Nine tools compared on method, account requirements and free limits. The first thing worth knowing is that "ranked choice" describes several different counting methods, and the tools here do not all use the same one.
For a quick group decision with no account for anyone, VoteGenerator runs instant-runoff free and shows every elimination round.
For dedicated ranked-choice features — voter lists, registration, duplicate detection, site embedding — RankedVote is purpose-built for it and requires an organiser account.
For formal correctness, CIVS uses Condorcet methods rather than instant-runoff, is free, and is run by Cornell. Its own site explains why it considers Condorcet fairer than instant-runoff, and the argument is a good one.
For a real election with single transferable vote or a verified voter list, OpaVote and ElectionBuddy are built for that and charge accordingly.
| Tool | Account to create | Account to vote | Method | Free tier | Best for |
|---|---|---|---|---|---|
| VoteGenerator | No | No | Instant-runoff | 100 responses, 3 active polls | Quick group decisions with no account for anyone |
| RankedVote | Yes | No | Instant-runoff | Basic RCV polls and elections | Dedicated RCV with voter lists and embedding |
| RCV123 | Optional* | No | Instant-runoff | Free, no published cap | Free RCV from a non-profit, classroom-friendly |
| RankedChoices | Not published | No | Instant-runoff | Free, donation-supported | Long-running free tool for informal group votes |
| CIVS | Yes | Email required | Condorcet | Free | Academic and organisational elections wanting Condorcet |
| OpaVote | Yes | Depends on type | IRV, STV, Condorcet, Borda | Polls have unlimited voters** | Formal elections needing STV or Condorcet |
| ElectionBuddy | Yes | Emailed ballot | Instant-runoff and others | Free for basic elections** | Clubs, HOAs and nonprofits running real elections |
| Helios | Yes | Yes | Approval and ranked methods | Free, open source | End-to-end verifiable elections, if you can self-host |
| StrawPoll | No | No | Not published | No published vote cap | High-volume public polls, if RCV isn't essential |
* RCV123 lets you create a ballot without signing in, but you cannot manage, edit, close or delete it afterwards without a free account.
** OpaVote states polls always have unlimited voters, while elections are priced at roughly $10 per 125 voters; third-party reviews report a free election tier of 25 to 30 voters and disagree with each other, so check OpaVote's own pricing page. ElectionBuddy's free tier is reported as covering basic elections, with published paid figures varying between sources. All details from vendor documentation as of August 2026.
No, and picking the wrong one produces a result that answers a different question than you asked.
One winner. Count first preferences; if nothing has a majority, eliminate the lowest option and redistribute its votes to each ballot's next choice. Repeat until something passes 50%. Used in Alaska, Maine and New York City.
Several winners, proportionally. Transfers both surplus votes from winners and votes from eliminated candidates. What you want for electing a committee that reflects the spread of preferences rather than just the majority.
Selects the option that beats every other option in a direct head-to-head comparison, where one exists. Formally stronger than instant-runoff, and harder to explain to a room.
Assigns descending points by rank position and totals them. Simple to compute, but vulnerable to strategic voting and to options being added or removed.
Most consumer tools mean instant-runoff when they say "ranked choice". CIVS means Condorcet. OpaVote supports several and lets you recount the same ballots under a different method, which is genuinely useful if you want to see whether the method changed the answer.
