With Research Flow, you can run AI-moderated research sessions through voice, voice and video, or text. Each format gives respondents a different way to share their perspectives, and the AI moderator can ask follow-up questions in all three.
The right format depends on your research goals, your audience, and the type of feedback you want to capture. Voice can encourage natural conversation, video adds visual context, and text lets respondents take part without speaking aloud or being on camera. This article explains when to use each format, so you can choose the one that fits.
To learn where to change the format, see How to provide context to your research study. To see what each format looks like for respondents, see What respondents experience in a Research Flow study.
If you're interested in Typeform's Research Flow for creating research studies, please reach out to our Sales team.
Compare session formats
| Voice | Voice and video | Text | |
| What respondents do | Speak their answers | Speak their answers with their camera on | Type their answers |
| What you receive | Audio recording, transcript, and highlights | Full video recording, transcript, and highlights | Text responses and highlights |
| Permissions needed | Microphone | Camera and microphone | None |
| What responses look like | Detailed and natural | Detailed and natural, plus facial expressions and reactions | Usually shorter and more considered |
| Best for | Customer discovery, product feedback, and exploring motivations and experiences through natural conversation when you don’t need to see respondents | Everything voice suits, plus concept and creative feedback, visual reactions, and interviews where seeing the respondent adds useful context | Sensitive topics, loud or public environments, and people who prefer to respond in writing |
When to use voice
Voice works well when you want detailed answers in a natural conversation. Speaking feels easier than typing, so respondents tend to say more. And without a camera, some may feel more relaxed. For example:
- Understanding why customers churn, upgrade, or choose a competitor
- Exploring experiences, opinions, and decision-making
- Onboarding or product feedback interviews
Keep in mind: Respondents need to allow microphone access and be somewhere fairly quiet. You won't get video clips in your results, but you'll still get audio recordings, transcripts, and highlights.
Tip! Turn on the Agent voice setting so the AI moderator reads questions out loud. This makes a voice session feel more like a real conversation. See How to provide context to your research study.
When to use voice and video
Voice and video is the closest to a live interview on a video call. You get the same detailed, natural answers as with voice, and you can see respondents' expressions and reactions too. For example:
- Gathering reactions to visual concepts, designs, or creative assets presented through Stimuli, where seeing respondents' expressions adds context to their verbal feedback.
- Exploring reactions to messaging, branding, or experiences when researchers want to see and hear respondents explain their perspectives. Watching session recordings to observe respondents' visible reactions and add context to their spoken feedback.
- Creating video clips and highlight reels to share findings with stakeholders.
Keep in mind: Some respondents aren't comfortable turning on their camera, and they need to allow camera and microphone access before the session starts. Tell respondents upfront that the session uses video. If respondents have trouble with permissions, share How to grant camera and microphone access for Research Flow studies.
When to use text
Text works well for audiences who would rather not be on camera or talk out loud. For example:
- Respondents answering in a language that isn't their first, who may want time to find the right words.
- Respondents who prefer to take their time, to read and write, or who find speaking harder.
- Sensitive or personal topics, where some people prefer to type and to review their answer before sending it.
- Respondents who may be at work, in public, or anywhere they can't talk.
Keep in mind: Text responses may be more concise than spoken answers, but the AI moderator can still ask contextual follow-up questions to explore participants' reasoning, clarify feedback, and uncover additional detail.
Questions to help you choose
Answer these questions to help you choose a session format:
Tips for any format
- Set expectations. Say which format respondents will use, and roughly how long the session takes, on your Welcome screen and in your invitation. See Best practices for inviting your own customers/contacts to a Research Flow interview.
- Choose before you share. You can change the format after launch, but switching mid-study, for example from Voice and video to Text, can create a mixed dataset.
Note! Your session format doesn't limit the questions you can ask. If you choose Voice or Voice and video, you can still add Open text, Multiple choice, and Dropdown questions. See How to add tasks to your research study.
Note! If you choose Voice or Voice and video, you may occasionally receive a text response. This happens when a respondent has a poor internet connection. The session switches to text so no data is lost.
FAQ
Can I change the session format after I publish my study?
Yes, but changing the format mid-study, for example from Voice and video to Text, can create a mixed dataset. If you're comparing results, it's best to settle on one format before you share your study.
Can respondents choose their own format?
You set the session format when you create your study and it applies to all respondents. This keeps everyone's experience the same, which makes your results easier to compare. The only exception is a respondent with a poor internet connection, whose session switches to text so no data is lost.
Can I still ask text based questions in a voice or voice and video study?
Yes. The session format sets how respondents take part in the conversation. You choose the answer format for each research question separately, so you can add Open text, Multiple choice, and Dropdown questions in any format. See How to add tasks to your research study.
Do respondents need to allow camera or microphone access?
It depends on the format. Voice and video sessions need camera and microphone access, Voice sessions need microphone access only, and Text sessions need neither. See How to grant camera and microphone access for Research Flow studies.
More resources
- How to provide context to your research study
- What respondents experience in a Research Flow study
- How to grant camera and microphone access for Research Flow studies
- Best practices for inviting your own customers/contacts to a Research Flow interview
- Recruit respondents to take your research study with a panel