Token Management

About tokens

Tokens are virtual representations of issued cards that are created when a cardholder:

• Adds a card to a digital wallet like Apple Pay or Google Pay

• Saves a payment method for their account at an online storefront or intermediary payment method

Customers can use tokens for payment, and they don’t expose sensitive card information every time they use them. As such, they function as substitutes for card details including the number, expiration date, and verification code, reducing the risk of stolen card information because of a transaction or a fraudulent actor. Because they don’t expose sensitive card information, tokens are generally considered a more secure form of payment than physical cards, or the manual entry of card details into a checkout form.

Tokens are most relevant for users that:

• Allow their cardholders to spend using Apple Pay, Google Pay, or Samsung Pay
• Have significant issuing volume on Card Not Present transactions (like online purchases)
• Want to incorporate token behavior into their business logic

Token management

Stripe Issuing allows you to view and manage details associated with all tokens issued through your program with the Tokens API. We provide these details to help you better understand the characteristics of your tokens and how they’re being used. For example, you can find details on the following key characteristics:

• The originator of a token: Whether a digital wallet provider or business is requesting the token.
• The predicted riskiness of a token: The card network’s assessment of risk and its recommendation for a given token.
• The device associated with a token: Whether a watch, phone, or other device is requesting the token, and the assessed risk of that device.
• The cardholder characteristics of the token originator: Whether the cardholder values such as the name and address match those of the cardholder in Stripe for additional verification.

In addition to visibility into token characteristics, the Tokens API allows you to activate, suspend, or deactivate tokens based on your desired workflows. For example, you might opt to do so to:

• Verify which existing tokens migrate to a new card when your program replaces a card.
• Deactivate tokens suspected of fraudulent activity without impacting the underlying card.

The issuing.authorization object also has a token field that populates if it uses a token.

Risk controls

When Stripe receives a tokenization request, we assess a variety of variables to determine whether or not we should approve the request. As a result of this assessment, we choose one of the following:

• Approve the tokenization request, creating the token and adding it to a digital wallet.
• Require additional authentication, triggering a one-time passcode flow through the digital wallet provider. After the authentication successfully completes, we create the token and add it to the corresponding digital wallet. In these scenarios, the status field is populated as requested until the additional authentication completes.
• Reject the tokenization request, preventing the token from being created.

Users also have the ability to put their own additional risk controls in place on top of Stripe’s. The Tokens API doesn’t allow for the outright rejection of a tokenization request at the point of creation, but users can deactivate or suspend tokens shortly after creation.

How tokens work

Understand how tokens work to give yourself a picture of token creation and the token lifecycle.

Create a token

Token creation, or tokenization, is a multi-step process involving cardholders, users, a digital wallet provider, Stripe, and a card network. The example scenario below shows the steps the cardholder must complete and the processes involved when using the Tokens API as part of your program.

Example scenario

A cardholder on your Issuing program wants to add their Stripe-issued card to a digital wallet, such as Apple Pay. To do so, they open their digital wallet app and complete prompts to enter their cardholder information (such as name and billing address), and card information (such as card number and expiration date).

This information is then submitted to the wallet provider, (in this case, Apple Pay) which is registered with the card’s underlying network (for example, Visa or Mastercard) as a ‘token requestor’ within the network. Next, the card network conducts a series of validations against this data, combines it with their own data into a tokenization request, and forwards it to Stripe to decide. Stripe conducts its own additional validation to determine how to proceed with the request. As noted earlier on this page, this validation step can result in three outcomes:

• Stripe approves the tokenization, which activates the token in the wallet, making it ready to use. Stripe sends out the issuing_token.created event to any listening webhook endpoints.
• Stripe requires additional verification, which prompts an authentication challenge to the cardholder. Stripe sends out the issuing_token.created event to any listening webhook endpoints. The token becomes active when the cardholder successfully completes this step. Stripe sends out the issuing_token.updated event to any listening webhook endpoints as soon as the token becomes activated.
• Stripe declines the tokenization request, which prevents the token from being added to the wallet. Stripe doesn’t send out the issuing_token.created event to any listening webhook endpoints.

The wallet provider or the card network can halt a tokenization request from proceeding further at any step in the tokenization process—Stripe doesn’t always receive notification when this happens.

The sequence diagram below helps to further illustrate the tokenization process.

