers can leverage NLP to conduct surveys and analyze user feedback. This allows them to iterate and improve the conversational aspect of a user interface.
If you haven’t jumped on the AI chatbot bandwagon already, FAQPrime assembled a beginner’s list of prompts to get you started. Tinker with these commands and see what suits your workflow best. Remember to follow these basic guidelines to follow when interacting with a chatbot or virtual assistant:
- Provide ample context
- Ask for multiple options
- Iterate on the output
- Build a prompt library
Search Your Own Research the Way You Search the Web
Marvin’s Ask AI product is the ChatGPT of Research. Trying to garner insights from old studies? A perennial pain in the behind, no more! Ask any question to begin searching your research repository and get the answers you need in seconds.
Aside from transcribing calls, Marvin’s end-to-end research assistant conducts foundational qualitative analysis for you. It creates automatic notes during and after interviews and is capable of articulating the gist of mundane and long conversations with AI generated summaries. AI synthesis allows researchers to annotate on the fly.
AI frees up a researcher’s time to focus fully on the interview and delve deeper into user pain points. Remember, using AI as a sidekick is like working with an intern. While fully capable of understanding instructions and completing tasks, you still have to check their work.
Learn about Marvin’s most popular products launched in 2023.
Extracting Actionable Insights
With AI’s help, designers can conduct comprehensive design reviews. Observing how users interact with a system (or predicting how they will interact) provides insight into the user journey. This allows researchers and designers to scrutinize UIs to identify usability issues and other areas that need improvement.
AI’s ability to handle vast and complex datasets accelerates the shift from raw data to actionable insights. AI extracts patterns and insights much faster than a human would. Using AI to conduct preliminary analysis and synthesis creates an immediacy in results. This helps answer questions like:
What changes can we make to the product based on this information?
It all leads to the development of designing smarter, modular design interfaces that are user-friendly.
Data-driven decision making improves the accuracy and quality of research. It provides insights using real-time data to identify areas of interest. Researchers and designers can then make informed choices based on the underlying data.
AI adds efficiency to the design process, however, its findings or results need to be interpreted by someone(human). You still (and always will) need a human being on hand to establish the direction of research and make sense of it all.
Future Trends and Innovations
AI will augment researchers’ work, not replace researchers themselves.
We’re more bullish on this assertion today than ever before.
Marvin CEO Prayag Narula believes AI is the perfect research assistant. AI-generated summaries help with a superficial understanding of the big picture. They also save researchers precious time ordinarily spent rewatching long videos and taking notes.
Our users tells us Marvin’s AI transcription is game-changing. It frees up their time and mindspace to focus on deepening their understanding of users.
AI in UX Research Technology to Watch For
However, transcripts miss important context — they don’t capture emotion. People don’t verbalize all their actions and describe everything that they’re thinking.
To tackle these shortcomings, some AI tech is still in the works:
- Biometric technology is (theoretically) capable of capturing human emotion from video or visual artifacts. Its current output is confusing, messy and misleading — it requires more training data. Emotion recognition has the potential to provide valuable insights into users.
- Synthetic users are largely unpopular among the UX community. Designers argue that they build for humans, not synthetic or artificial users. However, there are instances when it can be useful. During early days of a research practice, it can help smaller teams increase the scale of their work.
With AI technology developing rapidly, it might be a short time before these features are added to its growing arsenal.
To understand more about the UX industry’s continued evolution, Lattice’s VP of Design stopped by to share his predictions for 2024. Jared reminded researchers and designers to focus on:
- Efficiency — or how to do more with less. Companies will harness a combination of researchers, designers and AI capabilities to maximize research output.
- AI literacy is now a prerequisite. It moves from being a “nice-to-have” to a “must-have” skill for researchers and designers. Companies will look to hire UX professionals proficient or at least competent with AI.
AI’s Transformative Impact on User Research
AI has staggering potential to change our lives. It already has. Promising future applications of AI span across industries including meteorology (predicting weather) and healthcare (early symptom detection and diagnosis).
However, AI is at a nascent stage of its development. There are tasks it can’t perform, and things it might never do. It’s also bound to run into speed bumps with plenty of missteps along the way — think of the early days of dial up internet. There’s a considerable way to go. Best practices for designing with AI are still being developed. Education across industries will raise awareness as people become more attuned to concerns such as bias, privacy and exclusion.
It’s crucial to garner as much knowledge as we can about AI. Not only to keep up with the latest trends. Moreso to understand how humanity and tech can continue to push boundaries to generate deep user insights. To continue its development, AI needs more human oversight. Combining AI-powered analytics with human interpretation and logic produces high quality research.
It’s the potent combination of AI & human knowledge which will supercharge the future of UX.
Learn more about G2’s top-rated UX research repository
Find out how to integrate Marvin’s AI features into your UX research workflow.
Book a free demo today.
