BetterId™
Solution: For organizationsJoin the beta
Become a design partner

Solution · For organizations

One directory, every app. Nothing to steal.

A self-hosted, password-free org IdP on provenance logs.

Your people use dozens of internal and external apps, each with its own account. With BetterId IdP, your organization runs one directory for all of them. Add someone once and they get every app; remove them once and every app knows. Everyone signs in with a key on their phone, and no password exists anywhere. The first release is being built to work behind the IdP you already run, Microsoft Entra first; then you replace it one group at a time.

Identity is the employee’s work VLAD; proof methods are pluggable. Today that’s a device-bound passkey on their phone (WebAuthn over QR and Bluetooth) or a tap on the phone through the linked browser extension. Every method is checked against the employee’s provenance log and the organization’s own quorum-signed log, and a single Rust core does every cryptographic check. The first sign-in service speaks OIDC behind your current IdP; SAML, SCIM out, CAEP and short-lived certificates are designed for later.

  • Add someone once, and they get every app
  • No password database, and no password-reset desk
  • Joiners, movers, and leavers reach every app within seconds
  • Runs beside your current IdP, with no flag day
  • No passwords, OTP codes or plain push: every method is origin-bound and checked against the log
  • Each sign-in is designed to report its method and assurance level (amr/acr), so policy can require passkeys for a group
  • Directory = org provenance log, m-of-n admin quorum in HSM/KMS
  • Works behind Entra and AD FS first; Okta, Ping and Keycloak when a customer runs them

Add someone once. Every app knows.

In development

BetterId IdP is being built with a small group of design partners. Android sign-in works end to end in test builds; iOS is next. The first release is designed to work behind the IdP you already run, Microsoft Entra first. Single sign-on and SCIM in are on the pilot plan below, not shipped. If you run Entra, AD FS, Okta, Ping, or Keycloak and want to pilot it, talk to us.

Identity, proof, policy

Your identity lives in BetterId, not on a device.

Approve the device, not the password. Revoke it once and it’s gone everywhere. Losing a phone or adding a laptop never changes who someone is; it only changes how they prove it.

The identity is the work VLAD, not a device. Proof methods are pluggable, and each is checked against the employee’s log and the org log, so one revocation entry stops that key at every app and IdP at once. A passkey alone can’t do that.

LayerWhat it isChanges over time?
IdentityEach person’s work identity and its signed log. Every app knows them by itNever
ProofToday: a passkey on an Android 14+ phone, or a tap on the phone through the browser extension. Planned: laptop keys, and Windows and macOS logonYes: devices come and go, and better methods are added as they appear
PolicyThe org log lists which devices you approve and which methods each group may useYes: one entry revokes a device everywhere

One directory, every app

Add someone once. Remove them once.

App sprawl is where access goes wrong: joiners wait, people who change roles keep old access, and leavers keep accounts nobody remembered to close. BetterId IdP puts every app behind one directory. Here’s what each piece does, and the stage of the pilot plan it arrives in. Pieces marked “Later” are designed, and get built when a customer needs them.

Entra (or your IdP) feeds the org log by SCIM in, and a small OIDC sign-in service runs behind it. Each piece lands at a stage of the evaluation path; SCIM out, CAEP, a full SAML IdP and certificates are designed for later, when a customer pulls them.

NeedHow BetterId IdP covers itArrives
One sign-in for every appOIDC behind your current IdP, approved with the person’s phoneStage 1
Accounts in one placeSCIM in from Entra (or your IdP), or people added by an admin or a CSVStage 1
Changes everywhere, fastA departure is one signed entry, and every BetterId sign-in checks it; sessions are kept shortStage 2
Provisioning and instant session endSCIM out to your SaaS apps, HR feeds (Workday, SuccessFactors, BambooHR), and session-end events to apps (CAEP)Later
SSH, VPN and Wi-FiShort-lived certificates from the same identityLater
Apps without SSOCredential Vault fills them, approved on the phone; the secrets are managed by the organization and never seen by staffLater
AI agentsBetterId API Delegate gives agents scoped API access, logged against a person’s identityDesign partners
AgreementsBetter OpenSign signs documents with the same identity people sign in withDesign partners

Benefits

Okta already centralizes accounts. What’s different?

Every IdP sells one place to manage access, and Okta and Entra already offer passkeys. The difference is what the central directory is made of.

∅

No credential database to breach

The sign-in service holds nothing that proves who someone is. A stolen database or server can’t produce a login that a self-checking app will accept.

Postgres holds protocol state only: clients, codes, refresh families, SCIM cursors. Losing it logs people out; it doesn’t lose who works there.

