All Collections
OCR API Platform
API
Document Data Extraction Fields Explained
Document Data Extraction Fields Explained

JSON fields description and comments

Updated over a week ago

There are more than 120 fields Veryfi OCR API supports out of the box.

Please review the high-level description of the fields extracted.

Please note the most up-to-date information is available in your Veryfi account API Docs, which lists fields and parameters formats and limitations.

JSON field

Description

Comments

id

Veryfi Unique ID

id is Veryfi's unique ID for each document. Every time Veryfi processes the invoice, it is assigned a unique ID

img_url

URL for the submitted image

Image URLs are set to expire within 15 minutes, security requirements. The link is refreshed every time you GET this document.

img_thumbnail_url

Download URI for thumbnail image

The difference Between Thumbnail and Image is that a thumbnail is a small version of a larger graphic. By clicking on a thumbnail, it will display a larger image. Some invoices have a thumbnail, some have the whole image

Add custom fields for Regex under Data Transformations > Rules. Create custom rules under “Add Rule” with conditions at both the Document level and Line Item level.

You have the option to apply Regex so you can search for the presence of any static value.

document_type

Type of document

Possible values:

"receipt", "invoice", "purchase_order", "check", "other", "w9", "w2", "statement", "w8"

Type for document. Can be only one of these mentioned in the column on the left. Document types Veryfi works with Part of Veryfi Fraud Framework

document_title

Title on the document

"document_title" is the title on the document usually at the very top of the document: "Invoice", "Vendor Credit", "Purchase Order" etc. This field can be updated, but a validation step is required: if OCR text doesn’t match with the new value - an update is not allowed.

category

Document category/class default list

"Advertising & Marketing" "Automotive","Bank Charges & Fees "Legal & Professional, Services","Insurance", "Meals & Entertainment",

"Office Supplies & Software", "Taxes & Licenses","Travel",

"Automotive", "Rent & Lease","Repairs & Maintenance", "Payroll","Utilities","Job Supplies", "Grocery"

The default list of categories user account is been created. If you would like to use your own list and it’s different from the default one feel free to send a list of categories along with your API call, follow these instructions In API Documentation

tags

Document can have multiple tags, Tag is a JSON response attribute that is fully controlled by user. With Tags you can separate document by "campaign" "client" or any other. Create your first Tags here . Read more about Tags

default_category

Category from Veryfi default set of categories (being added if categories were provided by the user)

date

Document issue/transaction date

Date when the invoice/Receipt/W-2/W-9/Check was issued

due_date

Due date for invoice or bill

order_date

Order date

Date when the goods/services were ordered

delivery_date

Delivery date

Date when the order was/will be delivered.

delivery_note_number

Delivery Note Number

ship_date

Shipping date

Date when the order was/will be shipped

created

Creation date of the document (UTC)

Date when a document was sent to process

updated

Last update date (UTC)

Date with time when the last update was made to the document

currency_code

Currency code (ISO 4217)

payment_card_number

Credit/debit card number

Last 4 digits of CC number which the invoice has been paid by

payment_display_name

<payment_type> +<card_number>
e.g "American Express ***1234"

payment_type

One of the payment types
"visa", "master_card", "cash", "american_express", "other", "check", "discover", "bancomat", "interac", "girocard", "paypal", "bank-to-bank_transfer_bacs", "octopus", "jcb", "bankaxept", "alipay", "waon", "unionpay", "wechat", "bpay", "rupay"

Payment type of CC (visa, mastercard, etc)

payment_terms

Payment terms

NET if found on the document

tracking_number

Tracking number

A unique number assigned to packages when they are shipped.

subtotal

Subtotal amount

We try to find it on all the documents Veryfi support. If we do not find it, we calculate it as Subtotal = Total Spend - Tax - Tip
Computable

tax

Total tax amount

rounding

Rounding amount

Some vendors give a "discount" (rounding) so they do not have to pay with small monets (for ex, if a customer paid with cash and vendor did not have a change of 4 cent, he would round the number and return 5 cent)

cashback

Cashback amount

Cash amount a customer has withdrawn when making a purchase. Can be found on receipts, not present on invoices

tax_lines

Tax breakdown lines

Tax line (order, name, rate, total amount) for each tax.

order

Tax line sequence number

Order (position) of the tax line in the list of taxes

base

Tax base

name

Tax name

code

Tax code

rate

Tax rate

Tax percentage (10%, 9.95% etc.) Computable

total

Tax amount

Total Vat tax, tax, iva, they all will be extracted into this “tax” field. If the document has multiple taxes on it those taxes will be returned in the list inside “taxes” field. Note there are a couple of countries in the world that have 3 decimal places after the dot. If planning to support those, when choosing type for tax/total fields make sure it can handle 3 decimal places.

