Microsoft Authenticator: Free MFA With Push Approvals
Microsoft Authenticator on Android: TOTP beyond Microsoft accounts, backup restore limits, and why number-matching push approvals can go missing.
App summary
- App
- Microsoft Authenticator
- Developer
- Microsoft Corporation
- License
- Proprietary
- Version
- unknown
- Size
- unknown
- Price
- Free
Microsoft Authenticator Beyond Microsoft Accounts: Where TOTP Works on Android—and Where the Claim Breaks
The most common question about this app is some version of “does it only work with Microsoft accounts?” The answer starts with the function list, because that list separates two things people routinely merge into one. Push approvals and number matching sit on one side. TOTP codes and passwordless sign-in sit on the other. They are different mechanisms that happen to live in the same app, and they have different scopes.
A push approval is a prompt that arrives because a sign-in attempt was made against an account configured to send one. A TOTP code is generated by the app itself, on the device, from a shared secret established when the account was added. The first depends on a service pushing a request to you. The second depends on nothing but the app and the clock. That distinction is why “only works with Microsoft accounts” is the wrong conclusion: the code generator is listed as a function in its own right, not as a Microsoft-only feature.
What the supplied facts do not do is name a single third-party service. Google, Facebook, and every other provider are absent from the fact data. So the honest position is narrower than either extreme. The app generates TOTP codes, and a service will work with it if that service accepts standard TOTP. Per-service compatibility is unknown until you check that specific service’s own documentation. Anyone telling you it works with everything is guessing; anyone telling you it works with nothing but Microsoft is also guessing.
There is a second correction buried in the same confusion. A code generator is not credential storage. The fact data lists password managers and 2FA as a topic in its own right, which is a signal that the two are being treated as related but separate concerns. Do not add an account to this app and conclude that the password for that account is now managed. It is not. You have added a second factor, and the password still lives wherever it lived before.
On the question of whether you can use the app without a Microsoft account at all: the facts do not state an account requirement for opening the app or for generating codes. What they show is that push approvals and TOTP codes are listed as distinct functions, so an account requirement attached to one does not automatically describe the other. Treat that as unresolved rather than settled in either direction.
Microsoft Authenticator Backup on Android: What Restore Can Promise—and Why ‘All Accounts’ Is Unproven
Backup and restore appears in the fact data as a listed app function, grouped with app lock. That grouping matters, because it tells you what kind of claim is being made. A listed function is a capability statement: the app has backup, and the app has restore. It is not a coverage guarantee, and it is certainly not a promise that every account you have ever added will come back after a phone reset.
The specific unknowns are worth naming. The facts do not say which account types are included in a backup. They do not say whether TOTP secrets and any stored password entries are both covered or only one of them. They do not describe the encryption protecting the backup, where it is stored, or what is required to decrypt it on a new device. Every one of those details is load-bearing if you are relying on restore as your recovery plan, and every one of them is missing.
So when someone asks “does the backup restore all accounts?”, the defensible answer is that the facts support “the app has backup and restore” and nothing stronger. The leap from a listed feature to an all-account guarantee is exactly the leap that produces the search query “microsoft authenticator backup not restoring android.” A partial restore is not necessarily a bug. It may simply be the boundary of what the feature was ever described as doing.
App lock does not close that gap, and it is important not to let it appear to. App lock is a gate on the device: it controls who can open the app on the phone in your hand. It says nothing about what leaves the device, what comes back, or how complete the round trip is. An app can be locked and still have an incomplete backup. The two features are adjacent in the fact list, not causally linked.
The recovery question—what happens after you lose or reset your phone—runs into the same wall. The fact data does not describe a recovery procedure, a recovery key, or an account-recovery flow. If you are building a plan around this app, the plan has to include the possibility that restore covers less than everything, and it has to include a second factor method you can still reach when the phone is gone.
One more distinction belongs here. A restored authenticator is not automatically a full standalone password manager. Even a perfect restore returns an authenticator to a device. It does not turn the app into a credential vault with the scope of a dedicated password manager, because the fact data never describes it as one.
Microsoft Authenticator Number Matching on Android: Why Push Approvals Can Go Missing
Number matching is a component of push approvals, not a sibling of TOTP codes. It is the part of the approval flow where the prompt is tied to a number so that a blind “approve” tap is not enough. Understanding that placement resolves a lot of confusion, because it tells you when a number-matching prompt should appear at all: only when a service is sending a push approval to an account configured for one.
The first diagnosis for a missing prompt is therefore not a setting. It is whether a push was ever supposed to arrive. If the service in question is set up to use TOTP, the app will generate a code and no approval prompt is expected. Some “number matching not showing” reports are not failures. They are a mismatch between what the user expects the app to do and what the service actually asked for.
