Projects
When you need to compare terms across dozens of contracts, the manual approach is painful: open each document, search for the relevant clause, copy the value into a spreadsheet, repeat. A single comparison project can take days of paralegal time, and the results are only as accurate as the person reading each clause.
Pactly Projects automate this extraction. Select contracts from your repository, define the data points you need (governing law, liability caps, termination triggers, expiry dates), and let the AI read every document and populate a structured grid. You get a sortable, exportable table with values, source references, and AI reasoning for each cell.
What’s Covered in This Section
Section titled “What’s Covered in This Section”- Analyze Contracts - Create a project, add contracts, configure columns, and run AI extraction
- Review Analysis Results - Work with the results grid, view AI reasoning and source quotes, and handle errors or outdated cells
- Export Project Results - Download results to Excel with optional reasoning and source references
Project Types
Section titled “Project Types”Projects are designed to support different types of analysis. The first available project type is Contract Analysis (tabular review), which extracts structured data points from contracts into a spreadsheet-like grid.
Additional project types are planned for future releases, including regulatory monitoring and legal research.
Key Concepts
Section titled “Key Concepts”
Each column represents a data point you want to extract. A column has a question (the prompt the AI uses), a data type (Text, Number, Date, Yes/No, or Multiple Values), and a label. You can choose from predefined templates or write your own questions.
When you run analysis, the AI reads each contract and extracts values for every column. Results include the extracted value, the AI's reasoning, and a reference to the source clause in the document. Analysis runs in the background, so you can continue working while it processes.
Each cell in the results grid has a status: completed (value found), N/A (not found in the document), processing (extraction in progress), error (extraction failed), or outdated (the column's question was changed since this cell was last extracted). You can re-run individual cells or the entire project.
Column templates are predefined extraction questions grouped by category (General, Liability, Termination, IP & Confidentiality). They provide a starting point so you don't have to write prompts from scratch for common data points like governing law, expiry date, or liability cap.