Anete LusinaDiving into user research (without a product yet 🙈)
We have a product idea, but not the actual product. So, how the heck do you do any research for your potential target audience? 🤔 That has been a fun challenge we worked on in the past month or so.
Gavin (the Product Pro) recently got a new job so all the congrats 🎉, and I have been busy with my work and coding course which meant finding time for this project has been DIFFICULT. To say the least. But let me break down our steps for the user research stage.
Find people to interview
Where do you even begin finding people to interview for your product idea? Especially when you have no previous product insights or anything useful to go by? I went straight to the places I know best — Twitter, Linkedin, and marketing Slack channels.
I was looking for people who are already interested in skincare because that’s our target audience. I wasn’t fully sure about the age range or other characteristics. From researching competitors, I struggled to find information on their users beyond the fact that most are women.
Overall, I did eight interviews.
I was lucky to have people in my network who were more than happy to help. Maybe it’s because there’s always ongoing research, interviews, and article quotes in the marketing space. Beyond grateful for the support, though, especially because people happily contributed half an hour of their day to me without any tangible benefit to them. (Besides the chance to chat about skincare).
Here is a simple post I put up on Linkedin, and I got three interviews from it.
I used Calendly to set up my availability to share with the interviewees and set Zoom to record calls locally on my device so I could transcribe them later. You can find the setting in the Advanced Options in your Zoom.
Conduct interviews
I prepared for the interviews by noting down several questions in advance. I didn’t always get to ask all the questions because interviewees sometimes already covered them while discussing something else.
The way I prepped my interview notes looked like this:
- Product features that I want to keep in mind when we talk (but not mention to the interviewee directly)
- Warmup questions (in case my mind is completely blank when we start the call)
- Questions about routines to find out what users already do day-to-day (to help better understand where our product could fit)
- Questions about products interviewees use in their routine
- Questions about buying products
Interviewing someone is difficult, especially if you’re unsure about leading the call on a new project like this. But the goal was to uncover routines, pain-points (frustrations), and the positive highlights.
Luckily, I’ve interviewed people for my work as a writer and a journalist. The difference, however, is that I always knew the outcome and the direction of each conversation when I interviewed a subject for an article.
But with this project? I learned everything and anything about my interviewees through a set of open-ended questions. It’s definitely a skill that takes time to refine, so if you think you suck at this, it’s okay. Use each interview to reflect on improving the process next time. 😊
Process interview data
Once I recorded each call, I uploaded it on the transcription website Rev (here’s a $10 off your first order with my referral link). I used AI transcription (not human); for example, a 36-minute interview cost me $9 to transcribe. I know there are cheaper transcription services, but I tried a trial with Rev and found the process super fast and smooth, so that was a win for me.
Once the transcript is ready, I export it as a Word Doc. Now is the fun part of going through the interview and formatting it into an easily digestible format.
I used ChatGPT to help me get these interviews processed faster:
- I gave ChatGPT a prompt to “Make this text readable, remove any words between < and >, and keep the text in the first person.” This would remove transcribed verbal cues, like <affirmative>, and make the interview coherent.
- I copied several consecutive paragraphs from the interview and pasted them after the prompt.
- You can ask ChatGPT to ‘Do the same for…’ and paste chunks of interview text. After two or three requests like this, you may need to repeat the original prompt.
Here is an example of what this looks like in practice (just a small snippet of the interview):
I then created a new page on Notion for each interviewee and pasted all ChatGPT’s output to have all information readily available when we progress to the next stage.
What’s next?
- Sticky notes session to bring all the interview notes together and see if we can spot any themes or shared needs to help inform the next steps
/w=1920,quality=90,fit=scale-down)