⛓

A directory that can’t be quietly edited

Every hire, role change, and departure is a signed entry that needs several admins to approve. Auditors read the log instead of trusting screenshots.

Membership, roles, and device approvals live in the org provenance log. An m-of-n lock script enforces the admin quorum, so a compromised host cannot add itself an admin.

⚡

Leavers lose access in seconds

When someone leaves, one signed entry turns them off, and every BetterId sign-in checks it within seconds. Sessions are kept short, so there’s little left to chase.

The org node streams new log heads to validators, and the sign-in service refuses the next sign-in and revokes refresh tokens. CAEP session-revoked and SCIM-out deprovisioning are designed for later.

☎

No more “reset my password”

There’s no password, so there’s no reset, and no help-desk call for an attacker to talk their way through. A lost phone is handled by identity proofing and admin approval.

Recovery is an org-quorum key rotation on the employee’s work VLAD after policy-selected identity proofing. It is a signed, visible log entry.

◇

One identity for every door

The same work identity is designed to sign people into web apps first, then Windows and macOS, and later servers, VPN and Wi-Fi, when a customer needs them.

One work VLAD drives OIDC sign-in first, then PKINIT and Platform SSO; OpenSSH user certs, EAP-TLS and SPIFFE SVIDs from vlad-ca come later.

⇋

Partner access without federation projects

A partner’s staff prove who they are with their own company’s log. You decide which roles you accept; there’s no per-partner setup project.

/partners/<org-vlad> in your org log replaces per-pair SAML federation; partner membership is checked against their org log.

⌂

You run it, and it outlives us

The service and the network it depends on run in your environment. Sign-in keeps working if the internet’s public network, or Cryptid, is down.

The org node pins your org and employee logs, optionally on a private network. Logs verify without any Cryptid service.

↔

Built for the stack you have

The first release is a small sign-in service that sits behind the IdP you already run. Full editions in the languages your team uses come later, when customers ask, and share one set of security-critical code.

First, bs-orgd: a small OIDC service over one Rust core, plugged into Entra (then AD FS) as a factor. Java, .NET and Python editions come later, on request, gated by a shared conformance suite. See BetterId IdP.

Built from

The products behind it.

One directory, every app, from BetterId products working together. Each has its own page.

Evaluate, then replace

Move in stages. Stop at any one.

You don’t rip anything out to try BetterId. Each stage runs alongside what you have, rolls out one group of people at a time, and has a clear way back. Stages 0 and 1 are what we’re building now with design partners, Stage 2 follows, and Stage 3 is built as customers need it.

0

Prove it in a sandbox

A few weeks · IT team only · no production change

Steps

  1. Stand up the org node, validator, and reference host in a test environment.
  2. Create your org log and admin quorum, with keys in a test HSM or KMS.
  3. Enroll 5–10 IT staff on their phones (Android 14+ today).
  4. Connect one internal test app over OIDC or SAML.
  5. Run the conformance harness and try to break it: replay, wrong origin, revoked device, terminated member.

You measure

  • Sign-in success and time to sign in
  • Enrollment time per person
  • Seconds from “terminated” to access gone
  • Your security team’s review findings
Passwords
Untouched
Your IdP
Untouched
Way back
Turn the sandbox off
1

Add it as a factor

Pilot group on real apps · your IdP still in charge

Steps

  1. Register BetterId with your IdP as a phishing-resistant factor (see the table below).
  2. Pick a pilot group, for example IT plus one willing team, and push the app with your MDM.
  3. Let Entra (or your IdP) push joiners and leavers by SCIM, or add them by an admin or a CSV, and approve enrollments with identity proofing.
  4. Require the BetterId factor for the pilot group on a few real apps.
  5. Expand the group as the numbers hold.

You measure

  • Phishing-resistant coverage of the pilot group
  • Help-desk tickets per user, against your baseline
  • Lost-phone recoveries and how long they take
  • User satisfaction
Passwords
Still exist
Your IdP
System of record, first factor
Way back
Remove the factor from policy
2

Make it the only way in

Migrated users have no password at all

Steps

  1. Route migrated users to BetterId as the upstream identity provider (Entra domain federation, or AD FS); your IdP keeps issuing tokens to apps.
  2. Remove password sign-in for those users and delete their password hashes.
  3. Keep sessions short, so a revocation in the log takes effect quickly.
  4. Migrate group by group, starting with the people phishing targets most.