Token lifecycle

After a token is created, it can exist in four distinct states in the digital wallet:

• Inactive: The token request is outstanding, and the token can’t be used for authorizations yet. An inactive token in the Tokens API has a status of requested.
• Suspended: The token is temporarily unavailable to use in the wallet. A cardholder or a Stripe user using the Tokens API can trigger a token suspension. Cardholders can’t undo suspensions by a Stripe user (that is, through a digital wallet app). Users can only reactivate suspended cards directly through the Tokens API.
• Active: The token is available for use in the wallet it’s been added to.
• Deleted: The token has been removed from the wallet and you can no longer use it. You can’t modify tokens in this state.

The below state diagram helps show the different states, how they’re reflected in the API, and how you can use our APIs to modify them.

Stripe automatically synchronizes token status with the cardholder states and card states when they change. Stripe also migrates tokens between card replacements if the original card isn’t canceled first. To see which tokens are associated with which card, use the List API.

Merchant tokens

Businesses can also create tokens when saving a payment method for future use at a retailer (for example, a cardholder saving their card details for checkout at Amazon). In these scenarios, the business originates the token creation, and the Tokens API won’t have the wallet_provider field. To assess the underlying business that originated the token, we recommend examining the business details associated with authorizations conducted using the token. If you issue cards from MasterCard, tokens that originate from them might populate a readable name in the network_data.mastercard.token_requestor_name field.

Merchant tokens are tied to the specific business (token requestor in the card network) that originates them and can’t be used at other businesses.

Identify when a token was used for a transaction

Authorizations or Transactions that used a token have an expandable reference to the Token object in the token attribute. This field is null for Authorizations or Transactions that didn’t use a token. Combine this with the wallet attribute on the Authorization or Transaction object, or the wallet_provider attribute on the Token object, to determine whether a digital wallet token was used.

See the Authorizations and Transactions API documentation for more details.

Network data restrictions

The Issuing token object contains an optional, expandable field called network_data. It contains additional, sensitive card network details about your tokens, primarily related to the token creation process. Because this data is highly sensitive, you must have a restricted access key with the necessary permissions to access the data, and can only view the data on a token within the first 24 hours after token creation (based on the created value). This data is only available in the API to retrieve a token and the API to update a token status.

To access this data, configure your restricted access keys with the following permissions:

• Issuing tokens read access for Retrieve and List methods
• Issuing tokens write access for the Update Status method
• Issuing token network data read access for accessing network_data within the 24 hour time limit

If you need access to network_data beyond the initial 24-hour period that it’s available for, you must limit the IP addresses from which your restricted access keys will use.

Testing

An Issuer’s cardholder can create tokens for free at storefronts or in digital wallets. We recommend creating one in a digital wallet of your choice to understand the webhook events, API fields, and effects of updating a token. To do so, follow the Digital Wallets guide for manual provisioning first.

If you prefer to test Tokens API in test mode, you can create a test mode authorization with the wallet field set to one of the available choices. The token field is set on the resulting authorization. You can then use the API methods on this token as normal. In these scenarios, not all fields are set, including network_data, and this token isn’t used for subsequent test authorizations.

The Tokens API

Token data is only accessible through the Tokens API. Below are a few examples applications.

Verify a successful manual provisioning example

• In this example, you subscribe to the issuing_token.created and issuing_token.updated events.
• When you receive an issuing_token.created event, use the Retrieve API and expand network_data to look at provisioning details. Here’s an example:

{
  "id": "evt_1NxBn3FUQNp5XJkna0rkKU2r",
  "object": "event",
  "api_version": "2024-11-20.acacia",
  "created": 1691100189,
  "data": {
    "object": {
      "id": "intok_1NuMIZFUQNp5XJknPmDzEz0t",
      "object": "issuing.token",
      "card": "ic_1JDmgz2eZvKYlo2CRXlTsXj6",
      "created": 1691100179,
      "device_fingerprint": "intd_1JDmgz2OpvKigH2CxnEEs",
      "last4": "9203",
      "livemode": true,
      "network": "mastercard",
      "network_updated_at": 1691100170,
      "status": "requested",
      "wallet_provider": "apple_pay"
    }
  },
  "livemode": true,
  "pending_webhooks": 0,
  "request": {
    "id": "req_ARTvFhTufhHna9",
    "idempotency_key": "49a40678-8f45-4c91-9d6f-98a5bd569f9d"
  },
  "type": "issuing_token.created"
}

