One of the harder parts of the problem is that CIMs aren't standardized. The same metric can appear multiple times across a document in different formats (tables, narrative text, adjusted figures, projections, etc.).
Our pipeline extracts candidate values, normalizes them into structured metric families, and links them back to document locations so analysts can verify where each number came from.
We're still improving accuracy around projections and adjusted metrics since those are often labeled inconsistently across documents.
We recently launched Valedex, a tool for automating the first pass of private market deal evaluation.
Investment teams reviewing CIMs (Confidential Information Memorandums) often spend hours manually reconstructing financials from messy PDFs before they can even start evaluating the deal.
Valedex extracts financial metrics directly from CIMs, computes key ratios, and surfaces structural risk signals across the deal.
The goal isn't to replace analyst judgment, but to remove the mechanical work so teams can focus on decision-making.
We're currently opening a small number of early pilots with investment teams that regularly review CIMs.
Would love feedback from anyone who works with financial documents, investment memos, or messy PDFs.
One of the harder parts of the problem is that CIMs aren't standardized. The same metric can appear multiple times across a document in different formats (tables, narrative text, adjusted figures, projections, etc.).
Our pipeline extracts candidate values, normalizes them into structured metric families, and links them back to document locations so analysts can verify where each number came from.
We're still improving accuracy around projections and adjusted metrics since those are often labeled inconsistently across documents.