You measure

  • Password hashes deleted
  • Password-reset tickets for migrated users (target: zero)
  • Leaver propagation time across apps
  • Sign-in success across devices
Passwords
Gone for migrated users
Your IdP
Token issuer; trusts BetterId upstream
Way back
Route the group back and re-issue credentials from your IdP
3

Retire your old IdP

BetterId is the directory and the identity provider · each step is built when a customer needs it

Steps

  1. Make the org log the system of record; HR feeds it directly.
  2. Point apps at BetterId over OIDC and SAML, and provision them by SCIM out, with instant session-end events.
  3. Move SSH to short-lived certificates and VPN / Wi-Fi to certificate login.
  4. Move workstation logon, the hardest step: Windows and macOS are planned after Stage 2, Linux on request.
  5. Put legacy password-only apps behind managed secrets no person ever sees.
  6. Retire the previous IdP one business unit at a time, or keep it downstream for apps that can’t move.

You measure

  • Apps and users served directly
  • Remaining places a person types a password
  • Audit evidence pulled from the log
  • IdP licence and help-desk cost removed
Passwords
Gone
Your IdP
Retired, or a downstream app
Way back
Keep the old IdP downstream until you’re sure

How BetterId connects to your current IdP

Your IdPStage 1: as a factorStage 2: as the primary sign-in
Microsoft Entra ID (first)External authentication methodFederate the domain to BetterId (SAML / WS-Fed)
AD FS (next, if you run it)Additional authentication providerBetterId as the primary authentication method
Okta (on request)OIDC identity provider used as a factorRouting rule sends migrated users to BetterId
Ping (on request)PingFederate adapter or OIDC IdPPrimary authentication source
Keycloak (on request)Authenticator in the browser flowBetterId is the only step; the org log becomes the user store

Coverage

Every place a password lives today.

“Password-free” only counts if it’s everywhere. Web sign-in comes first. Windows and macOS logon follow. SSH, VPN and the rest come when a customer needs them.

Where passwords liveWhat replaces themWhen
SaaS and internal web appsSign-in over OIDC or SAMLEarly
Apps behind Entra, AD FS or OktaYour IdP trusts BetterId as a factor, then as the upstream IdP (Microsoft first)Early
Help-desk password resetsIdentity proofing plus admin-quorum recoveryEarly
Break-glass admin accountsQuorum keys in an HSM and sealed hardware tokensEarly
Windows logon and Active DirectorySmart-card-style certificate logon; AD rolls the old password awayNext
macOS logonPlatform SSO with a BetterId extensionNext
SSH to serversShort-lived SSH certificates after a phone approvalLater
VPN and Wi-FiCertificate login (EAP-TLS) with short-lived certificatesLater
Service accounts and workloadsWorkload certificates (SPIFFE) or mutual identity TLSLater
Legacy password-only appsA managed secret that’s generated, injected, and rotated; no person sees itLater
Linux workstationsPAM module that asks for a phone approvalLater

Questions

Frequently asked.

Is it available today?

It’s in development with design partners. Phone sign-in works end to end on Android in test builds, and iOS is next. The first release is being built to work behind the IdP you already run, starting with Microsoft Entra. Design partners start at Stage 0 and help shape Stage 1. Get in touch if you’d like to be one.

Where are the security questions?

On the BetterId IdP page: what happens if the server is compromised, whether the company can impersonate an employee, which devices it needs, and how it’s hosted.

Do we have to replace Okta or Entra to try it?

No. Stages 0 and 1 leave your IdP in charge; BetterId is added as a factor for a pilot group, behind Entra first, and Okta or another IdP when a customer runs it. You only move further if the results justify it, and you can stop at any stage.

Can our staff give AI agents API access without sharing credentials?

That’s what BetterId API Delegate does. A person signs in with their identity, grants an agent specific API permissions for a set time, and every API call is checked and logged against them. The agent never holds an API key or password, and a revoke cuts it off on its next call. It’s in preview for individuals today; we’re shaping the company version with design partners.

How is this different from BetterId Credential Vault?

BetterId Credential Vault, included in the BetterId app, keeps a person’s own saved logins on their phone. BetterId IdP removes passwords from your organization entirely. For apps that still demand one, it plans to fill managed secrets that no employee ever sees, approved on the phone the same way (later, when a customer needs it).

Some of our people already use the BetterId app. Does that help?

Yes. They already have the app and the habit of approving on their phone, so joining is a tap. Your organization gives each employee a separate work identity in that same app, which you can recover or turn off without touching their personal one. Contractors and partners can use their own identity instead.

Start with Stage 0.

A sandbox, your IT team, one test app. No production change, no commitment.