How do you handle conflicting user feedback obtained from interviews?

In qualitative research, there's this practice of converging, mapping, coding, and providing themes of your research participants' narratives or transcriptions. In that convergence exercise, you are aggregating these insights for high-signal findings and themes. As a result, there's typically going to be enough critical mass in a particular area for you to create a theme or a particular insight. If you're finding that you're getting a lot of conflicting information or you're not able to establish a theme, this may mean you need to look at your participants. Analyzing if their roles are mapping to the problem that you're looking for. Sometimes when you're getting that convergent insight, it's because you're not targeting the right audience. Ask yourself is this participant is from the right company, industry, or experience. Is the segment correct? Is the company size correct? Is the role or persona correct? And after you refine that, you might need to expand the number of participants per role, and then you may have different insights per kind of cohort.

This difference in cohort insights is a natural part of research. You'll learn that different users and different personas are going to have different needs in your product. So if you're interviewing, let's say, an admin persona who's responsible for setting up a particular product versus an end-user persona who's going to be interacting with the product, they're going to have different expectations of what a product should do, and therefore your insights are not necessarily going to be the same.