> ## Documentation Index > Fetch the complete documentation index at: https://docs.exoid.ai/llms.txt > Use this file to discover all available pages before exploring further. # Question types > Reference for the 17 block types: what each one asks, every setting it exposes and the research job it does. The block type decides two things: how the respondent interacts with the question, and what shape the data takes in analysis. You add blocks from the **+** menu on the canvas. This page is the reference for all **17 question types** — what each one does, every option it exposes, and the research case it fits. Beyond the 17 types there are two screens that aren't blocks you add — the **start screen** and the **Ending Screen** — plus **Create in bulk**, which is an action rather than a type. All three are covered at the bottom of the page. ## Quick orientation | Block | What the respondent does | Use it for | | --------------- | -------------------------------------------------- | ----------------------------------------------------------------------------- | | Multiple choice | Ticks one or more options. | Answers that can be true at the same time: brands known, channels used. | | Single choice | Picks exactly one option. | Mutually exclusive options: a preference, a segment, a screener answer. | | Slider | Drags a handle along a scale. | Intensity or agreement when a five-point scale is too coarse. | | Swipe | Rates images one by one with **LIKE** or **NOPE**. | Fast gut reaction to a series of visual stimuli. | | Open ended | Types free text. | Reasons and wording in the respondent's own words. | | Yes/No | Picks one of two labelled answers. | Screeners and binary filters early in the interview. | | Rating | Picks a point on a star or numeric scale. | Satisfaction, liking, recommendation — including NPS. | | Email address | Types an email address. | A contact for recontact, a prize draw or a follow-up. | | Date and time | Picks a date, optionally a time. | Purchase date, appointment, when an event happened. | | File upload | Uploads one or more files. | Visual proof: a receipt, a pack shot, a screenshot. | | Message | Reads. There is no answer field. | Section intros, instructions, a stimulus shown before the questions about it. | | Matrix | Answers several rows sharing the same columns. | Batteries of scales: many items rated on one scale. | | Dropdown | Picks one option from a collapsed list. | Long option lists: countries, cities, categories. | | Ranking | Drags items into order. | Relative priority, when order matters more than score. | | Number Input | Types a number. | Age, quantity, frequency, amounts used in calculations. | | Image Choice | Picks from a grid of images. | Pack tests, visual concepts, forced-choice on creative. | | Website | Types a website address. | Collecting a site URL from the respondent. | In the Italian interface **Dropdown** is called **Lista**. It is the same block. ## The 17 types in detail Every block also carries the settings shared across types — question code, question text, description, media, wait before advancing — described in [Block options](/en/builder/block-options). What follows is what is specific to each type. Several options, more than one selectable. **What the respondent sees.** A list of options they can tick, with an instruction telling them how many to pick. **Its options** * An option list — Add Bulk Options, Add Option, drag to reorder, per-option code, per-option visibility condition * **Selection mode**: `Unlimited` · `Exact number` (one numeric field: how many exactly) · `Range` (two numeric fields: minimum and maximum) * **Custom instruction text** — replaces the automatic instruction. The placeholder suggests `e.g. 'choose one or more', 'up to 2', 'at least 1'` * **Required Question** * **Allow 'Other' option** * **Randomize options** * **Exclusive options** — "Each selected option clears every other answer when chosen" This block has no Transport options. **In research.** The default block for awareness, usage and repertoire questions. Use Exclusive options for the "None of these" line so it can't be ticked alongside a real answer. Several options, exactly one selectable. **What the respondent sees.** A list of options where picking one clears any other. **Its options** * An option list, with the same controls as multiple choice * **Required Question** * **Allow 'Other' option** * **Randomize options** * **Transport options** — "Show only options/rows selected — or not selected — by the user in a previous question" Transport options only appears when an **earlier question with options exists**. On the first question of the interview the field is simply not there. **In research.** Screeners, single preferences, segment allocation. Also the block each task of a conjoint exercise is built on. A value chosen along a scale. **What the respondent sees.** A handle to drag between two labelled ends. **Its options** * **Slider Configuration** — Value: **Min**, **Max**, **Steps**; Label: **Minimum label**, **Maximum label** * **Required Question** There are no answer options and no randomization on this block. **In research.