The second diagnosis is Android itself. Push approvals depend on the app receiving a notification, and notification delivery on Android is subject to per-app notification permissions, background execution limits, and battery optimization behavior that varies by manufacturer and OS version. Those are the usual suspects when a prompt is late or absent. The supplied facts do not include specific Android settings, menu paths, or vendor-specific battery controls, so any single prescribed fix would be invented. Mark it unknown and work through the device’s own notification and battery settings rather than trusting a checklist.
There is a third possibility that gets less attention: the account was never enabled for push approvals in the first place. Push approvals are a function the app supports; they are not a state every account automatically occupies. If the account is configured for codes, codes are what you get.
Two related misconceptions should be retired here. Phone sign-in does not work for all personal and third-party accounts—the fact data lists passwordless sign-in as a function without claiming universal coverage, and coverage depends on the service and the account configuration. And TOTP codes do not sync seamlessly across all devices. The fact list includes backup and restore, not synchronization. Restore is a recovery event; sync is a continuous state. They are not the same thing, and the facts only support the first.
Microsoft Authenticator vs Google Authenticator, Authy, and Aegis: The Android TOTP Portability Gap
The comparison that actually matters on Android is not which app looks better. It is whether you can get your TOTP secrets out. The fact data lists TOTP codes, backup and restore, and app lock as functions. It does not list cross-app export. That absence is the whole story for anyone who has been burned by a locked-in authenticator before: export is unknown, and unknown should be treated as unavailable until proven otherwise.
Backup and restore is not export, and conflating them is the most expensive mistake in this category. Backup restores into the same ecosystem—it is a recovery path for you, on your devices, within the vendor’s arrangement. Export moves the seed itself to a different app entirely. A backup that works perfectly tells you nothing about whether you can hand your accounts to another authenticator, and the facts describe only the former.
Against Google Authenticator, Authy, and Aegis, the comparison has to be run on the dimensions the fact data actually supplies: TOTP generation, backup and restore, app lock, and license. The supplied facts describe this app as proprietary. They say nothing about the licensing or feature set of the other three, so any claim about how those apps handle export, backup, or audits would be invented. What can be said is directional: if portability of seeds is a hard requirement, verify it per app before you commit, and do not assume it from the presence of a backup feature.
The wrong audience becomes clear from that framing. Android users who need guaranteed portable TOTP seeds should not assume this app fits. People who require an open-source, independently audited authenticator are a direct mismatch with a proprietary license. Anyone wanting a standalone password manager without a Microsoft account is asking for something the fact data does not describe. And users who want a hardware security key instead of a phone app are in a different product category entirely.
What the Free, Proprietary License Means for a Security App
Two facts sit next to each other in the data and deserve to be read together: the app is free, and the app is proprietary. Free removes cost as a barrier, which is genuinely useful for a second factor—the cheapest security tool is the one people actually install. Proprietary means the source is not published for outside inspection, so the trust you place in the app is trust in the vendor’s process rather than trust in public review.
For a security tool, that trade-off is the whole argument. You are not evaluating code you can read; you are evaluating a company’s track record and the behavior of a binary you install from the official Android store listing. The fact data does not describe any independent audit, certification, or third-party review, so none should be assumed. Absence of a stated audit is not evidence of a problem, but it is also not evidence of assurance.
This is where the wrong audience is easiest to name. If an open-source, independently audited authenticator is a requirement rather than a preference, the proprietary license listed here is a direct conflict, and no amount of feature parity resolves it. That reader should look elsewhere and should apply the same scrutiny to whatever they choose, since the supplied facts do not cover the licensing of any alternative.
Push Approvals, TOTP Codes, and Passwordless Sign-In Are Not One Feature
The function list reads like four items, and it should be read that way. Push approvals and number matching. TOTP codes and passwordless sign-in. App lock, backup, and restore. Comparison with other authenticator apps. Password managers and 2FA. Treating that list as a ladder—“it does push, so it must do everything”—is the source of most of the confusion this app attracts.
Passwordless sign-in is the item most often overread. The misconception is that phone sign-in works for every personal and third-party account. The fact data lists passwordless sign-in as a function and stops there. It makes no claim about which services support it, which account types qualify, or whether a personal Microsoft account and a work account behave the same way. Coverage is a per-service question, and the facts do not answer it.
TOTP codes get the same treatment in the opposite direction. People assume that because codes are generated on a phone, they must appear on every device signed into the same account. The fact data lists backup and restore, not sync. Those are different promises. Restore happens when you move to a new device and recover. Sync would mean a code generated on one device is continuously visible on another. Only the first is described.
The practical consequence is that you should plan for one device being the primary authenticator, with restore as the recovery path rather than a second live copy. If a code fails—the perennial “why are my codes not working” question—the fact data offers no troubleshooting guidance, so the productive checks are whether the service actually expects TOTP rather than a push, and whether the account entry matches the service you are signing into.
App Lock Protects the Phone in Your Hand, Not the Account Behind It
App lock is listed as part of “app lock, backup, and restore,” and its scope is exactly what the name suggests: it gates access to the app on the device. Someone who picks up an unlocked phone cannot simply open the authenticator and read codes. That is a real and worthwhile protection, and it is the reason the feature appears in the list at all.