total_inclusive

Amount including tax on the tax line (tax base * tax rate)

tip

Tip amount

An amount of money that is given to someone for a service, also called Gratuity. Usually present on receipts not invoices. Computable

discount

Discount amount

shipping

Shipping amount

Cost of Shipping / Delivery

total

Document total amount

Grand Total amount including subtotal, tax, fees etc Computable

service_start_date

Start date of service, e.g. flight departure date or hotel arrival date

Typical for Hotel Folios

service_end_date

End date of service, e.g. flight arrival date or hotel departure date

Typical for Hotel Folios

invoice_number

Invoice number

Typical for Invoices, for Receipt we can return TR# or

Insurance

Insurance

Field specifies insurance terms for Invoice

Incoterms

Incoterms

Incoterms specify who is responsible for paying for and managing the shipment, insurance, documentation, customs clearance, and other logistical activities.

balance

Balance if found on the document

previous_balance

Balance before this the last invoice

final_balance

Final balance if found on the document

purchase_order_number

Purchase order number

Purchase order is A purchase order is a document given from a buyer to a seller that details the quantity, prices, and total cost of requested goods and services.

document_reference_number

Reference number

Reference numbers are commonly used to identify items for a particular customer

bill_to_name

Bill to name

Name of the person or business who is billed for the amount on the invoice

bill_to_address

Bill to address

Billing Address information of the customer. Veryfi can return parsed address for this field

bill_to_vat_number

Bill to VAT number

The customer's VAT number, usually present on the European invoices

bill_to__reg_number

Bill to REG number

The customer's REG number, usually present on the European invoices

bill_to_email

Bill to email address

Billing email address of the customer

bill_to_phone_number

Bill to phone number

Billing phone number of the customer

parsed address for bill_to

increments the address

ship_to_name

Ship to name

Name of the person or business who will accept the delivery of this order

ship_to_address

Ship to address

Address where the order has to be delivered.

parsed address for ship_to

increments the address

vendor

Vendor details

Vendor details

vendor_abn_number

ABN

Australian business number. It is only for Australian invoices.

vendor_vat_number

Vendor Vat number

vendor_name

Vendor name - cleaned

The name is an actual vendor name after cleaning and transformation, for example, "fb.com" would be converted to "Facebook, Inc."

vendor_raw_name

Vendor name - raw

The Raw version is what appears in the text whereas the enriched version is where we run post-processing and verifying if that vendor name and address actually exists or try to give more context based on the raw version. For example, if in a receipt we find “Walmart” with some address to be the vendor name, the enriched version could be “Walmart Supercenter” or “Walmart Neighborhood” etc… depending on the context of the receipt. Please note, if vendor_name was taken from VAT or ABN or other business registration number or any other online sources it is expected that vendor_raw_name to "vendor.raw_name": null

vendor_address

Vendor address

vendor.raw_address

Vendor address -raw

Same as vendor_raw_name, Please note, if vendor was taken from VAT or ABN or other business registration number or any other online sources it is expected that vendor_raw_name to "vendor.raw_address": null

vendor_lat

Vendor latitude

Latitude of the vendor location. It is extracted by Veryfi AI technology

vendor_lng

Vendor longitude

Longitude of the vendor's location. It is extracted by Veryfi AI technology

vendor_phone_number

Vendor phone number

Phone number of the vendor.

vendor_fax_number

Vendor fax number

Fax number of the vendor.

vendor_web

Vendor website URI

Website address of the vendor

vendor_email

Vendor email address

Vendor email address

vendor_logo

Vendor logo image download URI

Graphic mark, emblem, or symbol used to aid and promote vendor's recognition.

