What is multi-factor authentication (MFA)?

Multi-factor authentication checks multiple aspects of a person's identity before allowing them access to an application or database, instead of just checking one. It is much more secure than single-factor authentication.

Learning Objectives

After reading this article you will be able to:

  • Define multi-factor authentication (MFA)
  • Learn why MFA is more secure than using a password alone
  • Explore the various identity authentication factors

Related Content


Want to keep learning?

Subscribe to theNET, Cloudflare's monthly recap of the Internet's most popular insights!

Refer to Cloudflare's Privacy Policy to learn how we collect and process your personal data.

Copy article link

What is MFA (multi-factor authentication)?

Multi-factor authentication, or MFA, is a way to verify user identity that is more secure than the classic username-password combination. MFA usually incorporates a password, but it also incorporates one or two additional authentication factors. Two-factor authentication (2FA) is a type of MFA.

MFA is an important part of identity and access management (IAM), and it is often implemented within single sign-on (SSO) solutions.

What are authentication factors?

Before granting a user access to a software application or a network, identity verification systems assess the user for characteristics that are specific to them in order to make sure they are who they say they are. These characteristics are also known as "authentication factors."

The three most widely used authentication factors are:

  1. Knowledge: something the user knows
  2. Possession: something the user has
  3. Inherent qualities: something the user is

MFA refers to any usage of two or more authentication factors. If only two authentication factors are used, MFA can also be referred to as two-factor authentication or two-step verification. Three-factor authentication is another form of MFA.

What are some real-world examples of the three authentication factors?

  • Knowledge (something the user knows): This factor is a piece of knowledge that only one user should have, like a username and password combination. Other types of knowledge factors include security questions, ID numbers, and Social Security numbers. Even a "secret handshake" may be something a user knows.
  • Possession (something the user has): This factor refers to possession of a physical token, device, or key. The most basic example of this authentication factor is using a physical house key to enter one's home. In a computing context, the physical object could be a key fob, a USB device, or even a smartphone. Many modern MFA systems will text a temporary code to a user's phone and ask them to enter the code in order to access their account. This demonstrates that the user possesses a phone that no one else possesses, helping establish their identity (unless an attacker has hijacked the user's SIM card).
  • Inherent qualities (something the user is): This refers to a physical property of one's body. The most basic version of this authentication factor is the ability to recognize someone by sight or by the sound of their voice. Humans use this ability constantly in their daily interactions. Checking one's appearance against the photo on one's ID card is another example of verifying inherent qualities. In a computing context, one example of this authentication factor is Face ID, a feature offered by many modern smartphones. Other methods may include fingerprint scanning, retina scans, and blood tests.

Why is MFA more secure than single-factor authentication?

Single-factor authentication is the use of just one of the above factors to identify a person. Requiring a username and password combination is the most common example of single-factor authentication.

The problem with single-factor authentication is that an attacker only needs to successfully attack the user in one way in order to impersonate them. If someone steals the user's password, the user's account is compromised. By contrast, if the user implements MFA, an attacker needs more than a password to gain access to the account — for example, they will likely need to steal a physical item from the user as well, which is much more difficult.

This issue also applies to other forms of single-factor authentication. Imagine if banks only required the use of a debit card for withdrawing money — the possession factor — instead of requiring a card plus a PIN. In order to steal money from someone’s account, all a thief would need to do is steal their debit card.

It is important to keep in mind that it is the use of different factors that makes MFA secure, not multiple uses of the same factor.

Suppose one application prompts a user to enter a password only, while another application prompts a user to enter both a password and an answer to a security question. Which application is more secure?

Technically, the answer is neither: both applications are relying on one authentication factor alone, the knowledge factor. An application that requires a password and either a physical token or a fingerprint scan is more secure than an application that only requires a password and some security questions.

Which forms of MFA are the most effective?

This is a highly contextual question. Generally, any form of multi-factor authentication will be much more secure than single-factor authentication.

With that said, certain forms of MFA have been shown to be vulnerable to sophisticated attack methods. In one real-world example, attackers sent employees SMS phishing messages pointing to fake login pages for the organization’s single-sign-on service. If a user entered their username and password into this fake page, the following steps took place:

  1. The attackers used the stolen username and password on the organization's real login page.
  2. The real login page attempted to verify another authentication factor — possession — by sending a temporary code to the real user’s phone.
  3. The attackers redirected the user to another fake page, which asked them to enter the temporary code.
  4. If the user did so, the attackers used that code on the real login page and gained access to the account.

By contrast, another way of verifying possession — a USB security token — would not be susceptible to this particular attack. If all users are given unique security tokens to plug into their computers, and must physically activate those tokens in order to authenticate, attackers in possession of someone’s username and password would not be able to access accounts unless they stole that person’s computer. The same could be said of verifying identity using inherent qualities, e.g. a user’s fingerprint or facial scan.

Does this mean security tokens and fingerprint scans are more secure than one-time passwords? In a phishing context, yes. But organizations should evaluate their specific security risks and needs before selecting an MFA method. And again, any form of MFA is more secure than single-factor authentication, and would represent an important step forward in an organization’s security journey.

Are there any other authentication factors?

Some members of the security industry have proposed or implemented additional authentication factors besides the three main ones listed above. Though rarely implemented, these authentication factors include the following:

Location: Where a user is at time of login. For instance, if a company is based in the US and all its employees work in the U.S., it could assess employee GPS location and reject a login from another country.

Time: When a user logs in, typically in context with their other logins and with their location. If a user appears to log in from one country, then attempts a subsequent login from another country several minutes later, those requests are not likely to be legitimate. A system might also reject login attempts outside of normal business hours — although this is more like a security policy than an identity authentication factor.

If these are both considered to be additional identity factors — which is up for debate — then four-factor authentication and five-factor authentication are technically possible. Both fall under the umbrella of multi-factor authentication.

Implementing such strong security measures must be weighed against the toll this takes on the user, since overly stringent security measures often incentivize users to circumvent official policies.

How can users implement MFA for their accounts?

Many consumer web services offer MFA today. Most applications that do have MFA offer a form of 2FA that requires the user to use their smartphone when logging in. Explore the security settings in each application to see if it is possible to activate 2FA. In addition, Cloudflare allows all Cloudflare users to implement 2FA for their accounts.

How can businesses implement MFA?

Using an SSO solution is a recommended step for implementing MFA. SSO provides a single place for implementing MFA across all apps, whereas not all individual apps will support MFA.

Cloudflare Zero Trust integrates with SSO vendors who support 2FA. Cloudflare helps protect companies' websites and cloud applications by controlling what users are able to do and enforcing security policies for employees whether they work remotely or within controlled office environments.