It is not, however, protection for the account at the service. App lock does not change how a sign-in is verified, does not encrypt anything at the provider, and does not substitute for a strong device lock or a screen lock that actually engages. It is a second door on the same house. If the device itself is compromised or left unlocked and the app lock is weak or disabled, the gate provides little.
Nor does app lock say anything about backup completeness. This is the point where the two features in that grouped list get wrongly merged. A locked app can have an incomplete backup. A backup can be complete on an app with no lock enabled. The features are independent, and treating app lock as evidence that your recovery is solid is a category error.
Finally, app lock is not a password manager. The fact data lists password managers and 2FA as a subject in its own right, which is a reminder that the two are being considered separately. A lock on an authenticator protects codes. It does not store, fill, or manage the passwords those codes sit beside.
Password Managers and 2FA: Why One App Cannot Be Assumed to Do Both
The fact data lists “password managers and 2FA” as a function area, and the phrasing is deliberate: it names a relationship, not a merger. The app is an authenticator. The question of how it relates to password management is a question about boundaries, and the boundary is that a second factor and a stored password are different things serving different roles in the same sign-in.
Autofill is the clearest place to draw the line. The fact data does not list autofill as a function at all. So any expectation that the app will fill credentials reliably in Android browsers and apps is unsupported by the supplied facts. It may or may not do something adjacent, but nothing in the data describes it, and a reader should not plan around it.
The standalone-password-manager misconception follows directly. Someone who wants a password manager without a Microsoft account, with no dependency on an authenticator’s ecosystem, is looking for a product the fact data does not describe. This app is listed as an authenticator with push approvals, TOTP, app lock, and backup. That is a different job.
The useful mental model is a split: passwords live in a password manager, and second factors live in an authenticator, and keeping them in separate tools means a single compromise does not hand over both. Whether you consolidate is a decision about convenience and risk, not a feature the fact data grants.
Android Tablets, Wear OS, and Old Phones: Platform Claims the Facts Do Not Cover
The platform field says Android, and that is the entire platform claim. Nothing in the supplied facts describes a tablet layout, a Wear OS companion, or watch-based approval. So the question “how do I use Microsoft Authenticator on a tablet or Wear OS watch?” has to be answered as unknown. The app is listed for Android; anything beyond that is outside the data.
That matters because people plan around form factors. If your workflow assumes approving a sign-in from a watch, or reading a code from a tablet while your phone charges, the facts give you no basis for either assumption. Verify on the device you actually intend to use before you make it part of your recovery plan.
Old Android versions are a separate and more serious boundary. Devices running Android releases that no longer receive current security updates are the wrong audience for any security app, and this one is no exception. The fact data does not list minimum OS versions, so there is no supported floor to point at—but the general principle holds regardless of what the listing says. A second factor running on an unpatched device is a weak link in an otherwise reasonable setup.
The same applies to readers who want a hardware security key instead of a phone app. That is a different product category, and nothing in the fact data positions this app as a substitute for one.
Verdict
Microsoft Authenticator is a free, proprietary Android authenticator from Microsoft Corporation. The fact data supports a specific and fairly narrow description: push approvals with number matching, TOTP codes, passwordless sign-in, app lock, backup and restore, and a place in the broader conversation about password managers and 2FA. Everything in this review traces back to that list, and the list is thinner than the app’s reputation suggests.
The cautious verdict is the right one. This should not be treated as a portable TOTP vault, because cross-app export is not listed and backup is not export. It should not be treated as a full standalone password manager, because password managers and 2FA appear as separate concerns and autofill does not appear at all. And it should not be treated as a universal key to every account you own, because per-service compatibility with standard TOTP is a question each service answers for itself.
Who it fits: Android users who want a no-cost authenticator with push approvals, number matching, and app lock, and who are willing to verify their own services’ compatibility and to keep a separate password manager. Who it does not fit: anyone who needs guaranteed export of every TOTP seed to another app; anyone who requires an open-source, independently audited codebase, which a proprietary license rules out; anyone who wants a standalone password manager without a Microsoft account; anyone who would rather carry a hardware security key; and anyone on an old Android release without current security updates.
If you decide it fits, install it from the official Android store listing and treat backup and restore as a feature to test deliberately rather than a guarantee to assume. That single habit prevents most of the disappointment this app generates.
Pros
- Free to use
- Supports both Microsoft accounts and standard TOTP accounts
- Includes number matching and passwordless sign-in features
Cons
- Proprietary and closed source
- Some advanced features depend on the Microsoft account ecosystem
- Less transparent than open-source authenticator alternatives
The link goes straight to Microsoft Corporation. We do not host app files.
Frequently asked questions
Is Microsoft Authenticator free?
The recorded pricing model is Free.
What license does it use?
It is listed under Proprietary.
Where can I download the official version?
The official link points to https://www.microsoft.com/security/mobile-authenticator-app.