Reference (partial) — Options below match the Add Question dialog and question editor enums (June 2026). Per-brand screenshots for full-screen vs in-chat layouts are still pending.

Overview

When you add a question via the Question button, you choose three things: the question type (what the answer is used for), the response type (how the member answers), and the layout (how the question is displayed).

After you click Add, the question appears on the step list. Open it from there to configure answers, algorithm settings, pre-validation, and post actions on the question editor.

This page is a complete reference for all available options in the three add-dialog dropdowns.

Add question dialog

Opened from Question on the questionnaire step list.

FieldWhat to enterEffect after Add
Question name The question text members will see (required). Creates a new question row and appends it to the step list. You can refine wording later in the question editor.
Response type How the member answers—pick one, type text, upload a photo, and so on. See Response types. Controls the input widget in the member app and how answers are stored on the profile.
Question type What the answer is used for—bio, preference, predictor, other, or AI follow-up. See Question types. Feeds matching (bio vs preference), predictor jobs, or operational storage only.
Layout In Chat or Full Screen. See Layout types. Controls whether the question appears inline in Lora chat or opens a dedicated full-page view.

Cancel closes without saving. Add is enabled only when question name is filled in.

Type pairing rules

Some question types only work with certain response types. Wrong pairings may save in Admin but fail or behave oddly in the member app.

Question typeRequired response typePlain-language note
Predictor Scale only Predictor scoring reads scale answers and linked preference relations. Other response types are not processed by the predictor service.
AI Follow-Up Text only The server injects additional Lora messages until the free-text answer is sufficient. Configure the AI agent on the question editor after creation.
Bio, Preference, Other Any response type Pick the widget that fits the answer shape. Bio and Preference answers feed matching; Other does not affect scores directly.

Question types

Controls what the answer is used for.

TypeWhat it meansWhen to chooseMember / matching effect
Bio Stores a fact about this member (for example, their height, location, or denomination). Any question where the answer describes who the member is. Bio answers are compared against other members’ Preference answers when calculating match scores and deal-breakers.
Preference Stores what the member wants in a partner (for example, preferred height range, preferred age range). Any question about partner preferences. Preference answers are compared against other members’ Bio answers. Deal-breaker flags on linked attributes can block suggestions.
Predictor A scale-based question whose answers feed a separate predictive compatibility model. Use only with a Scale response type. Requires predictor relations and brand predictor settings. When predictors are enabled for your brand, predictor scores appear in matching and on potential-match screens.
Other The answer is stored on the profile but does not directly contribute to matching scores. Operational or informational questions that you need to collect but that are not part of compatibility calculation. Visible on the profile; does not drive bio/pref score columns.
AI Follow-Up A free-text question where the AI guide asks follow-up questions until enough information is collected. Use only with a Text response type. Member sees the initial question, then additional Lora chat lines from the configured agent until the server marks the answer complete.

Response types

Controls how the member answers the question.

TypeWhat the member doesStored on profileSave / matching notes
TextTypes a free-text answer (optional voice when brand enables transcription).String or string array when split by comma is on in response settings.Required for AI Follow-Up. Sensitive attributes encrypt at rest. Identity fields may also update login credentials on some brands.
Single ChoiceTaps one option from a list.Stored answer key on linked profile attribute.Most common bio/pref type. Use same keys in Singles Advanced Search filters.
Multiple ChoiceTaps one or more options.Array of answer keys.Supports forbidden pairs (two options cannot both be selected). Auto-response rules may append more keys on save.
RangeSelects min and max (slider or dialog per range adjustment mode on the question editor).Typically two numeric values or encoded range on the attribute.Common for age/height preferences. Matcher reads bio/pref relations for scoring.
ScaleSlides along a labeled scale.Numeric scale value.Predictor questions only. Feeds predictor jobs and potential-match predictor column when enabled.
File PickerUploads a file (certificate, document).File reference on the attribute.verified — Server runs a dedicated upload handler after save; not stored as plain text in the attribute blob alone.
ListScrolls and picks one row from a long list.Answer key.Same storage as single choice; different mobile widget for long lists.
Multiple Choice TagsTaps tag chips; multiple allowed.Array of keys.Visual variant of multiple choice—filter with Contains / In.
Single Choice DropdownOpens dropdown; picks one option.Answer key.Compact UI for long option lists in chat or full-screen layout.
Date PickerPicks a date from a calendar.ISO date string (often on birthmonth or similar attribute).Use Date after/before operators in search—not numeric Bigger/Smaller on the date string.
LocationConfirms city/region via location services or picker.Structured location object on the attribute (city/state/country fields).verified — Server normalizes location through a dedicated converter before persisting. Prefer Full Screen layout for complex pickers.
AutocompleteTypes and selects from suggestions.Display text or canonical value per brand rules.verified — Closed-name aliases may rewrite values before save (for example institution nicknames). Notify on new response on the attribute can alert admins when a new free-text value appears.
Image PickerUploads or selects a profile photo.Image id / URL on photo attribute.Triggers photo review workflows when your brand enables reviewed-photo flags.
BooleanTaps Yes or No.Stored true/false keys—not the button labels.Match filter using the stored keys, not localized Yes/No text.
Custom FormMulti-field form defined in response settings.Structured object per form schema.Confirm member UX with your brand lead; often Full Screen only.