• Check that the wallet_provider field is populated, which tells you that it originates from a digital wallet, and take note of the object’s id. Use that in the Retrieve API call:

curl https://api.stripe.com/v1/issuing/tokens/intok_1NuMIZFUQNp5XJknPmDzEz0t \
 -u 
sk_test_4eC39HqLyjWDarjtT1zdp7dc
: \
 -d "expand[]"=network_data \
 -G

This yields the following response:

{
  "id": "intok_1NuMIZFUQNp5XJknPmDzEz0t",
  "object": "issuing.token",
  "card": "ic_1JDmgz2eZvKYlo2CRXlTsXj6",
  "created": 1691100159,
  "device_fingerprint": "intd_1JDmgz2OpvKigH2CxnEEs",
  "last4": "9203",
  "livemode": true,
  "network": "mastercard",
  "network_data": {
    "device": {
      "device_fingerprint": "intd_1JDmgz2OpvKigH2CxnEEs",
      "ip_address": null,
      "location": "+30.22/-89.10",
      "name": "AB's phone",
      "phone_number": null,
      "type": "phone"
    },
    "mastercard": {
      "card_reference_id": "...",
      "token_reference_id": "...",
      "token_requestor_id": "...",
      "token_requestor_name": "APPLE PAY"
    },
    "type": "mastercard",
    "wallet_provider": {
      "account_id": null,
      "account_trust_score": null,
      "card_number_source": "manual",
      "cardholder_address": null,
      "cardholder_name": null,
      "device_trust_score": null,
      "hashed_account_email_address": null,
      "reason_codes": [],
      "suggested_decision": null,
      "suggested_decision_version": null
    }
  },
  "network_updated_at": 1691100170,
  "status": "requested",
  "wallet_provider": "apple_pay"
}

• In the example, card_number_source is manual, token status is requested, and it’s an Apple Pay wallet. This means the cardholder had the card details with them when they put the card in their Apple Wallet, and they need to complete additional verification before they can use the card in the wallet.
• A few seconds later, you can see an issuing_token.updated event for the same token. The token status is now in the active state. This means the cardholder successfully completed the verification and can use their card for Apple Pay.

Delete a suspicious merchant token example

• In this example, you subscribe to the issuing_token.created event.
• The webhook receives an issuing_token.created event.

{
  "object": {
    "id": "intok_1NuMIZuTQ2hhXJooNmDzEz0t",
    "object": "issuing.token",
    "card": "ic_1JDmgz2eZvKYlo2CRXlTsXj6",
    "created": 1691100179,
    "device_fingerprint": null,
    "last4": "9203",
    "livemode": true,
    "network": "visa",
    "network_updated_at": 1691100170,
    "status": "active"
  }
}

• The token has no wallet_provider field, so it’s a merchant token. Use the Retrieve API and expand network_data to look at provisioning details.

curl https://api.stripe.com/v1/issuing/tokens/intok_1NuMIZFUQNp5XJknPmDzEz0t \
 -u 
sk_test_4eC39HqLyjWDarjtT1zdp7dc
: \
 -d "expand[]"=network_data \
 -G

This yields the response:

{
  "id": "intok_1NuMIZFUQNp5XJknPmDzEz0t",
  "object": "issuing.token",
  "card": "ic_1JDmgz2eZvKYlo2CRXlTsXj6",
  "created": 1691100186,
  "device_fingerprint": null,
  "last4": "4674",
  "livemode": true,
  "network": "visa",
  "network_data": {
    "visa": {
      // ...other fields
    },
    "type": "visa",
    "wallet_provider": {
      "card_number_source": "manual",
      "cardholder_address": null,
      "cardholder_name": "abc",
      // ...other fields
    }
  },
  "network_updated_at": 1691100170,
  "status": "active",
}

• You can see that the cardholder name is an invalid value that doesn’t match the expected cardholder name.
• To avoid any fraudulent activity, use the Update Status API to delete the token before it can be used. Then follow up with the cardholder to see if the token was actually requested by them. If they didn’t request it, cancel and replace the card if the number was stolen, or the account compromised.

curl https://api.stripe.com/v1/issuing/tokens/intok_1NuMIZFUQNp5XJknPmDzEz0t \
 -u 
sk_test_4eC39HqLyjWDarjtT1zdp7dc
: \
 -d status=deleted

Device monitoring example

As with the previous examples, subscribe to the token events, and perform a Retrieve API request on the ID when you receive an event. In this case, you see that a device_fingerprint is populated, and check the network_data.device.location field. You see that the device was provisioned in a different country using the location coordinates. You see that you received prior notice that this cardholder was traveling abroad, and that it matches the country that they specified they were traveling to.