Data Integration Guide - Card Transactions

 

Transaction Data for Analysis

Transaction data fields, as well as customer profile data fields associated with the transaction, is an important driver of DataVisor’s ability to identify transaction fraud. Sensitive customer PII information is not required for our ability to detect. This document serves as a general integration guideline for identifying transaction fraud. Our team will work with the specific client on more detailed integration options.

Transaction Data

Fields that are helpful for transaction data include:

Category	Data Field	Description	Field Type	Required
Transaction Information	Transaction ID	Unique identifier of the transaction	String	X
	Customer ID	Unique identifier of the customer who performs the transaction	String	X
	Transaction type	Type of the transaction (e.g. CNP, card swipe, Apple Pay)	String	X
	Transaction time	Time of the transaction, including time zone	String	X
	Amount	Transaction amount	Numeric	X
	Currency	Transaction currency	String	X
	...	Any Additional Transaction Info
Merchant Information	Merchant Name	Name of the merchant who’s customer performs the transaction	String	X
	Merchant Category	What category this Merchant comes under	String	X
	Merchant Location	Location of the Merchant	String	X
	Merchant URL	URL of the merchant who’s customer performs the transaction	String
	MCC1	Merchant Category Code 1	String or Numeric
	MCC2	Merchant Category Code 2	String or Numeric
	MCC3	Merchant Category Code 3	String or Numeric
	...	Any Additional Merchant Information
Payment Information	Card Number	Card Number (Hashed is ok, with SHA-256 hashing)	String	X
	Card Last 4	Last 4 digits of the Card number (If card number is hashed)	String	X
	Card Name	Name on the Card (Hashed is ok, with SHA-256 hashing)	String	X
	Card Expiration	Expiration month of the card (MM/YYYY)	String	X
	BIN	BIN lookup value	String	X
	Billing Address Line 1	Billing Address provided by customer	String	X
	Billing Address Line 2	Billing Address 2 provided by customer	String
	Billing City	Billing City provided by customer	String	X
	Billing State	Billing State provided by customer	String	X
	Billing Country	Billing Country provided by customer	String	X
	Billing Zip	Billing Zip provided by customer	String	X
	...	Any Additional Payment Information
POS Terminal Information	POS Entry Mode	How the payment was initiated (i.e. swipe, insert, NFC, etc.)	String
	POS Terminal ID	Terminal ID for the POS system being used	String	X
	POS Device Type	POS device type	String
	POS Location	Location of the POS device	String
	...	Any Additional POS Terminal Information
Digital Information	Email	Email input by customer	String	X
	IP address	IP address associated with the transaction	String	X
	Phone	Phone number associated with the transaction	String	X
	Device type & version	Make and model of the device (e.g., Mac OS 10, iPhone 8)	String	X
	Device ID	The device fingerprint identifier for the account applicant’s device	String	X
	Operating system	The operating system and version of the device	String
	User agent	Incoming raw user agent string from the browser	String	X
	Browser cookie	The cookie of the browser	String
	Device name	The device name the account holder assigned to his/her device	String
	...	Any Additional Digital Information
Order Information	Order ID		String
	Order Item ID(s)		String
	Order Items category	Electronics, gift cards, resellable items, …	String
	Shipping Address Line 1	Shipping Address Line 1	String
	Shipping Address Line 2	Shipping Address Line 2	String
	Shipping Address City	Shipping Address City	String
	Shipping Address Zip Code	Shipping Address Zip Code	String
	Shipping Address State	Shipping Address State	String
	Shipping Address Country	Shipping Address Country	String
	Recipient Name	Recipient Name	String
	Recipient Phone	Recipient Phone	String
	Recipient Email	Recipient Email	String
	Delivery Instruction	Delivery Instruction e.g. “gate enter code 123”, “call me at XXX”	String
	..	Additional info about the order, the recipient, or the delivery details
Account Information	Account Number	Account Number held at Financial Institution	String	X
	Account Open Date	Date the account was opened at the Financial Institution	String	X
	Account Close Date	Date the account was closed at the Financial Institution	String
	Account Balance	Current Balance	Numeric
	Product Type	Determining whether the account is a checking, savings, or other	String
	Authentication Method	Any methods used to verify the account	String
	Account Status	Determining if the account is restricted or unrestricted	String	X
Authorization Results	AVS Response Code	Response code received through Address Verification Services for Card Not Present scenario	String	X
	CVV Response Code	Response code for Card Verification Value	String	X
	ECI Response Code	2 or 3 digit Electronic Commerce Indicator (ECI) code returned by the issuing bank and credit card specific networks, if 3DS is implemented	String	X
	Gateway/Acquirer Response	Response code returned by the payment gateway Approved/Declined, Decline reason.	String	X
	...	Any Additional Authorization Results

