NOTE: AI Import is currently available to a small group of users and will be available to everyone shortly.

This guide covers what to check and how to prepare your documents before starting an AI Import session. For a full walkthrough of the import process itself, see Using AI Import.

Table of Contents

Prerequisites

AI Import is available on the following Valcre plans:

  • Essentials + AI Import Credits
  • Professional + AI Import Credits
  • Enterprise + AI Import Credits

In addition to an active Valcre plan, AI Import requires AI credits. No configuration is required to get started.


Before You Begin

Once a document is imported via AI Import, it cannot be undone. Once you confirm an import, there is no undo or rollback. If you import into an existing Valcre record, its current field values are replaced by the extracted values for every field you have selected. There is no merge option.

This guide exists to help you prepare before you reach that step. Selecting the right document and checking its quality up front means less manual correction during Review and less risk when you commit.


Supported Document Types

AI Import is calibrated for the following document categories. Results on document types outside this list may vary. 

Lease Documents (PDF)

Single-property lease documents, including amendments. The system extracts lease terms, options, parties, rent schedule, and key dates, and creates Lease records linked to the subject property. Commercial rent rolls are also accepted and produce Lease records. Note: Valcre Online does not have a Rent Roll object type

Comp Lists (PDF, Excel, or CSV)

Multi-property comp lists. The system extracts each comparable and creates Lease records, Sale records, or both, depending on what the document contains. Mixed lists that include both lease and sale transactions produce both record types in a single session. 

Flat-table formats where multiple properties and their comps appear as rows in one table are handled automatically: the system detects and groups records by property without requiring you to reformat the source file first. 

Offering Memoranda (PDF)

The system extracts property details and any transaction data present in the document, creating Property records and, depending on content, Lease records, Sale records, or both. 

Documents That Produce Comparable Data 

The following document types are processed by AI Import and produce data that can be imported into the Valcre Workbook as comparable data. Note that imported records from these documents are intended for use as comparables, not as subject property records. 

  • Rent rolls from property owners: Data will be imported and create individual Lease records.
  • Expense Statements and Income Statements: Import matches data to Valcre expense fields.

Snag_60ad7b3.png Note: Multifamily rent rolls are not supported by AI Import.

Documents AI Import Does Not Process

The following document types are not recognized by AI Import. Uploading them will not produce usable output:

  • Multifamily Rent Rolls
  • RealQuest Property Detail Reports
  • Property Tax Statements
  • Federal Tax Returns

Document Requirements

Before uploading, confirm your document meets all of the following requirements. Files that do not meet these requirements are rejected at upload with an error message. 

Item Requirement
File type PDF, Excel (.xlsx), or CSV
File size 25 MB maximum
Page count 85 pages maximum
Password protection Remove password protection before uploading

Preparing Your Document for Best Results

The quality of what AI Import extracts is directly tied to the quality of the source document. The steps below reduce the amount of manual correction needed during the Review stage.

Use a Native PDF When Possible

A native PDF, one generated directly from a digital source, or a document saved to PDF from Word or Excel, will always produce the most accurate extraction. If a native version of your document is available, use it rather than a scan.

Snag_60ad7b3.png Pro-Tip: If your source is a data platform, request a fresh export in PDF or Excel format directly from the platform rather than scanning a printed copy.

If You Must Use a Scanned PDF

Scanned PDFs are accepted, but accuracy degrades on low-resolution or skewed scans, which are common in older CRE documents. Degraded scans may require more manual correction during the Review stage.

Before uploading a scanned PDF:

Do Avoid
Use the highest resolution scan available Uploading blurry or faded scans
Confirm the document is straight and not rotated or skewed on the page Scans where the document sits at an angle
Verify the text is legible at 100% zoom before uploading Compressed or reduced-resolution versions of scans

Check Your Document Against the Page and Size Limits

For large documents, check the page count and file size before uploading. Documents over 85 pages or 25 MB are rejected at upload.

If your document exceeds either limit:

  • Page count: Split the document into sections and run a separate AI Import session for each section.
  • File size: Check whether a lower-size export is available, or split the document as above.

One File Per Session

AI Import processes one file per upload session. If you have multiple documents to import, run a separate session for each file.


Pre-Upload Checklist

Use this checklist before starting each AI Import session.

  • The document is one of the supported file types: PDF, Excel (.xlsx), or CSV 
  • The document is under 85 pages and under 25 MB 
  • If it is a PDF, it is a native PDF where possible, or a high-resolution scan if not 
  • If it is a scanned PDF, the text is legible at 100% zoom and the page is straight 
  • Password protection has been removed, if applicable 
  • The document is one of the supported document types listed above 
  • You have reviewed the no-undo warning at the top of this guide and are ready to proceed

 

Ready to use AI Import? Get started with our Using AI Import guide. 


 

Additional Questions?  Valcre Support is ready to help with all your Valcre product questions! You can connect to us via phone, email, or chat! 

Was this article helpful?
0 out of 0 found this helpful

Comments

0 comments

Please sign in to leave a comment.