Skip to main content
InviziPoll
Sign inStart free trial
For HR and People teams

Get the answers people won't say out loud.

Employee surveys your people actually believe are anonymous, so you get the honest answer, not the careful one.

Start free trialAnswer a sample survey

No credit card. Workspace in < 2 min.
Used for engagement pulses, exit surveys and upward reviews.

Plans from $49/mo. See pricing

Pro $49Business $249Enterprise from $499/pricing
How the encryption works

Illustration: a crowd of anonymous respondents. Each answer is encrypted in the browser, held in a reserve pool, and joins the results only as part of a batch released together. No individual is identifiable.

Reserve pool2 heldillustrative
Q3 engagement pulse
open · batches of 3 to 5 · k=3 min group
152responses
WORKLOAD148 responses · min 3
41%of us have been asked to work time we can't log.
RETENTION146 responses · min 3
1 in 3of us is planning to leave this year.
Q3 STRATEGY141 responses · min 3
19%of us believe the current strategy will work.
Break down by department, tenure or location once the poll closes.

Example results, for illustration.

Why InviziPoll

A survey tool built so honesty isn't a risk

Engagement pulses, 360s, exit surveys, board polls: the questions that matter most are the ones people won't answer if they think someone is listening.

Respondents trust it, so they tell the truth

Responses are encrypted on the respondent's device before they're sent. Not fingerprinted, and not readable by us, even if we wanted to look. When your team sees "we literally cannot read this," the feedback stops being careful.

respondents see: a clean link: no login, no cookies, no tracker

Anonymity you can't accidentally break

Set a minimum group size and nothing shows until that many people have answered. Once the survey closes, any cut by team or location too small to protect someone is left out, including cuts that would single a person out by subtraction. No nervous manager ever has to do this math in their head.

k-anonymity built in (never fewer than 3, always enforced)

Built for the long term

Uses the post-quantum algorithms NIST standardized in 2024, designed to resist attacks from future quantum computers as well as today's. That matters when the questions are about people.

algorithms: ML-KEM-768 + ML-DSA-65

proof, not promises: verify it yourself

Security & threat model/securityEncryption docsML-KEM-768Anonymity guaranteesciphertext-onlyReviews & proof artifactsprimary-source

How it works

Readable on two devices. Sealed everywhere in between.

The path a single response takes from keystroke to your results, and the only part of it we ever hold.

Respondent's browserplaintext · device only
  1. 01 / RESPOND

    Answer in browser

    A signed link opens the poll. Selections live only in local memory: no login, no cookies.

  2. 02 / ENCRYPT

    Seal with workspace key

    The browser fetches the workspace public key and wraps the response with ML-KEM-768.

sealed before it leaves
InviziPoll serversciphertext only
  1. 03 / STORE

    An opaque blob

    Plus a blind signature. No IP, no fingerprint, no session. A subpoena to us gets ciphertext and account records, never an answer.

opened only here
Your browserplaintext · your device
  1. 04 / RESULTS

    Decrypt on your device

    Answers decrypt locally into charts and a shuffled list of written comments, with no name, time or order attached. Nothing renders until your minimum group has answered.

full architecture & threat model → /security · end-to-end encrypted survey

What you see

Results you can act on. Nobody you can single out.

Once the survey closes, break results down by department, tenure or location. Any group smaller than 5, or your minimum group size if that's larger, is withheld automatically.

  • Nothing shows until your minimum group has answered, and a few answers are held back while the survey is open
  • A group too small to protect someone is withheld, and so is any group that would let a small one be worked out
  • Written comments are shown word for word, in a shuffled list with no name, time or order attached

Try it: switch the breakdown, then open the verbatims and the audit packet.

app.invizipoll.com · Q3 engagement pulseillustrative data
Break down by:
“I can raise a concern without worrying about the consequences.”
4 to 5 · Agree3 · Neutral1 to 2 · Disagree
  • Everyonen=111
  • Engineeringn=42
  • Salesn=31
  • Supportn=19
  • Operations
    Withheld
    n hidden
  • Legal
    Withheld
    n hidden
111 responses releasedk = 3 · groups need 5+ · answers linked to no one

Where it matters

Built for the questions people dodge

Response quality depends on respondents believing (correctly) that answers can't be traced back.

Engagement & pulse surveys

Monthly temperature checks managers can't personally audit.

Why anonymityPeople score carefully when they think someone is watching.