vendor_category

Vendor Category

"Grocery", "Fuel", "Restaurant", "Hardware", "Online Shopping", "Fast Food", "Department Store", "Convenience", "Taxi", "Drugstore / Pharmacy", "Hotel", "Nurseries & Gardening", "Food", "Coffee", "Utilities", "Bakery", "Parking", "Bar", "Electronics Store", "Building Supplies", "Office Equipment", "Health", "Clothing Store", "Online Service", "Furniture Store", "Auto Parts", "General Contractor", "Pizza", "Car Repair", "Bank", "Shipping", "Arts & Crafts Store", "Airlines", "Cafe", "Home Goods Store", "Transportation", "Car Wash", "Insurance", "Car Rental", "Toll", "Shoe Store", "Book Store", "Auto", "Movie Theater", "ATM", "Laundry", "Airport", "Entertainment", "Hair Care", "Ice Cream Shop", "Market", "Rent", "Florist", "Farm", "Gym / Fitness Center", "Brewery", "Food Court", "Garage", "Catering", "Baggage", "Airport Parking", "Daycare", "Delivery"

vendor_reg_number

Vendor registration number

The vendor registration number, for the US it would be EIN (Employer Identification Number) Some countries require Vendor ID to be listed on every invoice

vendor_bank_number

Vendor bank number

Routing

vendor_bank_numbers

Other vendor bank numbers, if found on the document

vendor_bank_swift

Vendor bank SWIFT code

Society for Worldwide Interbank Financial Telecommunications, part of the Invoice Remittance information

vendor_bank_address

Vendor bank address

Bank Address

vendor_account_number

Vendor bank account number

Bank account number of the Vendor, part of the Invoice Remittance information. Read more about Bank details fields.

vendor_account_currency

Vendor Account Currency. Refer to the list of currencies Veryfi supports

vendor_bank_name

Vendor bank name

Bank name, part of the Invoice Remittance information

vendor_iban

Vendor IBAN

International Bank Account Number, part of the Invoice Remittance information

external_ids

External Ids for Vendor

vendors

Vendors list found on the document

line_items

Line items

Line items is the list of products or services purchased/ordered

line_items_id

Veryfi Unique Line Item ID

Veryfi's internal unique ID for this product/service

line_items_order

Line item sequence number

Order of the product in the list of line items (products purchased) Not related to PO

line_items_category

Line item category/class

Veryfi uses the broad list of categories but can be a custom list as well. Please talk to Veryfi support if line_item_category is important for you.

line_items_type

Line item type

"room", "tax", "parking", "service", "fee", "delivery", "product", "food", "alcohol", "tobacco", "transportation", "fuel", "discount", "payment", "refund"

Veryfi categorizes each line into one of those categories (types) listed in column "C"

line_items_tags

Line item Tag

Users can apply a "Tag" to both line items or whole document. More in API Documentation

line_items_section

Line Item section

Section that your purchase belongs to, extracted if found on the receipt. e.g "grocery", "baked goods", "liquor", "miscellaneous" Extremely important for CPG/DTC use cases

line_items_date

Line item date (often exists on hotel folios or utilities bills)

line items of service invoices usually contain the date of the service received. This date is extracted into its own field.

line_items_start_date

Line item start date

Often exists on hotel folios

line_items_end_date

Line item end date

Often exists on hotel folios

line_items_description

Line item description

Description of the product or service purchased on the invoice line excluding date, price

line_items_sku

Line item SKU

Stock Keeping Unit, a unique number associated with a product

line_items_upc

UPC

Universal Product Code

line_items_hsn

HSN

HSN code is a 6-digit uniform code that classifies 5000+ products and is accepted worldwide

line_items_discount_rate

Discount Rate. Typical for CPG/DTC use cases

line_items_price

Line item unit price

line_items_quantity

Line item quantity

Quantity for a line item. Computable

line_items_reference

Line item reference

Reference Number

line_items_unit_of_measure

Line item unit of measure

line_items_weight

Line item weight

Weight of the material product on the line item, usually present on logistic invoices

line_items_text

Line item text

Full text of an extracted line item including prices, dates, etc.

line_items_tax_rate

Line item tax rate

What tax rate is applicable for this product. For example some products may have 5% tax rate and some may be 15%

line_items_tax

Line item tax name

line_items_tax_code

Line item tax code

line_items_discount

Line item discount amount

Line item discount

line_item.subtotal

Line item subtotal amount

New subtotal for line item

line_items_total

Line item total amount

Line_items_country_of_origin

Code for Country of Origin, full list available in: API Schema

Line_items_batch/lot number

Lot number of a product if found on the document

Line_items_manufacturer

Manufacturer

Line_items_custom_fields

Custom fields for line item level

Add custom fields for Regex under Data Transformations > Rules. Create custom rules under “Add Rule” with conditions at both the Document level and Line Item level.

