How to transcribe a focus group for analysis
Short answer
Record with one good recorder in the center of the table (or the video call), ask participants to say their name before speaking, then upload it. Label the moderator and participants with pseudonyms, merge split speakers, and export to Word or TXT for coding.
A focus group is one of the hardest recordings to transcribe: six to ten people, overlapping answers, a moderator steering the conversation, and sessions that run 90 minutes or more. Done by hand, transcribing one session can take a full working day.
This guide covers how to record a focus group so the transcript comes out usable, how to get it transcribed by speaker in minutes, and how to prepare it for qualitative analysis.
Transcribe a focus group in 3 steps
- 1
Record for transcription
Use one good recorder in the center of the table, or record the video call for remote groups. Ask participants to say their first name (or participant number) before they speak the first time, and to avoid talking over each other.
- 2
Upload and label the speakers
Upload the recording (audio or video, up to 4 hours). Each voice is separated automatically; label the moderator first, then each participant with a pseudonym or number. If someone was split into two labels, merge them.
- 3
Export for coding
Export to Word to code by hand or to TXT to import into your qualitative analysis software. Every line carries the speaker and a timestamp so you can go back to the audio.
What makes focus groups hard
- Many speakers with short turns, which makes attribution harder.
- Crosstalk and laughter, where any transcription loses words.
- Participants far from the microphone.
- Long sessions with several activities.
Before you upload: ethics and consent
Check what your consent forms and ethics approval (for example, your IRB) say about recording and about using cloud transcription services, and tell participants how the recording will be used. Use pseudonyms or participant numbers as speaker names rather than real names. In Dolcenota recordings are stored encrypted, never used to train AI models, and you can delete them when you're done.
Working with the transcript
- Start with the summary for an overview of the main themes.
- Ask questions across the session ("what did participants say about price?") to find passages to code.
- Check key quotes against the audio before citing them.
- Combine the files of a session recorded in parts into one transcript.
Frequently asked questions
Can it separate 8 or 10 participants?
It separates distinct voices automatically. With many speakers and short turns, expect to review and correct some labels; introductions at the start help.
Can I use pseudonyms instead of names?
Yes. Label speakers with any name or code, such as P1, P2, and it's used in the transcript and exports.
Which format should I export for analysis software?
TXT imports cleanly into most qualitative analysis tools; Word is better for coding by hand.
Is it accurate enough for research?
It's a strong first draft. Review it against the audio, especially overlapping speech and key quotes.
Does it work for focus groups in Spanish?
Yes. English and Spanish are detected automatically.
Transcribe your next focus group
Upload the session and get a transcript by speaker, ready to code. Free to start.
Start free