| Tool | Instant-runoff | STV | Condorcet | Rounds shown | Embed | Voter verification | Other poll types | Slack app |
|---|---|---|---|---|---|---|---|---|
| VoteGenerator | Yes | No | No | Yes, every round | Yes | Single-use codes (Business) | 8 types | Yes, native app |
| RankedVote | Yes | Not published | No | Yes | Yes, paid plans | Voter lists, paid | RCV only | Not published |
| RCV123 | Yes | Not published | No | Yes | Not published | With free account | RCV only | Not published |
| RankedChoices | Yes | Multi-winner | No | Not published | Not published | Not published | RCV only | Not published |
| CIVS | No | No | Yes | N/A, Condorcet | Not published | Email ballots | Condorcet only | No |
| OpaVote | Yes | Yes | Yes | Yes, recount page | Not published | Email ballot links | Also approval, Borda | Not published |
| ElectionBuddy | Yes | Yes | Not published | Yes | Not published | Yes, paid | Election ballots | Not published |
| Helios | Not published | Not published | Not published | Cryptographic audit | Not published | Yes, verifiable | Election ballots | No |
| StrawPoll | Not published | No | No | Not published | Yes | Duplicate checks | Multiple choice | Not published |
"Not published" means the vendor does not state it clearly on their site — we have not marked it absent, only unverified. VoteGenerator is the only tool here with a native Slack app, offering slash commands, in-message voting and timezone-aware recurring polls; it is also the only one with poll types beyond ranked choice, including meeting scheduling, RSVP tracking, rating scales, image polls and multi-question surveys, plus 120+ editable templates.
It is the method VoteGenerator uses, and it is worth stating its known weaknesses plainly rather than leaving you to find them elsewhere.
Instant-runoff is non-monotonic. In rare constructed cases, ranking an option higher can cause it to lose. No amount of implementation care fixes that; it is a property of the method.
It fails the Condorcet criterion. It can elect an option that a majority would have ranked below another. CIVS says exactly this on its own site as the reason it uses Condorcet methods instead, and the criticism is fair.
Our own simulation found instant-runoff matched the head-to-head Condorcet winner 84.5% of the time with five options, against single-choice voting's 64.1%. Closer, not perfect. Arrow's impossibility theorem rules out any ranking method satisfying every fairness criterion at once, so the question is never which method is flawless.
Why most tools use it anyway: instant-runoff is explainable in one sentence and its result can be shown round by round. A method a team can follow gets used and trusted. A formally superior method nobody understands gets argued with. If formal correctness matters more than comprehension in your situation, use CIVS.
Worth knowing before deciding whether any of this matters for your group.
We simulated 660,000 group votes under a spatial preference model to measure it. With five options on the ballot, the single-choice winner held no outright majority 84.3% of the time. Single-choice and instant-runoff picked different winners in 28.7% of five-option votes, rising to 43.2% at eight options.
The sharpest finding is arithmetic rather than statistical: if a group who prefer similar things splits between two of them, they lose under single-choice voting unless they hold more than two-thirds of the vote. Disagreement between the two methods held at roughly 28% whether the group was six people or a hundred — group size barely matters, the number of options does.
The full methodology and the simulation code are published, seeded for reproducibility, so the figures can be checked rather than taken on trust.
These are not ranked best to worst. They are built for different jobs.
Whether the tool shows the elimination rounds. An instant-runoff result that appears as a single winner with no working shown is hard to defend to anyone who ranked something else first. VoteGenerator, RankedVote, RCV123 and OpaVote all display them.
How many options you're asking people to rank. Beyond about eight, people rank the top few carefully and the rest arbitrarily, which corrupts the later rounds. Shortlist before you ballot.
Whether you need identity verification. Cookie-based duplicate prevention deters casual double-voting and nothing more. If the outcome has legal weight, you need emailed individual ballots or a verified voter list.
What the tool stores about voters. Anonymity implementations vary widely. VoteGenerator's Anonymous Mode is opt-in per poll and enforced server-side — device, country, region and referrer are never collected rather than collected and hidden, and timestamps are stored as dates. We cannot audit how the others implement it, which is why it's worth asking each one directly.
Every free tool here is designed for decisions, not for elections with legal consequences.
For a board election, union ratification, HOA vote or shareholder resolution you need ballot secrecy, an audit trail and a one-person-one-vote guarantee. That means CIVS, OpaVote, ElectionBuddy, Simply Voting or a comparable platform, and usually a fee. Helios goes furthest with end-to-end verifiable cryptography, but requires self-hosting and technical expertise.
VoteGenerator's Business plan issues unique single-use access codes and can restrict embedded polls to approved domains, which sits between a casual poll and a formal election. For anything where the result could be challenged, use a platform built for it.
Instant-runoff, every round shown, no account for you or your voters.
Try Ranked Choice →