HeyMarvin
- Yu-Ri Chang
- Jul 29
- 4 min read
As a graduate student deeply involved in user research, I’ve come to rely on GenAI tools as essential collaborators in my process. One of the most transformative tools I use is HeyMarvin, a generative AI-powered qualitative analysis platform that has streamlined both my primary and secondary research processes.
In user research (especially during the early desk research phase, in my opinion), the most time-consuming part is locating the right sources in the first place. I often spent hours digging through articles, only to realize they weren’t even remotely aligned with the insights I needed. That’s where HeyMarvin made a major difference.
While I still manually comb through hours of interview recordings and academic texts, HeyMarvin enhances this process by helping me extract patterns, summarize key takeaways, and surface meaningful quotes across sources that I upload. It serves as a centralized, searchable research hub that helps me spot and identify recurring themes more efficiently. The first thing I do is desk research: compiling papers related to my research topic and questions, and uploading them to HeyMarvin. After skimming through the papers, I ask HeyMarvin the research questions (created and discussed before the search) and review the responses that surfaced. If I notice that I missed part of the reading, I go back and re-read to understand why HeyMarvin categorized certain themes or answered in a particular way, what it was referring to, and why (also ask follow-up questions). Then, I manually go in to take notes: adding labels, interview questions, and writing corresponding observations or answers in the input fields. I also tag relevant information as reference points to support my claims, hypotheses, or questions.
It’s a blend of manual and automated processes, but HeyMarvin has helped me work more effectively. Sometimes, it even sparks new curiosities, and I end up going down a research rabbit hole. Oddly enough, that deep dive helps me better understand the content, the audience, the authors, and how the paper was constructed.
Using HeyMarvin didn’t mean I stepped away from critical thinking- in fact, the opposite happened… I found myself using more of my time & brain to synthesize what was being surfaced. The AI helped me locate and organize relevant information quickly, but I always took time to cross-check those insights and make sure they aligned with the (broader) context. My first user research project’s questions focused on the reasoning and purposes behind presenting the same information (the Theory of Planned Behavior by Ajzen) through two different media: physical and digital. I was interested in exploring the differences in attitudes toward these two formats. Through a customer perspective study, I aimed to understand people’s preferences regarding accessibility, market trends, consumer segmentation, and social influence. Specifically, I wanted to uncover how and why digital media tends to be favored by pragmatic readers and consumers.
For instance, an example output was that, “...physical readers prioritize product quality, including the depth and breadth of news coverage. They value the tangible format and the craftsmanship involved in producing physical newspapers.” HeyMarvin enabled me keep my thinking grounded, and ensure that any conclusions I drew were well-supported by the data. In another case, while working on a recent research sprint (examining the effects of captions, transcripts, and reminders on learning and perceptions of lecture capture in a second language), I uploaded a set of interview transcripts to HeyMarvin. Within minutes, it revealed themes around user frustration and unmet needs. But instead of just copying those insights, I reviewed them manually, double-checked against original sources, and used my analytical lens to dig deeper into the ‘why’ behind each quote or behavior.
Beyond its AI capabilities, HeyMarvin’s intuitive interface and seamless integration with other platforms have made it even easier to stay organized. The platform is collaborative, meaning team members can easily access and understand shared insights, which has been essential for staying aligned in group projects or research teams. Like any tool, it isn’t perfect. Sometimes the AI misinterprets tone or misses nuance- especially when it comes to differentiating between an interviewer and a participant. However, I see that as a reminder of why human oversight remains so crucial. I treat HeyMarvin as a powerful assistant, not a final authority.
In short, GenAI has changed the way I study and work. It helps me automate the tedious parts of research so I can focus on what matters: interpreting, questioning, and generating insights. It’s not just about working faster- it’s about thinking deeper and learning better.
References:
Dommett, E. J., Dinu, L. M., Van Tilburg, W., Keightley, S., & Gardner, B. (2022). Effects of captions, transcripts and reminders on learning and perceptions of lecture capture. International journal of educational technology in higher education, 19(1), 20. https://doi.org/10.1186/s41239-022-00327-9
Gernsbacher M. A. (2015). Video Captions Benefit Everyone. Policy insights from the
behavioral and brain sciences, 2(1), 195–202. https://doi.org/10.1177/2372732215602130
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
Image Credits: https://heymarvin.com/product/qualitative-analysis