opens its technology to integrate with other CRE software applications

Who can benefit?

Software platforms that serve the commercial real estate market like banks loan management and servicing systems, deal pipeline management, due-diligence document management, cashflow analytics software, listing platforms with due-diligence document vault, CRMs and others.

Target Audience?

CXO, CTO, Product Managers, Senior Software Developers, Architect, VP Engineering, ML and AI engineers at innovative CRE focused software firms or financial institutions developing web applications.


This document details the approach of integrating ClikAI's Smart Extract with the third party applications. For the sake of simplicity, we are going to call these applications as Client/Lender applications.

The following points are detailed out as a part of the integration process:

  • About ‘Smart Extract’
  • Authentication
  • The Smart Extraction Process
  • Data Security & Compliance

About ClikAI's Smart Extract

Smart Extract helps other software providers extract the normalize data from financial documents such as Rent Rolls, Operating/Income Statements.



The rent roll and OS pdf data is extracted and presented in a spreadsheet that can be reviewed, edited, and saved for further processing by the underwriters and asset managers.

Integrating Smart Extract

The illustration below shows a detailed component diagram explaining how the Client application can integrate Smart Extract.

The key highlight is that the Smart Extract can be loaded in an iframe within the Client application, making the integration as simple as it can get.

Let us look at each of the above steps in detail.

Step 1 - Authentication


The authentication between the third party applications and Smart Extract involves the following steps:

  1. Once the Client/Lender application signs up for Smart Extract, they can generate API credentials required for authentication.
  2. Using the credentials, the Client/Lender application invokes the Auth API to authenticate the self.
  3. Upon successful authentication, the Client/Lender application receives a token from

Step 2 - Load Smart Extract in an iFrame



Using the auth token, the Client/Lender application makes a request to load the Smart Extract within an iFrame.

Step 3 - Extraction


The following steps are a part of the Extraction process:

  1. Once the Smart Extract is loaded within the iFrame, it sends a ‘Ready message’ to the Client/Lender application.
  2. The Client/Lender application upon receiving the ‘Ready message’, immediately sends the ‘Extraction Request’ with the document to be extracted to Smart Extract (via JavaScript PostMessage API).
  3. Smart Extract receives the request for extraction and forwards the same to’s extraction engine, which performs the ‘Smart Extraction’ process of extracting the relevant and actionable data from the document.
  4. Once the extraction job is completed, the data is loaded into a spreadsheet for further review and editing.


Step 4 - Editing & Saving the Extracted Data


Once the extracted data is available within the Smart Extract in the form of a spreadsheet, the underwriter can review the data and edit the same.


Upon save action, the entire extracted data with the user modifications is sent back to the Client/Lender application in the form of a JSON object (via Javascript PostMessage API).


An important thing to note here is that does not store any information from the extracted data ensuring that the data security & compliance requirements are met.


Data Security/Compliance understands that data security/compliance is of paramount importance to financial institutions and ensures that sensitive information extracted from the documents is not stored anywhere.


All the information extracted from the document is kept in memory and when the user does save after the edits he has made in the spreadsheet, the data from the spreadsheet is sent back to the parent application in the form of a JSON object.

While no sensitive information is stored, does need to store the following metadata for tracking and logging purposes.

  • Asset Type
  • Document Type
  • Document Hash - A Hash generated to identify each document uniquely
  • Status - Whether the extraction was successful or encountered any error.
  • Extraction Date - The date the document was extracted.
  • Extraction Time Taken - The time taken to perform extraction of the document.
  • Number of Pages - The number of pages extracted from the document.
  • Number of Rows - The number of rows extracted from the document.


The integration of Smart Extract is seamless and for a detailed integration of Smart Extract.