What's new
Carta Document Intelligence now analyzes and extracts key data from your fund documents so that you can search and analyze information that was previously locked inside fragmented PDFs.
Document Intelligence is currently live for Stock Purchase Agreements (SPAs), but we will be expanding this capability to more document types. For SPAs, we extract over 150 fields, including purchaser names, share classes, round amounts, and transaction dates, and make them searchable and queryable across your portfolio.
You can view the extracted data in a table, but the real power lies in making extracted data available inside the Carta Data Warehouse, or directly inside your Claude workspace with the Carta Investors Plugin. This allows you to ask questions like:
"Who are our top co-investors across Fund III and Fund IV?"
"Which portfolio companies are missing executed SPAs?"
"What was the average price per share across our Series B investments?"
Why it matters
Almost every piece of data on Carta is rooted in documents. Your ownership data is rooted in the SPA. Your fund terms, management fees, and carry structures are rooted in the limited partnership agreement (LPA). Your LP commitments are rooted in subscription agreements. Historically, these documents have been stored as files, unread and disconnected from the rest of the data.
When Carta can read your fund documents, that changes. The data inside them becomes structured as cells in a table rather than words on a page. This not only makes a wealth of data available for you to derive insights ("who are our top co-investors across our last three funds"), but it helps you power and automate a range of downstream business processes. For example, with structured extraction of your LPA terms, you could automatically verify that the management fees you've booked match what's in the agreement, or flag when a capital call notice doesn't align with the commitment terms on file.
We're starting with SPAs. LPAs and additional document types are coming next. Over time, every document you upload will be available to you to query, and make the system smarter about your firm.
How it works
When you upload a document onto Carta, Carta reads the file and extracts key fields into a structured view on the Document Intelligence page. Every field is linked to a bounding box in the original PDF so you can always trace an extracted value back to its source.
