Start by asking how many options you're testing. If you're comparing two or more options, use Head-to-Head, Ranked, Single Select, or Multi Select. If you're getting feedback on a single item, use Open-Ended, Star Rating, Emoji Reaction, Click Test, 5-Second Test, or Screen Recording.

The first decision is whether you're comparing multiple options or evaluating a single one. This determines which poll types are available to you.

If you have two or more options to compare, see the multi-option poll types below. If you're collecting feedback on a single item, skip to the single-option poll types.

Best for: Comparing exactly two options when you need a fast, definitive answer.

How it works: Respondents choose between Option A or B and explain their decision.

Try it when: Testing two thumbnails, evaluating two elevator pitches, or validating two pricing statements side by side.

Example question: "Which app icon would you tap first?"

Best for: Ordering 3–8 options when you need to understand preferences across the full list.

How it works: Respondents drag items into rank order, then justify their top and bottom picks. Instant-runoff voting produces a definitive winner.

Try it when: Prioritizing feature roadmaps, arranging product image stacks, or sequencing onboarding flows.

Example question: "Rank these hero images for a new kitchen scale listing."

For a detailed walkthrough, see Testing more than two options.

Best for: Mutually exclusive factual questions where only one answer can be correct.

How it works: Respondents choose a single option and explain why they selected it.

Important: Only use Single Select for factual or binary answers — like age brackets, yes/no, or subscription status. If you use it for creative preference testing with many options, votes may split evenly (e.g., 12.5% each across 8 options) and give you no clear winner. When you want a definitive winner among creative options, use Ranked instead.

Example question: "How many pets live in your household?"

Learn more in How to use Single Select polls.

Best for: Discovering which options resonate together or have broad appeal.

How it works: Respondents select all options they like and describe their selections.

Try it when: Doing feature prioritization, bundle planning, or gauging which ad angles resonate simultaneously.

Example question: "Which feature upgrades would you prioritize for our analytics dashboard?"

Learn more in How to use Multi Select polls.

Best for: Collecting detailed written feedback on a concept, question, or asset.

How it works: Respondents write a free-text response. There are no options to choose from — you're simply asking for their thoughts, reactions, or answers to your question.

Try it when: You want qualitative feedback on a concept, need to validate an idea before committing to designs, or want to understand how people describe or react to something in their own words.

Best for: Capturing a 1–5 numerical benchmark for a single item.

How it works: Respondents rate the item on a 1–5 star scale and explain their score.

Try it when: Assessing clarity, trustworthiness, or satisfaction for a single version of an asset — like a product description, landing page, or onboarding checklist.

Example question: "How would you rate this new onboarding checklist?"

Try it: Open sample Star Rating poll builder

Best for: Getting a quick read on emotional or gut response.

How it works: Respondents react with emojis (from negative to love) and explain what feelings the asset triggers.

Try it when: Testing mood boards, storyboards, ad concepts, or early creative exploration where sentiment is the signal.

Example question: "What is your gut reaction to this launch teaser script?"

Try it: Open sample Emoji Reaction poll builder

Best for: Understanding where people look or click on an image.

How it works: Respondents click on specific areas of a single image. Results show a heatmap of where attention focused, along with written explanations for each click.

Try it when: Testing layouts, packaging shelf placement, ad designs, or any visual where you need to know what draws the eye first.

Learn more in What is a Click Test?

Best for: First-impression testing — what do people notice and remember after 5 seconds?

How it works: Respondents view an image for exactly 5 seconds, then answer questions about what they recall. It measures instant comprehension and visual hierarchy.

Try it when: Testing whether your key message, branding, or call-to-action is immediately clear on a landing page, ad, or product image.

Best for: Watching how respondents interact with a page or prototype.

How it works: Respondents record their screen while completing a task you define. You get video recordings of their actual behavior, plus their spoken or written commentary.

Try it when: Evaluating website usability, checkout flows, app navigation, or any experience where seeing the full user journey matters.

You can test up to 8 options per Ranked, Single Select, Multi Select, Star Rating, or Emoji Reaction poll. Duplicate your poll if you need to explore more variations.

When you have more than two options, weigh what insight you need and how much effort respondents should spend.

Use Single Select only for mutually exclusive answers (age, household size, subscription status). Use Ranked when several creative or concept options could win and you need a majority-backed champion. Use Multi Select when respondents may like more than one option and you want to see the full set of appealing choices.

Yes. Duplicate your first poll, switch the poll type during setup, and upload the same assets. Many teams run Multi Select to spot appealing themes, Ranked to name the winner, and Single Select only when they later need a mutually exclusive factual response from a filtered audience.

Yes. When you build a survey with multiple questions, you can use a different poll type for each question — up to 16 questions per survey. For more on surveys, see What is a survey and when should I use it?