April 24, 2023

Is Artificial Intelligence (AI) the future of research?

We have all heard the recent buzz surrounding OpenAI’s ChatGPT and DALL-E. Artificial Intelligence (AI) was around long before OpenAI. (Even their GPT model is on iteration 4!) But never before has the curiosity and discussion around AI been as widespread.

Confronted with such awesome (and unknown) potential as AI holds, we as mere humans will naturally start musing: “How will AI help humanity?” or “What are the true dangers of AI?” Whether we openly admit it or not, what most people want to know is “How will this affect me?!”

Much like the histories of the assembly line or the computer or other transformative technologies before, it is worth contemplating what AI means for the future of our work. I find myself thinking a lot about AI and research — especially because distilling insights from, say, user interviews feels so inherently human. And if I am honest, also because I am selfishly wondering how AI will affect me as a builder of products and businesses.

What pieces of user research can and should we let AI tackle?

But First: What Do We Mean By User Research?

User research is an integral part of designing products and services that meet the needs of users. Ask any top business and they either have a user research team or regularly contract others to conduct research.

More specifically, User Experience (UX) research is the practice of collecting and analyzing information about how users interact with a product, service, or system to improve its usability, accessibility, and overall user satisfaction. The data collected from UX research can help identify areas for improvement, guide design decisions, and measure the success of design changes. Ultimately, the goal is to understand users’ needs, behaviors, preferences, and pain points, and use this information to inform design decisions.

UX research can take many forms: surveys, interviews, usability testing, observation, and analytics. So the question is…

Where Will AI Play in User Research?

As technology continues to advance, we are seeing a shift towards using AI to help streamline and automate certain aspects of user research.

First, it’s important to understand what we mean by AI in the context of user research, and understand the difference between AI, data science, and data analytics. Each of these play a crucial role in user research, but it is helpful to understand the difference between them.

Data analytics plays a crucial role in user research as it enables researchers to gain valuable insights into user behavior and preferences. By analyzing data generated through user interactions with websites, applications, and products, researchers can identify patterns, trends, and correlations that help them make informed decisions about design, marketing, and customer support. For instance, they can use analytics tools to track user behavior on a website and see how they navigate through the pages, which pages they visit most, and which ones they exit quickly. This information can be used to optimize the site’s design, improve the user experience, and increase engagement and conversions.

As data analytics becomes more sophisticated, it transitions into data science methods, which involve using statistical and machine learning techniques to extract insights from large, complex data sets. With data science, researchers can analyze not only user behavior but also other factors that influence user experience, such as demographics, social media activity, and purchase history. They can also use predictive models to forecast user behavior and preferences, which can inform product development, marketing, and customer support strategies. For instance, data scientists can use natural language processing (NLP) algorithms to analyze customer feedback and identify common complaints or issues. They can then use this information to improve the product or service and enhance customer satisfaction.

AI augments user research by providing even more sophisticated and accurate insights into user behavior and preferences. By leveraging machine learning algorithms and neural networks, AI can analyze massive amounts of data and identify hidden patterns and relationships that humans might miss. For instance, AI can analyze user interactions with a product and predict which features are likely to be most engaging for a particular user. It can also identify segments of users with similar preferences and behavior and personalize the experience to each segment. This can result in higher engagement, customer satisfaction, and loyalty. Overall, AI augments user research by providing a deeper and more nuanced understanding of user behavior and preferences, which can inform better design, marketing, and customer support strategies.

Photo by John Schnobrich on Unsplash

So Why Still Have Human Researchers?

Now that we have reviewed the role AI plays in user research, it might be more clear why we still need human researchers. AI does not eliminate the need for humans, it more so amplifies the effort spent by human researchers. I believe the most insightful and yet efficient research can happen when humans use AI tools to supplement, elevate, and augment what they are doing.

For example, there is something magical about exploratory user interviews. A good researcher can interview another human being and suss out what is important to them. Connecting on an emotional level can often result in the revealing of deeper truths that are ultimately more important to empathetic product design.

Not only this, but AI is only as good as the algorithm and the data it was trained on. Having a human researcher be able to catch strange or clearly inaccurate AI outputs is critical. Additionally, most product research cycles I have lived through involve only a handful of interviews at a time. AI may not be accurate enough to detect patterns from such low sample sizes.

What is the Optimal Way to Use AI in Research?

I believe the short answer is to use AI in ways that supercharge the powers of good human researchers, making research more insightful, unbiased, efficient, and ethical.

My company is currently building an AI tool for research that I hope accomplishes exactly this. Our goal is to utilize an AI engine — comprising Natural Language Processing (NLP) and generative AI models — to provide researchers and product builders with a competitive advantage by enabling them to make informed decisions on user research quickly and accurately.

As product builders, we know the countless hours that go into distilling and pulling out insights from multiple iterations of user interviews. So we are using AI to speed up and remove bias from the process.

A high-level summary of how it works:

  1. You state your research goals.
  2. You upload videos or audio files of your research.
  3. We transcribe them.
  4. Our AI engine identifies the key text within all the interviews, and classifies and detects themes and keywords (and rank by the project goals).
  5. We also detect user sentiment.
  6. Our AI also helps summarize key insights, quotes, and video snippets for easy sharing.

The AI saved many hours:

  • Manually transcribing.
  • Using highlighters and post-it notes over many days to determine and organize keywords and themes.
  • Editing highlight videos.
  • Creating a stakeholder presentation to influence business decisions.

What will researchers do with that extra time? We hope they will augment what our AI produced with their own uniquely human touches. And of course, do more research!

If you are curious to learn more, contact me!

Is Artificial Intelligence (AI) the future of research?

James Friscia

CEO of CoNote