StorySync AI Qual is a group of 9 voice-, video-, and image-first question formats in the palette's StorySync AI Qual group. They turn open-ended feedback into a guided, conversational response: the respondent speaks (or records video/photos), the platform transcribes and structures it, and — for the AI-moderated formats — can generate follow-up probes automatically instead of asking one fixed question.
Seven of the nine formats are configured through the StorySync wizard — a dedicated multi-step setup flow instead of the regular properties panel. The remaining two (Tap & Talk, Slide & Speak) are simpler image/video-tap-and-record interactions configured directly in the properties panel.
The StorySync wizard#
Adding one of the seven wizard-driven formats (Guided Story, Guided Story With Video, AI-Moderated Interview, Guided Story + Segmentation, Guided Story With NPS/CSAT, Guided Story With Image Capture, Pantry Checker) opens a configuration wizard instead of the plain question card. All seven wizards share the same three-step shape:
- Questions — greeting/instruction screens, the opening prompt(s), sub-questions or a discussion guide, and Guardrails (a free-text list of topics the AI should avoid steering toward in follow-ups).
- Display Settings — voice/media behavior: mandatory vs. optional voice, transcription, auto-probing, recording duration limits, language, text direction, button color, and (where relevant) camera facing, image scrolling, and image-height sizing.
- Text for Alerts — every button label and error/retry message the respondent can see (recording states, mic-permission errors for both mandatory and optional voice, "try again" copy).
A few settings worth knowing about because they change behavior significantly:
- Honor voice-permission globals (on by default) — makes the question follow the survey-level voice/mic-permission setting instead of its own Voice Mandatory toggle, and auto-skips the question if voice turns out to be unavailable. Recommended unless you have a reason to override per question.
- Enable Auto Probing — lets the AI generate its own follow-up questions from the respondent's answer instead of only asking your fixed sub-questions.
- Guardrails — steers what the AI does and doesn't probe into; leave it blank if you don't need topic restrictions.
- Voice or Text Preference — when voice isn't mandatory, you can ask respondents up front whether they'd rather answer by voice or by typing.
Guided Story#
Full guide: Guided Story →
Plugin type: unified_voice · Wizard-driven
The base voice-first format: the respondent sees a prompt and a large microphone button, speaks their answer, and can see a live transcript as they talk (with the option to re-record). One or more sub-questions can be asked in the same flow, each with an optional stimulus image or video.
When to use it: Open-ended feedback where a spoken answer feels more natural than typed text — general customer research, concept reactions, or anywhere you'd otherwise ask a long-text question.
Guided Story With Video#
Full guide: Guided Story With Video →
Plugin type: unified_camera_video · Wizard-driven
Same conversational flow as Guided Story, but the respondent records camera video instead of (or alongside) voice. Adds a camera-facing setting (front/rear) on top of the shared voice settings.
When to use it: Studies that want to see the respondent's face or environment while they answer, not just hear them.
AI-Moderated Interview#
Full guide: AI-Moderated Interview →
Plugin type: mini_idi · Wizard-driven
A short, AI-guided interview: instead of a fixed list of sub-questions, you set Objectives, a Discussion Guide (either as separate question fields or one free-form document the AI splits by paragraph), a target interview length, and two sliders — Moderator Guidance (stick-to-script vs. explore-openly) and Moderator Style (friendly/informal vs. neutral/professional) — that shape how the AI moderator conducts the conversation.
When to use it: Replacing a scheduled in-depth interview (IDI) with an always-on, scaled equivalent — exploratory research, concept testing, or jobs-to-be-done studies where a live moderator normally would adapt questions in real time.
Guided Story + Segmentation#
Full guide: Guided Story + Segmentation →
Plugin type: unified_single_multiple · Wizard-driven
Combines the voice narrative with a single/multiple-choice follow-up: the AI reviews the spoken answer and offers or scores it against a set of choice options you define, in the same frame as the recording.
When to use it: When you need both an open narrative and a structured classification of that narrative — e.g., "tell us why, then we'll bucket you into a segment" — without a separate follow-up question.
Guided Story With NPS/CSAT#
Full guide: Guided Story With NPS/CSAT →
Plugin type: nps_csat · Wizard-driven
Wraps a scored question — NPS (0–10, range configurable) or CSAT (1–10 scale, default 5-point) — with a spoken follow-up narrative explaining the score. You choose NPS or CSAT, whether respondents can change their scale selection, and the initial scoring prompt (with optional brand logo).
When to use it: Anywhere you'd run a standard NPS or satisfaction question but want the "why" captured as rich voice narrative instead of a short open-text box.
Guided Story With Image Capture#
Full guide: Guided Story With Image Capture →
Plugin type: unified_image · Wizard-driven
Guided camera capture (up to 4 images, each with its own instruction text, an optional overlay template, orientation requirement, and custom capture rules) with an optional voice narrative recorded after the photos.
When to use it: Visual research that also wants qualitative context — "show us your pantry/desk/closet and tell us about it" — without splitting it into two separate questions.
Pantry Checker#
Full guide: Pantry Checker →
Plugin type: pantry_checker · Wizard-driven
A purpose-built image-capture format: respondents take a flexible number of photos (you set a minimum and maximum, 1–4) with a "No More Photos" early-exit button that only appears once the minimum is reached, then optionally answer a voice follow-up. Each image slot supports its own instruction text, overlay template, orientation requirement, and image tags. Adds camera-specific display options (auto-capture, real-time image guidance, allow image upload as a fallback, AI source for guidance).
When to use it: In-home usage or pantry/product-presence audits where the number of relevant items varies by respondent, so a fixed photo count doesn't fit.
Tap & Talk#
Full guide: Tap & Talk →
Plugin type: tap_and_talk · Not wizard-driven (configured in the properties panel)
Respondents tap a concept image to register like/dislike, then record a short voice note explaining the tap — configured with separate like and dislike instruction text (and optional "next tap" follow-up instructions shown if the respondent taps again in the same phase).
Key settings: Concept image URL (supports piping and library upload), like/dislike instruction rich text, next-like/next-dislike follow-up instruction rich text, max taps per phase (1–20), max recording duration (10–600 seconds).
When to use it: Fast reaction-plus-reason capture on a single concept image — quicker to build than the full StorySync wizard when you don't need AI-generated probes.
Slide & Speak#
Full guide: Slide & Speak →
Plugin type: slide_and_speak · Not wizard-driven (configured in the properties panel)
A video slider the respondent scrubs through while recording a spoken reaction — pairs a video stimulus with voice narrative instead of a static image.
Key settings: Show instruction screen toggle, max recordings per phase (1–20), max recording duration (10–600 seconds), video max height (20–80% of viewport).
When to use it: Video-stimulus reactions where you want the respondent talking through the video as they watch, rather than answering after the fact.
Related guides#
- Question Types — overview and when-to-use table
- Custom Interfaces — mobile-first tap/swipe/drag plugin formats
- Display Blocks & Plugins — Text/Graphic and global-variable blocks
- Text & Contact Questions — standard long-text as a non-voice fallback
- Platform Feature List — plugin inventory and status
Was this page helpful?