A poll maker for one good question and a fast answer
A poll is the only form people answer for their own reasons rather than yours, which is why it gets response rates a survey never will. The whole craft is in restraint: one question, options that do not overlap, no sign-in, and a result the voter gets to see. Everything below is about protecting that.
A poll is not a short survey
The difference is not length, it is commitment. A survey asks someone to spend effort on your behalf; a poll asks them to register an opinion they already hold. That is why a poll can sit in a Slack channel or a tweet and pull hundreds of answers while a three-question survey in the same place pulls nine.
The moment you add a second question you have converted it into a survey and you should expect survey response rates. That is a fine trade if the second question is worth more than the answers you lose — just make it deliberately rather than by drift.
It also means a poll is the wrong instrument for anything you need to be representative. Whoever is in the room answers. Treat the result as a read on the people you asked, not on your market, and you will not be embarrassed by it later.
Writing options people can actually pick
Almost every unusable poll result traces back to the option list rather than the question.
Make them mutually exclusive
If a voter could honestly pick two, your counts mean nothing. "Under £50 / £50–100 / £100+" works; "Cheap / Mid-range / Under £100" does not, because the third overlaps the first two and different people will resolve that differently.
Cover the whole space, or say you have not
A voter with no matching option either abandons or picks the nearest wrong one, and the second is worse because it looks like data. Add "None of these" or "Something else" when the list is not exhaustive, and expect it to be picked more than you would like.
Keep the options the same shape
Three short nouns and one long explanatory sentence is not a fair contest — length reads as importance. Make every option roughly the same length and grammatical form, or you are voting on your own formatting.
Randomize nothing you need to compare
Order effects are real: earlier options get picked more. If you are running the same poll repeatedly to watch a trend, keep the order fixed so the bias is at least constant across runs.
Before you share the link
- One question. If there is a second, decide consciously that it is worth the drop-off.
- Options are mutually exclusive, similar in length, and cover the space or offer an escape hatch.
- No account required — a poll behind a login is a survey with extra steps.
- reCAPTCHA on if the link will be public. It is free and invisible to real voters.
- Anonymous responses on if you have implied the vote is private.
- You know what you will do with each possible outcome. If two results lead to the same decision, do not run the poll.
Ballot stuffing, and what you can honestly do about it
Any poll on a public link can be voted in twice by someone who wants to. There is no setting in any tool that fully prevents this without identifying voters, and a tool that claims otherwise is usually doing browser-side tricks a determined person clears in a second.
What is worth doing: turn on reCAPTCHA, which is free here and stops automated submission, and be honest with yourself about the stakes. For "which day suits the team" the risk is nil. For anything with a prize or a public outcome attached, a poll on an open link is the wrong instrument and you want identified voting instead.
The honest version of anonymity is worth getting right in the other direction too. If you have told voters the poll is anonymous, Zunoform's anonymous responses setting means no IP address, user agent, referrer, geolocation, browser language or screen size is stored, and those identifiers are stripped from partial rows as well. Note the tension with the paragraph above: the same setting that keeps your promise also removes the metadata you would have used to spot duplicate voting. You cannot have both, and you should pick on purpose.
Poll and voting templates
Three real templates you can open, see the questions in, and publish.
What a poll costs here
Nothing. Polls are on the free plan — unlimited forms, 500 responses a month, no card, no account for voters, with reCAPTCHA and anonymous responses both included. You would only pay if you wanted Zunoform's branding off the page, which is $14.99 a month.
Questions people ask
Do voters need an account to answer a poll?
No. Every Zunoform form is a public link — respondents never sign in and never create anything. That is the single biggest factor in poll response rate, so any tool that requires a login is solving a different problem.
How do I stop people voting twice?
You can raise the effort but not eliminate it. reCAPTCHA is free and blocks automated submissions, which is the common case. Blocking a determined human on a public link requires identifying voters, which means the poll is no longer anonymous. For low-stakes questions this does not matter; for anything with a prize attached, use identified voting instead of a public poll.
Can I see results as they come in?
Yes — responses appear in the dashboard as they arrive, with per-question summaries. If you want the numbers somewhere else, the free plan connects Google Sheets so every vote appends as a row live, which is usually the easiest way to put a running total in front of other people.
Poll or survey?
Poll if you want one number fast from whoever is around, and you already know what you will do with each outcome. Survey if you need to understand why, compare groups, or defend the result to someone. The dividing line is the second question — once you add it, expect survey response rates.
Keep reading
Better forms.
Better data.
Build your first form in under 60 seconds. Free forever for personal use, no credit card required.