Question types

Nineteen question types are available in the survey builder, plus a display-only element for text and images. This page covers what each one does, when to reach for it, and the settings that matter. For an overview of the builder itself, see Survey Builder.

All of them are available on every plan. The Free plan caps a survey at 10 questions; paid plans are unlimited. Quiz scoring is available on multiple choice, image choice, rating, and matrix.

Adding a question

1. Open a survey and go to Design.
2. In the left sidebar, select the Elements tab. Question types are grouped under Elements and Other types.
3. Drag a type onto the canvas to place it exactly where you want it, or double-click to add it to the end of the current page.
4. Hover the new question and click the pencil (Click to edit) in its toolbar. Its settings open in a panel on the right — that is where the options described below live.

You can also add a question from the canvas itself: hover between two existing questions and click the + that appears on the divider.

Hovering the ? beside any type in the sidebar shows a short description and a sample of what respondents will see. Clicking it opens that type’s section on this page.

Text input

An open-text field. Four formats: single line, paragraph, password, and multiple single lines (several labelled boxes under one question). Optional validation covers email, number, decimal, date, phone, postal code, US state, and length limits, and you can set a character limit with a live counter.

Use it when you can’t anticipate the answers — verbatim feedback, job titles, order numbers. Open text takes longer to answer and longer to analyse than a closed question, so it earns its place when the answer set is genuinely unknown.

Multiple choice

A list of predefined choices, set to one answer or many. Choices can display as a list or a dropdown, in one or more columns, with an optional “Other” write-in.

Use it when the answer set is known and you want countable data instead of free text. If you find yourself adding an “Other” box that most people use, the choice list is probably missing something — worth a round of open text first to find out what.

Image choice

Choices presented as images rather than text, single or multi-select, in a configurable column layout. Images can be sized to fit or fill their tile.

Use it when the thing being judged is visual — packaging, ad creative, layout options — or to keep a question light on language for international audiences.

Matrix

A grid that asks the same scale of several statements at once. Cells can be radio buttons (one answer per statement), checkboxes (several), a dropdown, a rating control, or free text. Scales can come from the built-in standard sets — agreement, satisfaction, frequency and others — or be written by hand, and the grid can be transposed, reversed, spaced out, or set to repeat its header row on long grids.

Use it for Likert batteries and any set of items sharing one scale. Keep the rows few — a long grid is hard work on a phone, and repeating the header row treats the symptom rather than the cause.

Text or image

A display-only block of rich text and images. It collects no response and cannot be piped into later questions.

Use it for section introductions, instructions, consent text, or a stimulus image that the questions after it ask about.

Slider

A draggable numeric scale — 0 to 100 by default, with configurable range and step size — that can carry several statements at once. An optional “Not applicable” control sits alongside it.

Use it when the exact number is the finding: budget percentages, price expectations, effort estimates. If you only need an ordinal read (“how satisfied?”), a rating scale is easier to answer and easier to compare.

Ranking

Respondents put items in order, either by drag-and-drop or through a ranking grid, with rules for how many items must be ranked.

Use it when relative order is the finding — top-three priorities, feature preference — and a rating scale would let everything tie at “very important”. Every extra item makes the ordering harder to hold in your head, so for a long list consider constant sum, which asks for a judgment per item rather than one judgment across all of them.

Group rank

Respondents drag items into groups and, optionally, rank the items inside each group. Supports up to 15 groups and 50 items, with rules for how many items must be placed.

Use it for card-sort work: sorting features into must-have and nice-to-have, or clustering ideas before ranking them. It answers “how do these relate?” where ranking only answers “which comes first?”.

Constant sum

Respondents divide a fixed total across several items, using either numeric inputs or sliders. An optional must-total rule holds the answer to a specific sum — typically 100.

Use it to force trade-offs, where a rating scale would let everyone call everything important. Budget allocation, time allocation, and preference distribution are the usual cases. Say what the total represents in the question text; “distribute 100 points” is clearer than “distribute 100”.

Net Promoter Score

The standard eleven-point scale (0–10) asking how likely someone is to recommend you, with Promoter, Passive and Detractor grouping applied automatically in reporting. The subject of the recommendation and the endpoint labels are configurable.

Use it once per survey as a relationship benchmark, and follow it with an open text question asking why. The score tells you the direction; the verbatim tells you what to do about it. Net Promoter Score and NPS are registered trademarks of Bain & Company, Satmetrix Systems and Fred Reichheld.

Yes or No

A binary choice, with a choice of icon pairs: like/dislike, check and cross, smiley/frowny, heart, or plus/minus.

Use it for screeners and qualifying questions, especially where the answer drives skip logic. Where a real middle ground exists, a three-point scale collects it instead of forcing a side.

Email

A single-line field validated as an email address.

Use it when you need to follow up, de-duplicate respondents, or tie answers to a contact record. Asking for an email address changes what people tell you — if the survey is meant to be anonymous, say so, and consider whether you need the address at all.

Date

A date picker in US (MM/DD/YYYY), European (DD/MM/YYYY) or ISO (YYYY-MM-DD) format, with optional 12- or 24-hour time.

Use it when the answer is a point in time — purchase date, event attended, appointment preference — so that results sort and filter properly instead of arriving as free text.

Rating

A scale of 2 to 10 points, drawn as stars, thumbs, hearts or emoji, with optional endpoint labels, a custom colour, and a “Not applicable” option.

Use it for a quick read on a single thing. Keep the same scale length across a survey, and across repeat waves of it: a five-point answer and a seven-point answer are not directly comparable, so changing the length mid-programme breaks comparability with everything measured before it.

Photo submission

Respondents take a photo with their device camera or upload an existing image. Three capture methods: camera and upload, camera only, or upload only. Files upload directly to storage.

Use it for field evidence — shelf shots, damage claims, receipts, before-and-afters. Camera-only is the setting to reach for when the photo needs to be taken at the moment of answering rather than found in a camera roll.

Video response

Respondents record video in the browser or upload an existing file, with configurable minimum and maximum durations (defaults: 5 and 45 seconds).

Use it for testimonials and usability walk-throughs, where seeing the respondent is the point. Recording on camera asks a great deal more of someone than typing does, so keep the maximum duration short — a 45-second ask reads as manageable in a way that a three-minute one does not.

Audio response

Respondents record audio with their device microphone or upload an existing audio file. Three capture methods: record and upload, record only, or upload only.

Use it when tone and detail matter more than a typed answer — spoken verbatims, pronunciation checks, or open ends on mobile, where the keyboard is the constraint.

Signature

A signature captured by drawing on a canvas, typing a name rendered in a signature font, or uploading an image of one.

Use it for consent, waivers and acknowledgements that need to sit on the response record. A typed or drawn signature is evidence of agreement, not a qualified electronic signature — if you need the latter, use a dedicated e-signature service.

File upload

Respondents attach files — documents, spreadsheets, images, presentations, archives — with configurable file types, a size limit between 1 MB and 50 MB (10 MB by default), and single or multiple uploads.

Use it when the answer is a document: résumés, invoices, entry submissions, supporting evidence. Name the accepted formats in the question text; the picker enforces them, but people would rather know before they hunt for the file.

CAPTCHA

A Cloudflare Turnstile challenge, verified server-side, with configurable theme, size and alignment.

Use it on public links and incentivised surveys, where bot traffic would contaminate the data. It isn’t needed on email invitations sent to a known contact list, and every extra step costs completions — add it where the risk is real rather than by default.