Conduct interviews to gain qualitative insights that explore things you might not have known.
Interviews are exploratory generative research, meaning that the insights from them might be complete news to you. Interviews should be largely informal conversations with individual members of your target audience.
Use quantitative surveys to generalize existing insights or beliefs about your audience’s demographics or attitudes.
Surveys allow you to confirm existing assumptions or beliefs with a broad subsection of your target audience, as long as that subsection adequately represents your broader audience.
Set your interview and survey goals based on the intended takeaways from your research. Directly relate these to both your business and research goals.
Examples of interview and survey goals include uncovering insights around your branding, or gaining attitudes towards a specific product line within your industry.
Create an interview data collection template based on your interview goals that allows you to plot in answers and analyze insights.
A data collection template also enables you to stay on track during the interview. Include:
- Basic information about your interview subject, including their name, contact information, employment information, years of experience, and industry.
- Information about your interview goals, like product usage or brand attitudes.
- Sufficient space for raw notes, covering topics that arise in the course of the interview.
Review your data collection template against your interview goals to make sure you cover all topics necessary to achieve those goals.
Try to prevent bias by asking questions with both a negative and positive slant. Use objective, simple language.
Avoid common question pitfalls, such as leading with a positive or negative question, using jargon or acronyms, asking double barreled questions, or asking users to predict their behavior.
Skip subjective language, which can push people in specific directions unintentionally. For example, ask tell me about your experience trying to do X rather than did you enjoy the process of trying to do X.
Practice engaging the user in natural conversation by conducting the interview without a script or consulting written topics and asking why questions, listening to responses, and following up with additional questions.
Interviews need to be natural conversations for maximum insights. Review recordings of your past interviews to see what you did well and where you could improve in engaging your audience.
Ask permission to record the interview and consider bringing a second person to the interview to take notes as a backup.
It is illegal in some states to record someone without their permission. But even in states where it is legal, asking permission is respectful and establishes credibility with your interview subject. A dedicated notetaker acts as a backup in case recording technology fails, while still allowing you to focus on the conversation.
For surveys, choose an audience sample that is both close enough to your audience’s demographics and large enough to be representative.
For example, a total audience population of 1,000 will have a margin of error of 10% if you ask 88 representative members, but only 5% if you ask 278 members. Use an online sampling calculator to estimate how large your survey audience needs to be.
Avoid common survey errors:.
- Coverage errors occur when you are not able to survey a representative sample of the population you are studying.
- Sampling errors arise from not measuring everyone in the target population, so your results could be far from the average you would get when running the survey hundreds of times.
- Measurement errors occur when the way you choose to measure something leads to inaccurate data, like asking poorly worded questions.
- Non-response errors occur when some groups within your survey complete the survey less than other groups.
Build a feedback protocol that allows internal stakeholders to provide productive input and improve your user research by refining the questions you ask over time.
Identify what stakeholders could benefit from your research. Then, provide them with a preferred way to give you actionable, specific feedback. The more clearly stakeholders like product engineers, sales agents, etc. can identify how to improve on your existing research, the more actionable your research can become over time.