A quick preference test for your designs

Upload 2 to 16 mockups, covers or layouts and share one link. Each person picks the better of two until their top list is settled. You get each voter's list and the overall ranking, for free.

Voters don't sign up. Files are deleted after 30 days.

Which page would you keep reading?

Pick 1 of 4

Sample pictures. Nothing you pick is saved.

When this helps

You have several directions for a landing page, a book cover, packaging or a slide template, and you need to pick one before you build it out. A preference test asks people which version they like better. Pairpoll runs that test with the people you choose: your users, your client’s team or a few people from your audience.

How a poll for designs works

  1. Export each design as an image at the same size. For a page, a screenshot of the top of the page is usually enough.
  2. Create a poll and write the question for the job the design has to do, for example “Which page makes you want to read on?” or “Which cover would you pick up?”
  3. Choose a top 1, 3 or 5. A top 3 shows the runner-up directions as well as the winner.
  4. Send the link. Voters see two designs at a time, side by side, and pick one until their top list is settled.
  5. Open the results and look at both the overall ranking and each voter’s list. If most people put the same design first, you have a clear answer.

Pairpoll and UX preference test tools

User research tools run preference tests too. Lyssna’s compares up to 6 designs, and Lyssna can also recruit participants for you from its paid panel. Pairpoll does one thing: people compare designs two at a time, up to 16 designs in a poll, and each voter ends with their own top list.

UXtweak’s preference test also compares up to 6 designs, or up to 2 images on its free plan.

What it doesn’t do

Pairpoll doesn’t collect comments, so it tells you which design people prefer but not why. If you need the reasons, talk to two or three voters afterwards or use a survey tool. It also doesn’t test clickable prototypes. You can record a short screen video of each version and run a video poll instead.

Tips

  • Change as little as possible between versions. If the copy, the colours and the layout all differ, you won’t know which change people liked.
  • Use real content. Placeholder text makes every layout look alike.
  • Ask the people the design is for. Designers and clients judge differently from users.

Common questions

What is a preference test?

A test where people see two or more versions of a design and say which they prefer. It’s quick and it measures opinion, not behaviour. To see which version works better in use, test it with real traffic or watch people use it.

Can my client vote without an account?

Yes. Voters open the link, type a name and start. Only the person who creates the poll signs in.

Is my work kept private?

Poll links are unlisted and hidden from search engines, so only people with the link can open the poll. Files, votes and the poll itself are deleted 30 days after the poll is created.

Can I share the result with the client?

Yes. You can export the results as CSV or JSON, or turn on a read-only results link. Voter names stay hidden on that link unless you choose to show them.

Other things to rank

Make your own poll

Free. Upload 2 to 16 images or videos and share one link. Voters don't sign up.

Create a poll

Pairpoll uses Google Analytics cookies to see which pages people visit, only if you accept. Details are on the privacy page.