Voice interviews: why participants who speak tell you more

An in-depth interview lives on what people volunteer. That's exactly where the typed interview has a built-in brake: writing is work. Anyone answering on their phone cuts it short — and the most interesting half-sentence falls victim to the typing.

The typing barrier

• Typed answers are short — side thoughts get dropped before they're ever spelled out.
• On mobile, the risk of abandonment rises with every question that demands a long answer.
• Writing is edited: anything that seems rude or messy gets smoothed over before it's sent.

What changes when people speak

In a voice interview the agent asks the question out loud, and the person answers the way they'd actually talk — with false starts, corrections, and exactly the asides where the real reasons hide. Spoken language is less edited, and that matters for research: the first thought comes along with it.

„Oh, and one more thing — this only just occurs to me: the main reason was actually something completely different …“

— From a voice interview, transcribed live

Nobody types sentences like that. People say them all the time.

What matters technically

• The transcript stays the source of truth — every analysis points back to the exact spot; voice changes nothing about the evidence requirement.
• Interrupting has to be allowed: someone who cuts the agent off is engaged — the system should listen, not keep talking.
• Typing stays available at any time — not every environment suits speaking, and switching should cost nothing.
• Audio is especially personal: EU hosting and clear deletion windows aren't optional extras with voice, they're a precondition.

Voice doesn't replace the typed interview — it lowers the barrier for everyone who finds talking easier than writing. And outside our bubble, that's most people.