Generate more accurate, explainable responses by combining data from your Q/A Library, Project Responses, and uploaded Documents, all in a single click.
The Challenge
Due diligence questionnaires often contain dozens, sometimes hundreds, of questions that teams must answer repeatedly across different projects. Historically, answering each question meant manually searching through past responses, sifting through internal documents, and copy-pasting information from multiple places. Even with standard autofill, the system could only match one similar question at a time, often missing richer context buried in related responses or supporting documents.
The Goal
Instead of relying on a single similar question, enable the system to search across multiple sources, understand both questions and actual response content, and combine information to form a complete, well-sourced answer.
Getting Started
- Open Auto-Fill: From any questionnaire, click the Auto-Fill button in the top toolbar and select Responses.
- Select AI Mode: Toggle to AI Mode at the top of the Auto-fill dialog to enable multi-source retrieval and AI-powered synthesis.
- Choose Your Sources: Select one or more sources — Library Q/As, Project Q/As, and/or Documents. When Documents is selected, you can use an existing document or upload a new one.
- Configure Filters: Set your content date range, enable exact-match prioritization, and optionally scope to specific entities or projects.
- Auto-fill as Drafts: Click Auto-fill as Drafts to run. The AI will generate responses across all applicable questions and save them as drafts for your team to review. Since these responses are in draft, users need to finalize them.
How Multi-Source AI Autofill Works
When you trigger AI Autofill, the system retrieves relevant information from up to three distinct source types, evaluates the content for relevance and recency, and synthesizes a cohesive draft response. All of this, without requiring you to leave the questionnaire.
Sources
| ✓ Library Q/As | ✓ Project Q/As | ✓ Documents |
Library Q/As draws from your firm’s centralized Q/A Library, the canonical repository of approved responses that your team maintains over time.
Project Q/As retrieves answers from previous and ongoing projects, capturing how your firm actually responded in specific contexts.
Documents allow you to include information from supporting materials such as pitch decks, fund overviews, compliance policies, or any file relevant to the questionnaire, giving the AI access to information that may not yet exist in structured Q/A form.
You can select any combination of these sources, and the system pre-filters content to your firm automatically, ensuring responses are drawn only from data you own or have been assigned.
The Retrieval and Generation Process
- Multi-Source Search: For each question, the AI searches across all selected sources simultaneously, like Library Q/As, Project Q/As, and Documents, to find the most relevant matches.
- Content Understanding: The system analyzes not just question similarity but the actual content of past responses and documents, identifying passages that directly address the question being asked.
- Intelligent Synthesis: Information from multiple sources is combined into a single, coherent draft response. It is not just copied from one place, but synthesized to form a complete and contextually appropriate answer.
- Draft with Provenance: Every AI-generated response is saved as a draft for human review, accompanied by source details showing exactly where the information came from, along with a confidence score and reasoning.
Smart Filters for Precision
AI Autofill includes a set of configurable filters that let you fine-tune which content the AI considers, ensuring responses are accurate and relevant to your specific context.
Content Date Range Restrict source content to a specific time window so the AI prioritizes the most current information available. | Prioritize Exact Question Matches before AI When an exact question match exists, use that response directly instead of generating an AI answer, reducing review overhead. |
Include Responses from Specific Entities/Investor firms Limit source content to responses from specific entities within your firm for fund-specific or strategy-specific accuracy. | Select Specific Projects Select specific projects to draw responses from, ensuring the AI references only the most contextually relevant prior work. |
Built-In Transparency and Trust
Every AI-generated response includes a layer of explainability designed to make review faster and build confidence in the output.
AI-generated responses include:
Source attribution detailing which Library Q/As, Project Q/As, or Documents contributed to the answer; a confidence score indicating the AI’s assessment of response quality; and reasoning that explains how the response was assembled.
All AI-generated content is saved as a draft — never published directly. This ensures your team retains full editorial control, reviewing, editing, and approving every response before it reaches the investor.
A complete Auto-Filled History log is also available, providing a full audit trail of every autofill action: when it was run, which sources were used, and what was generated. You can also delete auto-filled responses in bulk if a re-run with different parameters is needed.
Standard vs. AI Mode
AI Autofill operates alongside the existing Standard autofill mode. Here’s how they compare:
| Capability | Standard Mode | AI Mode |
| Source types | Library Q/As only | Library + Project Q/As + Documents |
| Matching approach | Question-to-question similarity | Question + response content understanding |
| Multi-source synthesis | — | ✓ |
| Confidence scoring | — | ✓ |
| Source attribution | Matched question reference | Detailed provenance per response |
| Exact-match prioritization | — | ✓ |
| Document ingestion | — | ✓ |
| Output format | Direct fill | Saved as draft for review |