What is a rating scale question?
A rating scale question gives people a fixed run of numbers and asks them to pick one. The scale is the same for everyone who answers, so the numbers can sit beside each other and mean the same thing.
It answers any question shaped like how much. How satisfied, how clear, how difficult, how likely. If the honest answer is a label you could write down as a list, multiple choice will serve you better. If the honest answer is an amount, this is the question type.
Use a rating scale when the answer is a number, not a name.
When to use a rating scale
Rating scales earn their place when you need each thing measured on its own terms, and you are content for several things to score equally well.
Use a rating scale to:
- Measure satisfaction with a feature, a flow, or a support experience
- Check how clear a message, a price, or an instruction reads
- Gauge how much effort a task took
- Ask how likely someone is to recommend the product
- Take the same reading again after a change, using the same scale
Skip it when the options need to compete. If only one of them can go first, a ranking question forces the trade-off a rating scale lets people avoid.
And a number on its own is only half the finding. Follow it with an open text question when you need the reason behind the score.
Choose the method that matches the decision
| Rating scale | Ranking question | Matrix | |
|---|---|---|---|
| The question it answers | How much? | What comes first? | How much, across several statements? |
| Best for | Measuring one thing on its own scale | Relative priority across a short list | Rating several statements on one shared scale |
| Response shape | One number, or a skip | One complete order | One chosen label per statement |
| Main trade-off | Several things can score the same, so nothing is forced to lose | Forces a difference, even where people feel none | Denser to answer, so keep the list of statements short |
Start with a rating scale when the score itself will change what you do next. Move to ranking when the decision is which one goes first, and to a matrix when you are asking the same question about several statements at once.
How to write a useful rating scale question
1. Name the decision the number will serve
Know what a 4 would change and what an 8 would change. A score with no decision behind it is a tidy figure that sits in a doc.
2. Ask about one thing at a time
A question with two subjects gets one number, and you will not know which half it belongs to.
Prefer: How clear was the pricing page?
Avoid: How clear and trustworthy was the pricing page?
3. Choose the scale length on purpose
Fiuto offers 5, 7, or 10 points. Five reads quickly and suits a light check. Seven picks up finer differences. Ten gives the most room, and is the right choice for how likely are you to recommend.
4. Label both ends in plain words
The end labels are optional, but a bare run of numbers leaves each respondent to invent their own meaning. Poor to excellent, or not at all to extremely, tells everyone which direction the scale runs.
Prefer: 1 is very difficult, 10 is very easy.
Avoid: Rate this from 1 to 10.
5. Follow it with an open question
The score tells you where people landed. A short text question after it tells you why, and that is usually the part you act on.
Worked example
Question:
How easy was it to set up your first project?
Scale:
7 points, low end labelled very difficult, high end labelled very easy.
Why it works:
- It asks about one moment, not the whole product
- The labels fix the direction, so a 6 means the same thing to everyone
- Seven points leave room for people who found it mostly fine
- An open follow-up can ask what got in the way
Rating scale examples
Satisfaction after a task
How satisfied were you with how that went? Five points, from very dissatisfied to very satisfied.
Likelihood to recommend
How likely are you to recommend us to a colleague? Ten points, zero-based, so the scale runs 0 to 10, the same scale NPS surveys use.
Clarity of a message
How clearly did that page explain what the product does? Seven points, from not at all clear to completely clear.
Perceived effort
How much effort did that take? Seven points, from very little to a great deal.
Each example asks about one thing, on one scale, with both ends named. Replace the wording with language your respondents already use.
How to read rating scale results
Fiuto shows you the average score and the spread of answers across the scale. Most people read the average first. The spread is usually the more interesting half.
Two questions can share an average of 6 and describe completely different situations. One where almost everyone answered 6, and one where half answered 3 and half answered 9. The first is a mild, shared verdict. The second is two audiences, and the average is describing neither of them.
So look at the shape before you quote the number. A tight cluster means people agree. A split means you are probably looking at more than one kind of respondent, and that is worth chasing with an open question.
Keep the finding tied to the wording, the scale, and the people who answered. A score is a reading taken with one instrument, not a fact about the world.
Run a rating scale question in Fiuto
Write your question and pick the scale: 5, 7, or 10 points. Turn on the zero-based option to start the scale at 0 instead of 1, which is how you get the 0 to 10 scale used for likelihood to recommend.
Add labels for the low and high ends if you want them, such as poor and excellent. Make the question required, or leave it optional so respondents can skip it.
Every answer comes back as a single number, or a skip. Fiuto reports the average and how the answers spread across the scale.
Share the study link with your own users, customers, waitlist, or community. Fiuto does not supply respondents, and the people you already have are usually the ones whose answer matters.
Frequently asked questions
01What is an example of a rating scale question?
“How easy was it to set up your first project?” answered on a 7 point scale from very difficult to very easy is a rating scale question. Every respondent picks one number on the same scale.
02Should I use a 5, 7, or 10 point scale?
Fiuto offers all three. Five reads quickly and suits a light check where you only need a rough direction. Seven picks up finer differences between people who broadly agree. Ten gives the most room and is the usual choice for how likely someone is to recommend something. Pick one and keep it, because scores taken on different scales are not comparable.
03Can I run an NPS style question in Fiuto?
Yes. Choose the 10 point scale and turn on the zero-based option, which makes the scale run 0 to 10, the eleven-point scale net promoter score surveys use. Fiuto does not have a feature called NPS. It reports the average score and how the answers spread across the scale, so if you want the NPS number itself you work it out from that spread.
04When should I use a rating scale instead of a ranking question?
Use a rating scale when you want each item measured on its own and it is fine for several items to score the same. Use a ranking question when you need relative priority and want people to make a trade-off. A rating scale can tell you five features are all valuable. A ranking question tells you which one to build first.
05Do I have to label the ends of the scale?
No, the low and high end labels are optional. They are worth adding anyway. Without them, each respondent decides for themselves what a 3 means, and you inherit that inconsistency in the results.
06Can respondents skip a rating scale question in Fiuto?
Yes, if you leave the question optional. Respondents then see a skip option instead of being held on the question. Make it required when the number is the reason the study exists.
07Does Fiuto recruit people to answer the rating scale question?
No. You bring your own users, customers, waitlist, community, or panel, then share the Fiuto study link with them.
