Anonymous voting means a vote counts without the voter saying who they are. On a product roadmap it means something narrower: feature voting without a login, where a visitor votes without signing up and the board still recognises their browser so one person cannot vote twice. The question worth asking is what turning it on costs.
Which tools allow it is in the table below. The harder question is what you agree to when you turn it on, and that is a trade with something real on both sides.
Which roadmap tools allow anonymous voting?
Twelve of the nineteen tools whose own pages settle it, read on 3 October 2026. Three allow a vote without an account by default, seven as a setting the board owner turns on, two only on a paid plan, and seven never do.
| Tool | Vote without an account | What its own pages say | Source |
|---|---|---|---|
| Features.Vote | Yes, by default | Voting needs no signup | home |
| Feature Upvote | Yes, by default | Voting needs no account; a new suggestion or a comment asks for an email | page |
| VoteFirst | Yes, by default | A random cookie identifies each visitor, and a setting can require sign in | below |
| Featurebase | A setting | Posting, voting and commenting without a login are three separate switches; on 3 October 2026 voting was on in 1,134 of 1,218 live workspaces and posting in 113 | help |
| Frill | A setting | Its Anonymous mode tracks a visitor by cookie; the default Guest mode asks for a name and email. 69 of 117 live workspaces had votes on Anonymous | help |
| FeatureOS | A setting | Guest votes are a switch, on for 29 of 79 live workspaces | help |
| Sleekplan | A setting | The board asks for an email unless the anonymous option is on | help |
| Convas | A setting | A switch on each board | post |
| Noora | A setting | A switch on each board lets a visitor post, vote and comment with no account, under a random name, and its pricing page lists it on every plan | changelog |
| FeedBear | A setting | Its features page offers sign in through single sign on or a social account, or anonymous voting turned on | features |
| Supahub | From Starter, $15 a month billed yearly | Guest posting, commenting and voting is left off the free plan | pricing |
| UserJot | From Starter, $29 a month | Guest posting, which covers posting and voting without an account, is left off the free plan | pricing |
| Canny | No | Anonymize users hides a name behind an alias; the voter still has an account and the team still sees who it is | help |
| Aha! Ideas | No | An anonymous ideas portal shows Guest in public; the requester still has an account | post |
| Fider | No | Voting needs an email sign in, and anonymous voting is an open request on its own board | request |
| UserVoice | No | Voting needs an account, and anonymous suggestions are an open request on its own board | request |
| Jira Product Discovery | No | Voting needs a contributor account, which is free | answer |
| Rapidr | No | Its help page answered no and pointed to single sign on that assigns each visitor a made up identity; the help centre was offline on 3 October 2026 | archived help |
| Savio | No | A vote comes through a Savio account or single sign on from your own app | help |
Three more could not be settled from their own pages. Nolt and Upvoty name an anonymous setting only in posts on their own feedback boards, not in a help page that could be read. ProductLift's marketing page says voting without an account ships on every plan, while a request for it is still open on ProductLift's own board.
Two rows carry the most common confusion. On Canny and Aha! Ideas, anonymous describes what the public sees, a hidden name, and not how the vote was cast. Everyone voting there still has an account.
Where the switch exists, boards use it. Of 1,218 live Featurebase workspaces read on 3 October 2026, 93.1 percent let a visitor vote without signing in, and 9.3 percent let one post. On Canny, 104 of the 868 boards on paying sites show voters under aliases. On Frill, where a new account asks for a name and email, 69 of 117 workspaces take a vote with nothing asked. The settings on all 3,275 boards are in one open file.
What does anonymous feature voting mean here?
That a visitor can press vote with nothing asked of them, and that the board still knows they are the same person next time.
An identity is minted in their browser the first time they land. It holds sixteen random bytes and nothing about them, it lasts a year, and everything they do is attributed to it. That is enough to stop one person voting twice for the same thing, and it is not enough to write to them.
