There are actually 20+ question types available to a survey designer. Defaulting everything to "let's just use multi-choice" or "let's just use open text" drops the quality of the data and makes analysis harder.
This article walks through each type and when to use which.
Categorizing question types
| Category | Types |
|---|---|
| Selection (quantitative) | Single-select, multi-select, dropdown |
| Rating (quantitative) | 5-point, 7-point, star rating, NPS |
| Open (qualitative) | Short text, long text |
| Numeric / formatted | Number, date, email, phone, URL |
| Structured | Matrix, ranking, slider |
| Other | File upload, signature, address, appointment time |
Selection types
Single-select (radio buttons)
The most-used type. Choose one.
Q. How did you hear about us?
○ Search engine
○ Social media
○ Referral
○ Advertising
○ Other
Tip: Make options MECE (no gaps, no overlap). Always include "Other."
Multi-select (checkboxes)
When multiple items should be selectable.
Q. Which features are you actively using? (select all)
□ Dashboard
□ Reports
□ API integration
□ Team features
□ Mobile app
Tip: Avoid making "Select all" attractive (put "Other" at the end).
Dropdown
For 10+ options (industry, country, role).
Caveat: Awkward on mobile. For ≤5 options, prefer single-select.
Rating types
5-point (Likert)
The standard rating scale.
Q. How satisfied are you with our service?
1 (Dissatisfied) ─ 2 ─ 3 ─ 4 ─ 5 (Satisfied)
Tip: Watch for central bias. If you really need to see directional differences, consider an even-point scale.
7-point
Common in academic research. Right for fine-grained analysis.
Star rating
Widely used in e-commerce. Intuitive but less academically rigorous.
NPS (11-point)
Asks recommendation likelihood on 0–10. The standard loyalty metric.
Q. How likely are you to recommend us to a friend or colleague?
0 (Not at all likely) ─ ... ─ 10 (Extremely likely)
Open types
Short text
Single-line answer. Names, short comments.
Q. Your name
[__________________]
Long text (textarea)
Multi-line free response. Improvement requests, open feedback.
Q. What would you like us to improve?
[ ]
[ ]
[ ]
Tip: Always make open text optional. Required open text crushes completion rate.
Numeric / formatted
Number
Numeric input only. Age, count, amount.
Q. Annual budget (USD, thousands)
[_________]
Tip: State the unit. You can also add range validation.
Date
Calendar picker.
Q. Start date
[YYYY-MM-DD]
Email / phone / URL
Each comes with format validation. On mobile, the appropriate keyboard (email, numeric) appears automatically.
Structured types
Matrix
Rate multiple items on the same scale in one block.
Please rate the following on satisfaction:
Diss. Some. Neut. Some. Sat.
Price ○ ○ ○ ○ ○
Quality ○ ○ ○ ○ ○
Support ○ ○ ○ ○ ○
Delivery ○ ○ ○ ○ ○
Tip: ≤5 rows. More than that breaks on mobile.
Ranking
Rank a list of options.
Rank these by importance:
1. Price
2. Quality
3. Support
4. Features
Tip: 4–5 items max for usability.
Slider
Visual numeric range input.
Q. Estimated budget ($k): [slider: 0 ─ ● ─ 100]
Other types
File upload
Customer uploads images, PDFs, video.
Use cases: Job applications (resume), quote requests (specs), support (screenshots).
Signature
Touch-input signature.
Use cases: Consent, contract substitute.
Address
ZIP code → auto-complete address.
Appointment time (calendar integration)
Pick from available time slots.
Use cases: Sales call booking, interview scheduling, support appointments.
Decision flowchart
Want analyzable quantitative data?
├─ Yes → Selection or rating
│ ├─ 10+ options → dropdown
│ ├─ Pick one → single-select
│ ├─ Pick multiple → multi-select
│ └─ Measure degree → rating
│ ├─ Standard → 5-point
│ ├─ Fine-grained → 7-point
│ └─ Recommendation → NPS
└─ No → Open or formatted
├─ Free comment → long text
├─ Short info → short text
├─ Number → number
├─ Date / time → date or appointment
└─ Email / phone / URL → formatted
Repoan's question types
Repoan supports all of the above types.
- AI chat picks an appropriate type for each question automatically
- You can change the question type later while preserving data compatibility
- All types render correctly on public forms, embedded forms, and mobile
You get the right type for the use case without fiddling with settings.
Things not to do
❌ Make everything open text
You lose quantitative analysis and pay for it later.
❌ Make everything single-select
"None of the above" becomes the most-selected option and the data distorts.
❌ Over-use required questions
5 required questions = a sharp completion drop. Keep required to the minimum, mark everything else optional.
❌ Pile 10 rows into a matrix
Breaks on mobile. Stick to ≤5 rows.
Summary
Three rules for picking a question type:
- Aggregability comes first (quant vs. qual)
- Mobile UX matters (matrices used sparingly)
- Minimal required to prevent drop-off
Repoan's AI chat picks an appropriate type for each question by default. You can override manually after generation, so the smoothest workflow is to let AI build the first draft and then adjust.