Questions it can answer
- Which style direction should we go in?
- What tone of voice is most appropriate?
- Which design variation should we go with?
What is a preference test survey?
A preference test survey presents variations of the subject, asking questions around the variations and ultimately having the participant choose a preference.
Related guide: This page is the template. For when to use preference testing, how many participants you need and how it compares with A/B testing, read our preference testing guide.
Do it faster with a Claude skill: Unmoderated test builder. Builds a first draft of this study in Claude that you can edit before you send it.
The questions in this template
These are the questions in the Great Question preference test template. Replace the text in angle brackets with your own.
How often do you use <insert product or service>?
Single Select
- Daily
- Weekly
- Once a month
- Every 2-3 months
- 2-3 times a year
- I do not use it
Which of these <insert product or service category> have you used before?
Multi Select
- Brand A
- Brand B
- Brand C
- None of these
Which option do you trust more?
Single Select
- Option 1
- Option 2
- Option 3
- I don't trust any of them
Which option feels more secure?
Single Select
- Option 1
- Option 2
- Option 3
- None of them
Which option feels more informal?
Single Select
- Option 1
- Option 2
- Option 3
- None of them
Which option gives you the most confidence?
Single Select
- Option 1
- Option 2
- Option 3
- None of them
Which option would you prefer to buy from?
Single Select
- Option 1
- Option 2
- Option 3
- None of them
Which option provides the most important information?
Single Select
- Option 1
- Option 2
- Option 3
- None of them
Why did you choose that option?
Long Text
What did you not like about the alternative option?
Long Text
How would you make your chosen option even better?
Long Text
Advantages
- Brings user opinions into early stage decision making
- Helps set direction in a way that’s user focused and not stakeholder focused
- Simple to setup and analyze results
Limitations
- Does not reflect real world usage because participants aren’t using the product for real
- Focuses exclusively on the opinions of participants and not behaviors
- The variations are often presented out of context as it’s not possible to preference test an entire product
Frequently asked questions
What questions should a preference test ask?
Which option people trust more, which feels more secure, which gives them the most confidence, which they would rather buy from, and which provides the most important information. Then ask why they chose it, what they didn't like about the alternative and how they would make it better. The "why" is the most useful answer.
How many versions should I compare?
Two or three. The template has room for three options. More than that and people struggle to hold the differences in mind.
Should I randomise the order of the designs?
Yes. People lean towards the first or last option they see, so randomise the order for each participant.