UEQ+: the modular questionnaire for user experience

What is the UEQ+?

The UEQ+ is a modular questionnaire that measures user experience not as a single number, but as a profile of individual qualities. Martin Schrepp and Jörg Thomaschewski published it in 2019 as a further development of the User Experience Questionnaire (UEQ).

Instead of a fixed questionnaire, there is a collection of validated scales. You choose the ones that are relevant for your product. Each scale consists of four contrasting word pairs — such as “slow / fast” or “confusing / clear” — that are rated on a seven-point scale.

When does the UEQ+ fit — and when not?

The UEQ+ fits when you need to know which quality is the problem. Typical occasions:

• one system replaces another, and the question is: better or worse — and in what way?
• a relaunch is to be measured before and after on the same scales
• two products or competitors are compared directly
• a poor SUS score needs to be narrowed down to one quality

The UEQ+ does not fit when you only need a fast overall score — then the SUS is cheaper. And it does not answer questions about motives or causes; for that you need open questions or a JTBD study.

Which scales are there?

A collection of validated scales from which you choose to suit your product. Commonly used:

For each scale, you can also ask how important it is to the respondents. This shows whether a weak score concerns a quality that matters at all.

The selection must be justified. Which qualities are decisive for your product? POS software needs efficiency and perspicuity; an online shop needs trust and attractiveness more.

How is the UEQ+ analyzed?

For each scale, the four contrasting word pairs are averaged into a score from −3 to +3. Negative scores stand for a poor rating, positive ones for a good rating; scores around zero for a neutral judgment.

The UEQ+ becomes meaningful above all in comparison:

• Against a benchmark: places the scale score relative to other measured products.
• Against an earlier measurement: the same scales before and after a redesign show the difference per quality.
• Between segments: an unremarkable average can hide two groups with opposite experiences.

With small samples, every scale score needs a confidence interval. If the intervals of two measurements overlap heavily, a difference is not proven.

Example: four scales better, two clearly worse

In the POS software case study (in German), a retail chain is about to roll out new POS software across 168 stores. Feedback from the pilot is contradictory. 289 employees rate six scales — for the old and the new system:

“The system is bad” is not true — “the system has two specific weaknesses” is. The segment comparison also shows that long-serving employees rate intuitive use considerably worse than new ones. Canceling the rollout is rejected; instead, two workflows are corrected and the rollout is postponed by four weeks.

What are the limits of the UEQ+?

• Longer questionnaire. Every scale costs four ratings. More than a handful of scales overwhelms participants.
• More participants. Because each scale is analyzed separately, the UEQ+ needs considerably more responses than the SUS.
• The selection is a decision. If you choose the wrong scales, you measure precisely past the question.
• No cause. The UEQ+ shows which quality is weak, not which screen or workflow is behind it. For that you need open questions.

How does the UEQ+ work in usabiq?

UEQ+ is included from Research Core.

• Up to 15 scales and three presets as a starting point for your selection.
• Grades from A to F per scale, plus confidence intervals — scored without a language model.
• Transparent intermediate values. All scale means can be viewed and exported individually, not just the grade.
• Additional open questions are analyzed separately; AI theme detection runs only on JTBD Switch, JTBD Story and the free-form questionnaire.
• Compare segments from Decision Pro — with statistical confidence instead of colorful bars. If a group is too small to support a conclusion, usabiq says so.