Pulse surveys

360° & upward feedback

Peer and manager reviews protected by cohort-size guardrails.

Why anonymityNobody critiques their boss on the record.

Anonymous employee survey

Exit & stay interviews

Honest departure feedback with no HRIS trail linking rows to names.

Why anonymityReferences outlive the exit form.

Anonymous exit interview

Board & exec polls

Close-quarters decisions where attribution chills candor.

Why anonymitySmall rooms have long memories.

Browse templates

On-device AI

Themes from written answers, without uploading a single one.

Most engagement tools send comments to a model vendor to find themes. Ours runs a small open-source model inside the admin's browser tab, on answers already decrypted on that device. No answer, prompt or summary leaves it, and summaries add no AI subprocessor to your vendor list.

Typical toolComments uploaded to an AI vendor for summarising
InviziPollSummarised in the admin's tab. Nothing sent.
What runs where: on-device AI summaries

Pricing: published, in USD

No "contact sales" wall for a price

Three tiers and a consultants licence, quoted on one page. Annual billing saves ~20%.

Pro
$49/ workspace / month

For teams that need professional anonymous feedback.

  • 200 active polls, 500 recipients per poll
  • End-to-end encrypted polls
  • Post-quantum cryptography
Start trial on Pro
Business
$249/ workspace / month

For organizations scaling a culture of feedback.

  • Everything in Pro
  • SAML 2.0 & OIDC single sign-on
  • SCIM 2.0 user & group provisioning
Start trial on Business
Enterprise
From $499/ month

For compliance-heavy orgs with procurement requirements.

  • Everything in Business
  • Advanced Audit Logs (expanded tenant visibility)
  • Dedicated account manager
Get a quote
Annual cost at your size

Priced per workspace, not per head.

Per-seat tools bill you for every employee, every month. InviziPoll is one flat price for your workspace.

Typical per-seat engagement platform
$8 per employee per month × 150 × 12 months
$14,400 / year
Goes up with every hire.
InviziPoll Profits up to 500 per poll
$39 / month flat, billed annually
$468 / year
Post-quantum encryption, AI authoring, Slack, cross-tabs.
$13,932 a year back in your budget, about 31x lessPer-seat figure is an illustrative list price for per-employee engagement platforms, not a quote from any vendor. InviziPoll figures use annual billing.

Running surveys for client companies? There's a licence for that, with 5 client workspaces included. See the consultants licence.

full comparison, limits & annual pricing → /pricing

Questions

Asked by every security review

Is InviziPoll actually anonymous, or "anonymous" like other survey tools?
Most tools promise anonymity in a privacy policy while storing readable responses next to respondent metadata. InviziPoll is anonymous by architecture: responses are encrypted in the respondent's browser before they reach us, and our servers persist ciphertext only. See the full argument at /trust/anonymity.
Can an admin see who submitted a response?
No. Respondent links are cookieless and untracked, and nothing stored with an answer says who sent it. Admins see totals and charts, and every written comment word for word, in a shuffled list with no name, time or order attached. Nothing shows until your minimum group has answered, and a cut too small to protect someone is left out entirely.
What does "post-quantum" mean here?
Responses are sealed with ML-KEM-768 and signed with ML-DSA-65, NIST-standardized algorithms designed to resist attack by future quantum computers. Feedback about people has a long shelf life; the encryption should too. Details in the encryption docs.
Your AI reads written answers. Does it send them anywhere?
No. Most engagement tools send comments to a model vendor to find themes (compare tools). Ours runs a small open-source model inside the admin's browser tab, on answers already decrypted on that device, so no answer, prompt or summary leaves it and summaries add no AI subprocessor to your vendor list. Summaries are labelled as drafts and stay off shared links and exports unless you opt in. What runs where, in detail: on-device AI summaries.
Where is our data stored?
In the United States (AWS). We process and store customer data in the United States, and we don't offer regional hosting yet. Answers are encrypted on each respondent's device before they reach us, so the answers we hold are unreadable to us wherever they sit. The full list is on subprocessors.
Do you publish pricing?
Yes. Pro at $49 and Business at $249 per workspace per month (USD), Enterprise quoted individually, all on one page at /pricing. No demo call required to learn a number.

FAQ answers also link to anonymity, encryption docs, and pricing.

Give your team a voice.
Keep their identity.

Create your workspace in under two minutes. No credit card. We don't track respondents: ballots stay cookieless and unlinked.

Read the security model