** Fine-grained intensity: purchase intent, price sensitivity, agreement with a statement where you want more resolution than a five-point scale gives. A series of images judged one at a time. **What the respondent sees.** One image after another, each answered with **LIKE** or **NOPE**. **Its options** * **Swipe configuration** — a list of images, each with **Select image** (`SVG, PNG, JPG, GIF`, max **10 MB**), **Image title**, **Image description**; the **New Image** button adds another * **Required Question** This block does not have the shared Media field: the images live in the swipe configuration. **In research.** The fast test on visual stimuli. Ideal for screening many concepts, packs or claims quickly, especially on mobile, where the effort per stimulus has to stay near zero. A free text answer. **What the respondent sees.** A text box, optionally with a placeholder and a length constraint. **Its options** * **Placeholder** * **Minimum characters** * **Maximum characters** * **Required Question** * **Enable AI follow-up** — "the AI will adjust the next question based on the user's previous answer". Turning it on reveals: * **Maximum follow-up questions**: `1` / `2` / `3` * **Goal** * **Tone/Style** **In research.** Verbatims, reasons behind a rating, spontaneous associations. With AI follow-up it becomes a short moderated exchange — see [AI moderation](/en/builder/ai-moderation). A binary answer with two renameable labels. **What the respondent sees.** Two buttons, "Yes" and "No" by default. **Its options** * **Response Configuration** — **First Response** (default "Yes") and **Second Response** (default "No"), both renameable * **Required Question** **In research.** Screeners and eligibility filters. Rename the two responses when the natural wording isn't yes/no — "I have" and "I haven't", "Agree" and "Disagree". A point on a scale, shown as stars or as numbers. **What the respondent sees.** A row of stars, or a row of numbers to click. **Its options** * **Rating Configuration** — **Number of Stars** from `1` to `10` * **Display type**: `Stars` / `Numeric` * **Required Question** **In research.** Satisfaction and liking on the star display. Set **Display type** to `Numeric` with a 10-point scale and you have an **NPS** question: that combination is how recommendation is measured in Exoid. A field that expects a valid email address. **What the respondent sees.** A single-line input for an address. **Its options** * **Email Configuration** — **Placeholder text** * **Validation pattern** — restricts which addresses are accepted, for example to one domain * **Required question** **In research.** Recontact for a follow-up wave, prize draws, panel building. The validation pattern is what keeps a B2B study to company addresses. A date, optionally with a time. **What the respondent sees.** A date picker, plus a time picker when time is enabled. **Its options** * **Date/Time Options** — **Placeholder** * **Enable time** — "Date only when off" * **Min** and **Max** — the range of dates accepted * **Required Question** **In research.** Date of last purchase, when an event happened, appointment slots. Use Min and Max to keep answers inside the period you're studying. An upload of one or more files. **What the respondent sees.** "Select your file", with the accepted formats stated: `.pdf, .doc, .docx, .jpg, .png (max. 10MB)`. **Its options** * **File Upload Configuration** — **Placeholder** * **Maximum number of files** * **Maximum file size (MB)** * **File types**, selectable from: PDF, Word Document, Word Document (DOCX), JPEG Image, PNG Image, GIF Image, Text File, CSV File, Excel Spreadsheet, ZIP Archive This block does not show Required Question. **In research.** Proof of purchase, shelf photos, in-home pack shots — anything where the evidence has to come from the respondent. An informational block with no answer field. **What the respondent sees.** A title, a description and optionally a media item, then the button to continue. **Its options** * Question code, Question, Description, Media (optional), and the three buttons at the bottom of the panel It asks for nothing, so it has neither Required Question nor Wait before advancing. Its question code carries the `STM_` prefix. **In research.** Section intros, instructions before an exercise, and the stimulus screen shown before the questions that ask about it. A grid: rows that all share the same columns. **What the respondent sees.** A table where each row is answered on the same scale. **Its options** * **Selection Mode**: `Single Choice` / `Multiple Choice` * **Rows** — Add Bulk Options, **Add Row** * **Columns** — "Add columns for the matrix (minimum 1)", Add Bulk Options, **Add Column** * **Required Question** * **Randomize Options** — "Randomize the order of the rows" * **Transport options** **In research.** With **Selection Mode** set to `Single Choice` this is the **battery of scales**: a list of attributes, brands or statements all rated on one scale, in one screen. It is the workhorse of image and attribute studies. Randomize Options shuffles the rows to cut order bias. A collapsed list of options. **What the respondent sees.