In a standard integration with DataVisor system, transaction events should be sent to DataVisor Detection API. Please refer to DataVisor API Integration Guide for sending data events to DataVisor real time APIs and what API responses would be returned by the APIs.

Transaction Feedback Data

Fields that are helpful for transaction feedback data include:

Category	Data Field	Description	Field Type	Required
Transaction Information	Transaction ID	Unique identifier of the transaction	String	X
	Customer ID	Unique identifier of the customer who performs the transaction	String
	Transaction type	Type of the transaction (e.g. CNP, card swipe, Apple Pay)	String
	Transaction time	Time of the transaction, including time zone	String
Transaction Outcome*	Transaction outcome	Merchant decision for Approved, or declined.	String	X
Label  Feedback**	Chargeback label	Whether the transaction is reported as chargeback	String	X
	Other fraud label	Whether the transaction is confirmed as fraudulent transaction for other reasons	String	X

*Transaction outcome should come in as a separate event to DataVisor Update API, if the auto decisioning logic is not implemented in DV platform.

** Label feedback should come in as a separate event to DataVisor Update API, at the time when the chargeback or fraud report is available. 

Please refer to DataVisor API Integration Guide for sending data events to DataVisor real time APIs and what API responses would be returned by the APIs.

Customer Profile Data

Customer profile data is also important for fraud detection, when merging with transaction level data. Fields that are helpful for customer profile data include:

Category	Data Field	Description	Field Type	Required
Customer Profile Data	Customer ID	Unique identifier of the customer who performs the transaction	String	X
	Registration time	The time the user first registered in the platform	String	X
	Product type	The type of product customer has	String	X
	Customer tenure	Tenure of the customer	Numeric
	Email	Email address of the customer	String	X
	Phone	Phone number of the customer, or phone prefix	String
	Full Name	The full name of the customer	String	X
	Address	The address of the customer	String
	Country	The country of the customer	String	X
	City	The city of the customer	String	X
	State	The state of the customer	String	X
	Zip Code	The zip code of the customer	String
	Year of Birth	Year of birth of the customer	String

An example of customer profile data:

Customer ID	Name	Customer Tenure	Product Type	Email	Registration Time
53277897	Alice Jane	14 months	Premier Credit Card	[e-mail]	2014-01-01 12:03:03
40098005	John Doe	60 months	Premier Credit Card	[e-mail]	2017-01-02 12:00:00

In a standard integration with DataVisor system, customer profile and profile update events should be sent to DataVisor Detection API. Please refer to DataVisor API Integration Guide for sending data events to DataVisor real time APIs and what API responses would be returned by the APIs.

Data Structure Report

DataVisor needs a historical data set (3-6 months) to train and warm up the detection models. Before sending the full historical data set over, please send over a data structure report to your Technical Account Manager. This process allows DataVisor to ensure a faster turnaround time and a higher quality of results during the fraud assessment. This data structure report should contain the following for each data set:

• Schema of each the data event types
• Preview of 10 rows for each of the data event types
• Earliest date and latest date in the full historical data set
• Number of rows for the full historical data set

Data Anonymization

DataVisor’s algorithm is able to work with anonymized data fields if required. For sensitive PII information, such as name and address, anonymized information can be processed by our algorithm as long as the data structure remains.