Templates Create a poll

How to run an anonymous feedback survey

People give more honest feedback when they trust it can’t be traced back to them. That trust depends on the settings you choose, the questions you ask and what you tell participants.

What “anonymous” means here

In VotePoll, participants don’t need an account, and no IP address is stored with a response. Repeat voting is limited with a random browser cookie, not by identifying people. Participants are only asked for a name if you turn that on.

Who can see what is set per poll:

  • Other participants never see who answered what, unless you turn on Show who voted for what.
  • You, the creator, see every response. You only see names next to answers when that setting is on.
  • Whoever runs this VotePoll site can see any name a participant entered, from the admin panel.

The full details are on the privacy page.

The settings to use

  • Don’t require a name, and leave Show who voted for what off.
  • Set result visibility to only you if participants shouldn’t see each other’s answers — usually right for staff feedback.
  • If you turn on comments, keep them private — the default: only you and the site administrator can read them.
  • Add a password if the link might be forwarded beyond the people you want to hear from.
  • Set a closing date, so people know how long they have.

Avoid questions that identify people anyway

A survey can be anonymous in name only. In a team of eight, a combination like “department + years of service + age group” can point to one person. Ask only for the background details you will actually use, and use broad groups.

Instead of “Which team are you in?” (8 teams of 3–6 people)

Better “Do you work mainly in the office, mainly remotely, or a mix?”

Small groups make answers identifiable. Ask for the background that matters for your decision, in groups large enough that nobody stands out.

Open text answers can identify people too — someone describing “the meeting on 3 March where I presented” has named themselves. Mention in the description that written answers are read as written, and don’t ask for personal details: VotePoll’s terms don’t allow personal or sensitive data in polls apart from a first name or nickname.

Word sensitive questions carefully

Ask about experiences and behaviour rather than asking people to judge colleagues, and give a clear way to say “no opinion”. Use a 0–10 rating for satisfaction and a text answer for “what would you change?”, so people can explain their score.

Instead of “Is your manager doing a bad job?”

Better “How satisfied are you with the support you get from your manager?” (0–10) and “What one thing would help most?” (text, optional)

The first version is loaded and invites a yes/no verdict. The second asks for an experience and leaves room for a constructive answer.

Tell people what happens next

Use the poll description to say who will read the answers, whether the overall results will be shared, and when. Then do it: sharing a short summary of what you heard — and what you will change — is the best way to get honest answers next time.

Templates to start from

The employee satisfaction, event feedback and anonymous suggestion box templates already use rating and optional text questions. Edit the wording to fit your group before you create the poll.