After save, answers appear on the member profile under the linked attribute name. Bio and Preference types feed the matcher; Other does not change bio/pref score columns directly.

What happens when a member answers

End-to-end effect after the member taps submit (chat, slides, or full-screen page).

  1. Validate — The app checks required selections, min/max counts, and forbidden combinations before sending.
  2. Save response row — A row is stored in questionnaire response history (used for Res Rate and audit).
  3. Update profile attribute — When the question is linked to a profile attribute, the answer is written to that member's stored attributes (encrypted when the attribute is marked sensitive).
  4. Auto-responses — If you configured automatic follow-up answers on the question, additional attributes may be filled in the same save.
  5. Special handlers — File uploads, location normalization, and autocomplete institution aliasing run on the server during this step.
  6. Advance flow — Chat questionnaires emit the next Lora message; slide decks advance the pager; full-screen questions return to chat after submit.
  7. Batch completion — On the last active step, the assignment may mark the questionnaire complete (except mid–AI-follow-up turns).

Matcher refresh: Bio/pref changes eventually affect suggestions on the next matching run—they are not instant on every save unless your brand triggers a separate refresh job.

Configure links on the question editor: profile attribute, deal-breaker flag, algorithm tab, auto-responses, and post actions.

Skip, empty, and system-skipped answers

SituationWhat happensRes Rate / profile
Member taps Skip on an optional question Response saved with skipped flag where applicable; profile attribute is not overwritten if it already had a value. May count as answered for analytics depending on step; attribute stays unchanged when skip would write empty.
Pre-validation hides a step (chat) Step never offered; no member-facing error. Hidden steps do not appear in the thread; see pre-validation walkthrough.
Pre-validation hides a slide at open Slide omitted from deck; on finish, hidden questions may receive a system skipped marker. Profile records that the step was not shown—see slide vs chat.
Empty text submit on sensitive attribute verified — Empty/skipped answers do not clear sensitive values accidentally. Prior value retained.

Layout types

Controls how the question is displayed to the member.

LayoutMember experienceWhen to choose
In Chat The question appears as a chat message in the Lora conversation. The member answers inline without leaving the chat view. Default for most onboarding chat questions. Feels natural in a conversational flow.
Full Screen A button appears in the chat that opens the question on a dedicated full-screen page. After answering, the member returns to the chat. Use for complex questions with long answer lists, date pickers, location inputs, custom forms, or rich single-choice layouts where chat width is too narrow.

Slide-style onboarding: On On Boarding Slides questionnaires, layout still affects published JSON, but the member experience is the slide pager rather than Lora chat bubbles—see questionnaire types and visual guide.

Question answered inline inside Lora chat with chip options
Waiting for screenshot. Capture: In Chat layout—question with inline chips or text field inside the Lora thread (slug questionnaires-in-chat-question).
Open question button in chat leading to full-screen answer page
Waiting for screenshot. Capture: Full Screen layout—Open question button in chat, then the dedicated full-page answer screen (slug questionnaires-full-screen-question).