Last Thursday here at Vox Media, we held one of our regular Documentation Days, where our entire Editorial Products team spends the day churning out documentation on process, teams, workflows, projects, and best practices.
I took the time to flesh out some thoughts on one of my favorite research techniques—interviewing.
I have conducted thousands of interviews in my career so far, first as a reporter and now as a product designer, and am convinced that interviewing is deceptively complex. The results can be incredibly insightful, but a poorly planned and executed interview is at best, time poorly spent, and at worst, misguided insights that may mislead your end product.
Here are eight steps I follow when I interview in a design or research capacity:
1. Figure out which type of interview you are going to conduct.
This will help you determine who to talk to and how to structure your questions. In the research phase of a project, I’ve found that interviews typically fall into three main categories—user interviews, stakeholder interviews, and informational interviews.
- User interviews are great for gauging general attitudes and thoughts about the problems you are trying to solve. Target the end users of whatever you’re designing, and probe for specific instances and examples. You should come away with a sense of your users’ needs, goals, and behaviors.
- Stakeholder interviews can help you understand the requirements, objectives, and business constraints of your project. Talk to whoever has stake and final say in the project, be they leadership, outside clients, etc. Ask about priorities, success metrics, resources, and best case and worst case scenarios for the project. You should come away with a clear idea of the requirements and goals of whatever you are building.
- Finally, in informational interviews, you should target experts in subject matters related to your project in order to learn context and insights that can help inform your product decisions. These are the most free-form of interviews. You should come away with a better understanding of the space you’re working in.
2. Write down what you hope to accomplish in your interviews.
Each question or set of questions should serve one of these goals. Questions should not be asked arbitrarily.
3. Plan your questions ahead of time.
This puts you in a better spot to ask thoughtful questions that are consistent and help you reach your goals.
4. Don’t ask leading questions.
For instance, rather than asking: “Is this survey tool easy for you to use?”
Ask instead: "Tell me about your experiences using the survey tool. What works well? What does not work well?”
5. People tend to idealize their workflow.
In user interviews especially, ask for specific examples to get a better sense of their actual workflow. For instance, rather than asking: “What is your process when you are making a survey?”
Ask instead: “Have you made a survey before? When was the last time you made a survey? Walk me through how you made that survey.”
6. Ask about process and problems rather than asking about potential features, opinions about what users say they want, or what users like and don’t like.
Don't ask: “If we had a feature that randomized questions in a survey, would you use it?” Or “Do you like the idea of a survey question randomizer?”
Ask instead: “What are you trying to achieve when you write questions in a survey? How do you organize and prioritize questions in a survey? What problems do you encounter when writing questions in a survey?”
7. Your role as an interviewer is not to get buy-in or support for a project.
Remember, you are there to learn. Be open to the information that comes your way.
8. Don’t overly smile, affirm, or otherwise encourage the person you are interviewing to go down a certain path.
People want to please, and if you show that you like what they are saying, they may just say what they think you want to hear.
Though user testing, or usability testing, is not the same as interviewing, they share many of the general principles: