A 42 NPS tells you something is wrong. It doesn't tell you whether it's onboarding, support, pricing, or something you haven't thought of. The follow-up open-end is supposed to answer that — but typed responses give you "bad service" and "could be better." Voice responses give you the full story.
Typing feels like work. When asked "why did you give that score?", most detractors write one sentence. Most promoters write nothing at all.
"Support wasn't great" and "the agent was rude and I had to call three times to solve one issue" are the same sentiment — but only one tells you what to fix.
Tone, urgency, and emotion are lost in text. Two analysts reading the same response often reach different conclusions.
Example responses — illustrative
Before — typed open-end
"Good product but support could improve."
6 words. Promoter. Score: 8.
After — voice response, auto-transcribed
"I love the product itself, honestly I'd recommend it to anyone. The one thing holding me back from giving a 10 is that the last time I had a billing issue it took four days to resolve — I sent three emails and had to follow up each time. If support was faster I'd be a 10 without question."
Promoter. Score: 8. Now you know exactly what to fix.
Works in Alchemer, and any survey tool that allows custom HTML questions. No developer needed.
One button. No app installs, no account creation. Works on mobile. If they prefer not to speak, the text fallback is always available.
Every voice response is transcribed automatically and appears in your survey results as text — ready to read, export, or analyze.
Voice Capture handles collection and transcription. Once you have a set of verbatims, Survey Coder Pro (part of the same suite) can code them automatically — grouping detractor responses by theme, identifying the top drivers of your score, and giving you a structured breakdown in under an hour.
Learn more about Survey Coder ProVoice Capture embeds directly into your existing survey via a one-line HTML snippet. No platform migration required.
Start with 250 free credits, then buy a one-time credit pack when you have a project. No subscription, nothing to cancel, and unused credits roll over to your next study.
See all plansNo credit card · One-time credits, never expire · GDPR compliant · EU database
No — voice is offered alongside the text box, never instead of it. Respondents who prefer to type still type. In practice, completion rates tend to hold or improve because speaking is lower-friction than writing a paragraph.
No. Voice Capture runs in the browser. Respondents click the microphone button and allow microphone access when prompted — the same permission any website asks for. No apps, no plugins, works on mobile.
Within seconds of the respondent finishing. The transcription is produced automatically via OpenAI Whisper and appears in your dashboard as text — ready to export to CSV alongside the rest of your NPS data.
No. Voice Capture processes audio in memory for transcription and discards it — only the text is kept, in EU-based infrastructure. That keeps your data-protection footprint small and is GDPR-friendly by default.
Voice Capture gives you the transcripts; for automatic theme coding and driver analysis, pair it with Survey Coder Pro, a companion product in the same suite built specifically for analyzing open-ended responses.
Collecting broader customer feedback, not just NPS? See our voice of customer surveys page
We use cookies to improve your experience and analyze site traffic. Learn more