“Overall, how easy or difficult was that to complete?”
From Jeff Sauro · Founding principal of MeasuringU; popularized the Single Ease Question (SEQ)
Why it works
The SEQ is a single item — 'Overall, how difficult or easy was the task to complete?' — on a 7-point scale labeled only at the ends. Its power is that one well-placed question correlates with task completion and task time about as strongly as much longer, more elaborate difficulty questionnaires; the extra items mostly add length, not signal. Asking it immediately after the task is what makes it work: perceived ease decays fast, and a rating taken right after the attempt reflects the actual struggle rather than a tidied-up memory. Because every respondent answers the identical item the same way, the scores are comparable across tasks, releases, and competitors — which is the whole point of a standardized measure.
When to ask
Immediately after a user attempts a single, discrete task — completing checkout, configuring a setting, finishing a flow. The SEQ is a post-task measure: ask it while the experience of doing the thing is still in working memory, before the user rationalizes it.
Good follow-ups
- What made it feel that way?
- Where exactly did you get stuck or slowed down?
- What would have made it a 7 — completely easy?
Watch out for
Reading one person's score as a verdict. A single SEQ rating is a conversation-starter, not a measurement; the mean and its benchmark (around 5.5 out of 7 across typical tasks) only emerge at sample size. The other trap is letting the rating float free of a specific task — asked about 'the product' in general, the SEQ loses the anchor that makes it meaningful. Tie it to one concrete task, gather enough responses to trust the average, and always pair it with 'why?' so a low score tells you where the friction is, not just that it exists.
Where to ask
- In-product surveygreat
The native habitat — fire the single ease item the instant a task finishes, while the difficulty is still felt. The SEQ was designed to run immediately post-task, exactly here.
- Long-form surveygreat
Travels cleanly in an emailed post-task survey; the one 7-point item stands on its own and the mean only means something at sample size, which a survey delivers.
- User interviewworkable
Excellent in a moderated usability test, where it was born — but treat one rating as a prompt for the 'why', not a measurement. The SEQ benchmark (~5.5 on a 7-point scale) only exists at sample size.
Reworded for this context: “Overall, how easy or difficult was that to complete — and why?”
- Support chatpoor
SEQ rates the ease of a task the user just performed in the product, not a support exchange. Asking it mid-conversation has no task to anchor to, so the number measures nothing comparable.
- Sales callpoor
A prospect who hasn't completed the task has no ease to rate. The question lands as hypothetical and the answer carries no usability signal.
Pairs well with
Stage: Get value · A question popularized by Jeff Sauro
Source: https://measuringu.com/seq10/