# Open-Ended Questions: Definition and Examples for Surveys

An open-ended question lets someone answer in their own words: “What influenced whether you contributed during the workshop?” Compare it with closed options.

- Canonical URL: https://www.harmate.com/en/blog/open-ended-questions-get-actionable-data-without-bias
- Author: Harmate Team
- Published: 2025-12-12
- Updated: 2026-11-02T00:45:05.603768+00:00
- Language: en

## Content

# Open-Ended Questions: Definition and Examples for Surveys

An open-ended question lets someone answer in their own words instead of choosing from preset options. It can surface context, exceptions, or a topic the questionnaire did not anticipate. It does not guarantee a complete answer, freedom from bias, or a representative set of responses. Its value depends on what you want to understand and how you review the answers.

## Open-ended or closed-ended: choose for the job

Use an open-ended question when you want to explore an experience, understand someone’s reasoning, or discover possible themes. It gives respondents more freedom, but answers may be brief, vary in detail, and take more work to analyse.

A closed-ended question offers options or a scale. It can make it easier to measure or compare predefined categories, but it cannot by itself explain why someone chose an option or what the listed choices leave out. When the topic is not yet well understood, start by collecting people’s own wording; then decide whether a closed-ended question would help.


### Closed-ended vs. open-ended questions: three paired examples

| What you want to learn | Closed-ended question | Open-ended question |
|---|---|---|
| Experience after training | “Have you used the method since the course? Yes / No / No opportunity yet.” | “Tell us about the most recent occasion when you could have used the method. What happened?” |
| Customer effort | “How easy was it to complete your request? Very difficult to very easy.” | “Walk us through what you did to complete your last request.” |
| Workshop participation | “Did you contribute during the workshop? Yes / No.” | “What influenced whether you contributed during the workshop?” |

These are different instruments, not interchangeable versions of the same measure. Use the open response to explore the experience; use a closed response when the categories serve a defined comparison. If the person has not had the experience, let them say so rather than invent an answer. If a closed-ended effort question feeds a score, this [worked Customer Effort Score example](/en/blog/how-to-calculate-customer-effort-score-formula-and-example) shows how to read its scale without confusing it with the open comment.

## Write a question for its context

Before drafting, decide what you want to learn, who will answer, and which situation the question refers to. A question focused on one subject and a recognisable context is often easier to understand and review. Treat this as a drafting cue to test, not a guarantee of comparability.

Avoid building the expected explanation, a preferred answer, or a list of examples into the question. Add only the context needed to make the subject clear. A time period or specific activity can help locate an answer, but a length limit or prescribed format will not suit every situation.

| Context | Open-ended example | Check before sending |
|---|---|---|
| Training | “Describe a recent situation where you tried to use this method at work.” | Have respondents had a chance to use it? If not, adapt the question. |
| Cross-functional project | “What information did you have when you picked up the client follow-up file?” | Does the question refer to a step and file everyone understands? |
| Activity feedback | “Describe one part of the activity that affected how you took part.” | Is the wording neutral, and is the experience recent enough to recall? |
| Customer experience | “What were you trying to accomplish during your most recent contact with our service?” | Can the respondent describe a different goal or an experience the question did not anticipate? |

These examples are starting points, not universal templates. Adapt the wording to the actual situation and check that the question does not ask for unnecessary personal information. For a sensitive topic, explain who will see the responses and how they will be used.


**Put one question to work.** Choose the decision the answers should inform and adapt an example. [Create an open questionnaire in Harmate](/en/product/open-questions), or, if you already have responses, import a CSV, JSON, or XLSX file through [Response Import](/en/product/response-import). Check the columns, then review themes alongside their source excerpts before deciding what to change. Import requires an account.

## Pretest before sending the questionnaire

Ask someone close to the intended audience to read the question and explain what they think it means, what they believe they should describe, and whether any term is unclear. If they see several possible interpretations, add context or rephrase. A pretest can reveal comprehension problems; it does not prove that a question is unbiased or that the answers represent the wider audience.

Also check whether the question contains two separate requests. If someone could answer “yes, but…”, split the topics or reword the prompt so that nuance can emerge. Do not require a long answer when a few words are enough for your purpose.

## Read answers without turning them into profiles

An open-ended response first tells you what someone chose to write in a particular context. It does not by itself establish a personality trait, general ability, or compatibility with other people. When grouping answers, keep the original wording, exceptions, and disagreements available for review.

For the difference between codes, themes, and interpretation, see the guide to [thematic analysis](/en/blog/thematic-analysis-codes-themes-and-interpretation). If you need to examine a larger corpus, the guide to [analysing 200 open responses](/en/blog/analyzing-200-open-responses-without-bias-a-method-for-actionable-decisions) covers a different task from writing questions. The [verbatim guide](/en/blog/verbatim-how-to-use-exact-quotes-without-distorting-their-meaning) explains how to quote while retaining context. AI assistance may help with some steps, but a summary does not replace checking source excerpts and counterexamples; see [AI survey analysis: don’t confuse a summary with a decision](/en/blog/ai-survey-analysis-dont-confuse-a-summary-with-a-decision).

Harmate’s [Open Questions workflow](/en/product/open-questions) lets teams review responses and the excerpts associated with themes. It does not make a question automatically valid or representative, or decide what the answers mean on the team’s behalf.

## Key takeaways

A useful open-ended question connects to a concrete goal, makes sense to the people answering it, and does not announce the conclusion you expect. Examples can help you start; pretesting, reading in context, and retaining the original wording help you check what the answers can support. To distinguish survey quality from response volume, see [how to interpret a survey response rate](/en/blog/survey-response-rate-why-this-number-can-mislead-you).