How to Create a Free Feedback Survey: A Simple Step-by-Step Guide

Step 1: Start with one decision

Most weak surveys fail before the first question is written. They try to learn everything, so they ask about everything. Start instead with a single decision the answers will change: whether to keep an exercise, shorten a section, move a session earlier in the day, or run the workshop again.

Write that decision as a sentence and keep it next to you. Every question you add later has to earn its place by helping with that decision. If a question would not change anything you do, cut it.

Step 2: Pick your audience and your moment

Decide who should answer and when. For a workshop, that is the people who attended, and the best moment is usually the last few minutes of the session or the same day, while details are fresh. For a product or a service, it is people who have just used it.

Be honest about how many people you can reach. Eight responses can tell you about that group, but not reliably represent a wider audience. Report counts such as 6 of 8 and avoid broad claims about all learners.

Step 3: Write five to seven focused questions

For a short session, five to seven questions is a useful starting point. Treat ten as an editing guideline, not a technical limit or a guarantee of completion. A practical shape looks like this:

• One overall rating first, such as: How would you rate this session overall, from 1 (poor) to 5 (excellent)?
• Two or three questions about specific parts, such as: How useful was the group case study for your own work?
• One question about delivery or logistics, such as: How well did the pace suit you?
• One open question near the end: What is one thing you would change?
• An optional follow-up question for anyone who wants a reply.

Pew Research Center's guidance on writing survey questions makes the same core points: keep wording neutral and ask about one thing at a time, so you know what each answer means.

Step 4: Remove the wording traps

Read each question aloud and check it against three traps. First, double-barrelled questions: Was the content clear and engaging? is two questions, and a person who found it clear but dull cannot answer. Split it.

Second, leading questions: How much did you enjoy our excellent speaker? tells people what answer is expected. Ask How would you rate the speaker? instead. Third, vague words like good or often. Label the ends of your scale, for example 1 means not useful and 5 means very useful, and use the same scale for every rating question.

Step 5: Build it without paying

You can build a survey in a spreadsheet-linked form tool, a template site, or in debrief. In debrief, you can start without an account: add your slides, notes or a short description of what you covered, review the generated questions, and edit them. An account is only requested when you press Continue after reviewing the questions.

The free plan includes 5 surveys and 150 responses in total per UTC calendar month, with no card required. Participants never need to sign in. Material you provide is deleted once your questions are ready.

If you would rather start from a standard set, the free feedback templates page lists editable question sets by session type.

Step 6: Test it on yourself and one other person

Open the survey on your phone and time yourself. If it takes more than two or three minutes, cut questions. Ask one colleague to complete it and tell you whether any question made them hesitate. Hesitation usually means a question is unclear or asks for two things.

Step 7: Share one link and ask at the right time

Share a single link where people already are: on the last slide, in the follow-up email, or in the chat at the end of a session. Say how long it takes and that answers are anonymous, and then keep that promise by not asking for names.

A short reminder the next day is usually enough. Repeated nudges tend to annoy the people who already answered.

Step 8: Read the results as counts

Look at the overall rating first, then the specific parts, then the open comments. Report counts, such as 5 of 9 rated the case study 4 or higher, instead of adjectives like most. With small groups, avoid reading too much into one answer; look for patterns that repeat.

Finally, write down the one change you will make, and mention it to the next group. People answer more carefully when they can see that feedback leads somewhere.