You have the option to apply Regex so you can search for the presence of any static value.

account_number

Account number

Customer account number. If it is on the document, we extract it.

ocr_text

OCR text extracted from the submitted document

Full text extracted from the document

total_quantity

Total quantity of if mentioned on the document. Total quantity found on a document. Should be equal to the sum of line item quantities, but depends on the vendor.

total_weight

Total weight of all the items listed on the invoice

total_weight_list

List of all the weights (net, gross) on the invoice

Some invoices have both weights of the items ordered Gross and Net, in this case, this list will contain both of the weight

Barcodes

A Barcode is a method of representing data in a visual, machine-readable form.

data

Extracted data from the Barcode

type

QR + Barcodes, PDF417, EAN, UPC, Code128, Code39, I25

exch_rate

Exchange rate

Exchange rate

weights

All weights found on the document

license_plate_number

License plate number

vin_number

VIN number

A vehicle identification number (VIN) is a unique code assigned to every motor vehicle when it's manufactured. The VIN is a 17-character string of letters/

additional_charges

Additional charges

Additional fees and charges included on receipts or invoices (R&D)

summary

Summaries

Summary charges on receipts or invoices (R&D)

status

Document Status

Status: in_progress┃processed┃reviewed┃reimbursed┃archived

meta

Document Meta

total_pages

Total sent pages

owner

Owner of the document

processed_pages

Total processed pages. please look at max_pages_to_process guidelines

meta fraud

is_lcd

LCD Detection

is_lcd field - LCD (Liquid Crystal Display) detection in images refers to the process of identifying whether an image is a screenshot of an LCD display. This task is performed to detect forged or manipulated images, especially in scenarios where screenshots are used to deceive or misrepresent information. Part of Veryfi Fraud Framework

warnings

Data Reconciliation

Field called warnings is part of the meta object. The idea is that some of those signals can help our clients to detect fraud. Part of Veryfi Fraud Framework

Example :

  • Tax + Line items = Total

  • Repeating line items (SKUs or Descriptions)

  • Total Items sold != SUM of Line Items

handwritten_fields

Handwritten Detection

Service in the data extraction process that looks at the selected fields and its areas and tries to identify whether there are any traces of handwriting. Veryfi Machine learning algorithms are trained on a labeled dataset of handwritten samples. The algorithms learned to recognize different handwriting styles and characters based on the extracted features.

Example: User has added a new line item by hand and/or modified the total by adding a 0 to the total to exponentially increase their spend claim. Many use cases would want to flag such receipt (main application is expense management and cpg loyalty campaigns)
Part of Veryfi Fraud Framework

color

Color indicator where green is no fraud while yellow is fraud. How confident the model is about color prediction

Allowed: green┃yellow
Part of Veryfi Fraud Framework

updated_date

The date when the document was updated with counts PUT API call or any update vie web portal (if)

guest_count

Number of guests

Hotel/restaurant’s number of guests/seats

store_number

Store Number

server_name

Server Name

Restaurant’s server name

vending_person

notes

Field users can add notes with Rules or regular expressions

is_approved

is_document

Document Classifier

Document Classifier. Helps you identify what type of document was captured into memory.
Use is_document and document_type keys to base your decision on whether you trust this submission.

is_money_in

True if refund or credit note

is_duplicate

Duplicate detection

Existing duplicate detection looks at many fields and assets.
Duplicate detection can be customizable per your specific use case. If a document has been processed before, Veryfi will identify it and notify. Duplicate detection is a part of Veryfi Fraud Framework.

duplicate_of

This field provides "id" for the document that might be original.

flags blurry images

We recommend checking API Docs for the most recent API Schema.

Additional request parameters

JSON Parameter

Comment

confidence score for extracted values

a rectangle that surrounds an extracted value, that specifies its position for value

Flag that tells Veryfi whether boost mode should be enabled. When set to 1, Veryfi will skip data enrichment steps, but will process the document faster. Default value for this flag is 0.

parameter to breakdown vendor, bill_to, ship_to address into pieces

asynchronous processing/ webhooks

Limit processing to a number of pages. A page is a pdf page or an image. When sending a long document to Veryfi for processing, this parameter let you control how many pages of the document are to be read and processed. Max pages to process 15.

external_id

Use this if you would like to assign your own ID to documents. one of use cases is to separate clients in one environment, read more here.

Whether to use additional calculation e.g. for line_items.quantity

List of categories, if you would like to use your own

List of tags

country

New A special hint for the Veryfi data extraction model to make more accurate predictions regarding currency. Please note this is different from the currency preferences settings that are available in your Veryfi account. A suitable use case is having international users. country codes are according to ISO 3166-1 alpha-2

Please note, that not all the fields from the list above are enabled for user accounts from the start. If you would like us to add more fields or parameters to your account, please contact support@veryfi.com

Did this answer your question?