Edit

Understand use case and input documents

Report

This is a POC "Report". It is necessary to study and process documents of different templates and extract some fields that contain key information which will be entered into inner systems. The timeline of the project is 2 months.

Number of documents	Number of templates	Names of templates	Number of fields	Success criteria
619	4	New Hampshire Massachusetts Georgia California	15	Accuracy: 90% Automation rate: 60%

Business logic

All entities conducting business are required to file annual reports to the Secretary of State. Reports must be signed, dated and accompanied by a filing fee. They confirm the principal and registered agent/office. There are also names and addresses of the entity's officers, directors, members, managers; and necessary changes. Some annual reports are filled in online, and part is printed and submitted by mail. Such reports arrive as PDF attachments and have to be proceeded manually. All the necessary data from each report is defined and entered manually into a data store. The monthly distribution is not equal; there are approximately 500–600 reports submitted monthly from January to June, 0–100 documents monthly from July to September, and 200–300 documents from September to December. For this use case, only four states are chosen: California, Massachusetts, Georgia and New Hampshire.

Team

There are two SMEs (part-time availability) which perform this task. On the project, they will be available four hours daily.

List of fields

Field name	Values	Data type	Meaning/Notes	Required	Multivalue
Secretary of State	California, Massachusetts, Georgia, New Hampshire	Alphabetic	Secretary of State where the report is submitted	Yes	No
Trademark Name	Any	Character	Company name, which submits the report	Yes	No
Corporate Number	Usually there are 9 digits, but there can be letters as well	Alphanumeric	Company ID	Yes	No
Entity Type	Corporation, Domestic Profit Corporation	Alphabetic	Type of ownership	Yes	No
Business Type	Any	Character	Type of company's activity	No	No
Filing Date	Any	Date	Date when the report was signed	Yes	No
Effective Date	Any	Date	Date when the changes become valid	No	No
Authorized Person	Any	Character	Person that signed the report	Yes	No
Authorized Title	Authorized Person, Incorporator, CEO, Secretary, President	Alphabetic	Title of the authorized person	Yes	No
Agent Name	Any	Character	Name of the company or person-founder	Yes	No
Street Address	Any	Alphanumeric	Street address of the principal office	Yes	No
City	Any	Alphabetic	City of the principal office	Yes	No
State	GA, CA, MA, NH	Alphabetic	State of the principal office	Yes	No
Zip Code	Any	Numeric	Zip code of the principal office	Yes	No
Filing Fee	It can take any value, but usually fee equals to $100 or $25	Numeric	Filing fee, which is charged when reports are sent after the due date	No	No

Documents

To see the original documents download and unpack this archive.

For better understanding of the workflow, here is the list of action items that need to be done on the use case at this point which are not shown in this demo.

1. From the customer, request access to the folder with documents, working instance and S3.
2. Look through the documents and study the business logic of the documents and use case.
3. Organize the kickoff meeting with the team of SMEs, discuss business logic of the documents, and explain next steps.
4. Write an email to the team with the results of document analysis.