Veryfi AI is trained to extract out-of-the-box a wealth of fields from documents like invoices, bills and receipts in seconds using machines end-to-end. For you, this means you will be up and running much faster than alternative solutions on the market. After you signup for the Veryfi API self-serve portal with just your email address, you are setup immediately with API keys, docs etc to go go go!
Veryfi processing runs these 3 steps:
1. OCR your document (common function),
2. Data Extraction using a pre-trained machine model to give context to your document and turn it into structured data and then finally
3. Data Transformation categorizing the document, checking for duplicates and executing post-processing rules you may have setup.
List updated January 2021
To see a list of fields and their key names available in your Veryfi account, login to Veryfi API Portal and under the “Inbox” drop a document for data extraction, then open it up and press the JSON tab for a breakdown of fields (as shown below).
Veryfi API will Categorize this document using default list of Tax Categories. eg. Travel. This list can be altered inside your API Portal account or the list of categories added during an API call for data extraction. See API Portal documentation for detail.
Veryfi API will identify a possible duplicate document that may have been previous uploaded. This helps reduce Accounting Errors and fraud in CPG loyalty programs. The logic behind this is simple (ocr_phone, invoice_number, total, date, vendor) but can be extended (upon request) with your own algorithm.
Data Transformation on Line Items
You can run your own Rules AFTER the Data Extraction triggered by various vectors. Think of Rules as IF THIS THEN THAT (IFTTT). Apply this to any dimension like Category, Tag, Payment Type or Customer/Project. Learn more
Common Questions & Answers
Q. A field I need is missing. What to do?
A. If you need a field which Veryfi doesn’t support yet, then let us know and we can train the system with you or alternatively provide you with API calls you can make to train the Veryfi system.
Q. Why is Veryfi so fast in extracting fields when alt solutions take much longer?
A. Veryfi’s data extraction and OCR is machines end-to-end. Veryfi is powered a neural network combining different custom architectures and training on a colossal data set of documents of different origins, currencies, languages, layouts.
Q. How do I speak to a human?
Q. What use cases are there for this data?
A. Many solutions but here’s just a few…
Expense Management: extract data for reimbursements or tax bookkeeping or build your own expense app in half the time.
CPG Loyalty: gain insight into what consumers buy and why from SKU product data found on CPG receipts.