Automatic Data population from PDF files to Azure Database for Insurance Company

data population


Insurance service providers operate in a highly competitive environment wherein customers and stakeholders expect on-demand settlement of claims.

Despite digitization of processes, the insurance industry is prone to delay in grievance handling and claims processing resulting into rise in frauds and a dissatisfied consumer. One of the unresolved bottlenecks here is the errors in data population due to manual tasking.

In the following case study, we walk you through a solution wherein the data population from uploaded PDF forms to the Azure CRM’s database is automated. Besides nullified manual errors, the solution empowered the business to perform quick transfers for thousands of files.

Business Requirement

The customer is a popular insurance services provider and currently using the Azure cloud account. The business experiences huge volumes of customer data upload on the services portal. Given the increasing workloads, the business wanted a solution wherein the data from the uploaded PDF is automatically copied to the Azure SQL database.

By automating the data population task, they wanted to cut down on the tedious manual tasking and fasten processing at abbreviated costs.

Solution Overview

The Intellinez team used Azure Form Recognizer to extract the text and tables from the uploaded files.

The Form Recognizer labelling tool consumes input from the PDF file and analyses it using the trained model to provide JSON output.

The JSON output is then used to extract the required data and copy it to SQL table(s).

Solution Flow

The Azure functions trigger and log whenever a file is uploaded to the Azure Storage account. Then, use Form Recognizer REST API and Python to get the analyzed results in JSON format.

Once the data is received In JSON format, use Python code to extract only the required fields as table format and copy them into SQL table.

Azure Form Recognizer

Form Recognizer is part of Microsoft Azure Cognitive Service. It is a prebuilt AI feature that can be easily used via an API call. The Form Recognizer uses machine learning to parse the source file structure and extract data into components.

Azure Blob storage

It is Microsoft’s object storage solution for the cloud. Blob storage is optimized for storing massive amounts of unstructured data. Unstructured data is data that does not adhere to a data model or definition, such as text or binary data

Azure Functions

Azure Functions integrates with Azure Storage via triggers and bindings. Integrating with Blob storage allows you to build functions that react to changes in blob data as well as read and write values


The Azure Functions were used to trigger Blob storage and log in to the Azure storage account whenever a file upload event is generated. The following pre-requisites were required

  1. Python
  2. Form Recognizer model
  3. Visual Studio Code

Initial Setup

Install Python, Visual Studio and give configuration details

Integrate with Blob Storage

Perform Azure functions using Triggers and Bindings

Connect Form Recognizer

Using Python code to get data for every file

Initial Setup

The output can be seen in the SQL table while the logs are generated in Azure Functions


The output can be seen in the SQL table as the data gets copied. Also, logs are generated in Azure Functions every time a new blob is processed and triggered.

Business Benefits

The business was able to perform automatic updating of their Azure database from an external PDF file thereby saving time and operational cost.

With instantaneous access to customer data on demand, the insurance services provider can now fasten processing of claims and implement fraud detection protocols more effectively.



Got a Project in Mind? Get Started Now