** A closed field that opens into the list. **Its options** * An option list * **Required Question** * **Randomize options order** * **Transport options** **In research.** Long, flat lists where buttons would flood the screen: countries, regions, cities, product categories. Items put in order by the respondent. **What the respondent sees.** "Drag items to rank them from first to last". **Its options** * **Options to rank** — "Add the items that users will need to rank by preference or importance", with Add Bulk Options and Add Option * **Required Question** * **Randomize options** — "Show options in random order at the start" **In research.** The forced-choice exercise: the respondent can't call everything important, they have to trade off. Always randomize the starting order, otherwise the initial layout drives the result. A numeric answer. **What the respondent sees.** A field that only accepts numbers. **Its options** * **Placeholder text** * **Minimum value** * **Maximum value** * **Allow decimals** — "When enabled, respondents can enter decimal numbers" * **Required Question** **In research.** Age, household size, frequency, spend. A real number you can average and segment on, instead of a band you have to interpret. A choice made on images instead of text. **What the respondent sees.** A grid of images — four columns by default, two when Supersize is on. **Its options** * **Required Question** * **Image options** — "Add images; labels are optional. Respondents choose from a grid layout", each with Select image (`SVG, PNG, JPG, GIF`, max 10 MB); **Add option** adds another * **Selection mode**: `Unlimited` · `Exact number` (one numeric field) · `Range` (two numeric fields, minimum and maximum) * **Custom instruction text** * **Supersize** — "Display larger images in a 2-column grid, default is 4 columns" * **Allow 'Other' option** * **Randomize options** * **Exclusive options** **In research.** Pack tests, concept screening, creative selection. Set **Selection mode** to `Exact number` and it becomes a forced-choice task: "pick exactly 3" gets you a discrimination you can rank, where an unlimited choice would get you everything ticked. Turn on Supersize when the detail in the image matters. A field where the respondent types a website address. **What the respondent sees.** A single-line input for a URL. **Its options** * **Website Configuration** — "Configure the placeholder text shown in the website input field": **Placeholder** * **Required question** This block **collects** a site address from the respondent. It does not display a web page inside the interview. Earlier versions of this page said the opposite. **In research.** B2B profiling — the company site — or any study where the respondent has to point you at a page. ## The start screen Every interview has a start screen. It always exists as the `Start` node, so you never add it from the **+** menu, and it is the first thing a respondent sees. Its fields: * **Question** — the title of the screen * **Description** * **Skip welcome screen** — "If you skip the welcome screen, the survey will start immediately" * **Scripts** — see [Scripts](/en/advanced/scripts) Use it to say who is asking, roughly how long the interview takes and what happens to the answers. Turn on **Skip welcome screen** when the study is short and you'd rather open on the first question. ## Ending screens The **Ending Screen** closes the interview. Ending screens sit in their own section of the question list, their codes carry the `END_` prefix, and a campaign can have **more than one**. Its fields: * **Question code** * **Auto Redirect** — "Automatically redirect to the CTA link after the survey is completed" * **Title** * **Description** * **CTA Button Text** * **CTA Link** * **Screen Out** — "Mark this as a screen out to filter out disqualified respondents" * **Scripts** The default text is "Survey completed. Thank you for your time…", and "Powered by Exoid" appears at the bottom of the player. Multiple ending screens plus **Screen Out** is how you separate outcomes. Send people who complete the interview to a thank-you screen, and people who fail the screener to a second one marked as a screen out — so they are filtered out of the results instead of counted as completes. The Ending Screen has none of the shared block fields except Question code and Scripts. ## Create in bulk **Create in bulk** sits at the bottom of the same **+** menu, but it is not a question type: it creates several questions in one go instead of adding them to the canvas one at a time. In the Italian interface it's called **Aggiungi in serie**. Use it when the question list already exists on paper and you want it on the canvas in one move. Once created, each question behaves like any other block and keeps its own options. ## Next steps Configure question code, media, required answers, randomization and transported options. Turn an open-ended question into a conversation with automatic follow-ups. ## Related topics - [Block options](/en/builder/block-options.md) - [Payload reference](/en/webhooks/payload.md) - [Carry forward options](/en/logic/carry-forward.md)