If the cookie is missing, because it was blocked or cleared, the fallback is their network address. Everybody behind one office connection then looks like one person. It is the rare case rather than the normal one, and it is worth knowing if the people you are collecting from all sit in the same building.
Is anonymous voting the same as a secret or blind vote?
Not quite, and the general definitions disagree with each other. Loomio treats blind and anonymous voting as one mode: a vote that cannot be connected to the person who cast it, with the results hidden until the poll closes. eBallot defines an anonymous vote as one that cannot be traced to its voter, even by the organisation running it, and also sells weaker versions where an administrator can see the pattern. A secret ballot, in the dictionary sense, is a known and registered voter marking a choice in private.
On a feature board it usually means something narrower: a vote cast without an account. The board still ties the vote to one browser so the same person cannot vote twice, which makes it closer to a pseudonym than to a secret ballot, and the counts are shown as they come in, so nothing about it is blind.
If that is the job you have, a trial is 14 days with a card at signup and $9 a month after it.
What do you get for it?
Volume, and a different kind of volume than you expect.
Every step between having an opinion and recording it loses people, and the losses are not random. A sign up form does not filter out the frivolous, it filters out the busy. What survives a form is the unusually motivated, which is a real population and not the one you were trying to measure.
Turning the barrier off does not just give you more votes. It gives you votes from people who would never have made an account and were never going to write in, which is the group your roadmap is least likely to be hearing from otherwise.
What do you give up?
The ability to write back, which is a bigger loss than it sounds.
When a release ships, VoteFirst writes to the people whose vote is in it and names what they asked for. That message can only go to somebody who has an account with an address on it.
An anonymous voter is skipped, not by choice but because there is nowhere to send it. Frill and Featurebase say the same of their own anonymous voters. So the loop closes for some of your voters and not for others, and the ones it does not close for are exactly the ones the anonymous option brought in.
The public answer still reaches them if they come back. The card moves, the release appears on a page anybody can read. What they do not get is the arrival, and arriving is most of what a notification is for.
The demo board is open and it asks for no account.
Does anonymous voting work with revenue weighting?
Mostly not.
A weighted vote carries the multiplier of the tier its voter sits in. An anonymous visitor sits in no tier, because a tier is something you assign to somebody you can identify. So their vote weighs one, which is the correct and deliberate floor, and no amount of weighting will move it.
That means the two ideas pull against each other in the same product. Anonymous voting widens the sample and revenue weighting sharpens it, and a vote can be inside one or the other but rarely both. Worth knowing before you build a plan on having both at once.
The way through is to let identity be optional and late rather than required and early. A customer who signs in through your own product votes as themselves and carries their tier. Everybody else votes anonymously and counts as one. You get the width and the sharpness in the same board, on different rows.
Does it invite abuse?
It invites less than you would think, and the defences are not the ones people assume.
There is no spam filter in the product. What exists instead is a review gate.
Every suggestion from a visitor is held until you approve it, whichever route it arrived through, and there is no setting that turns that off. Nothing embarrassing appears on your board without a person seeing it first.
Votes are different, because a vote is one press with nothing to moderate. One identity holds one vote on one card and that is a rule in the database rather than a check in a handler.
Somebody determined can clear a cookie and vote again. Somebody determined can also make five accounts on a tool that requires them, so this is a smaller difference between the two designs than it is usually sold as.
Should you turn it on?
Yes if you are trying to find out what people want, and no if you are trying to find out who wants it.
Those are different jobs. A board collecting anonymously gives you a wide, cheap, slightly noisy read on demand, and it will not tell you which account is behind any row. A board requiring identity gives you a narrow, expensive, precise one.
VoteFirst is anonymous by default and you can layer identity on top where it matters. One setting, Only signed in visitors can vote, comment and suggest, turns anonymous voting off while anyone can still read the board. The board is nine dollars a month, and what a vote is worth once you can see who cast it is the next thing to read.
14 days, a card at signup, then $9 or $39 a month.
