AI-Powered Document Intelligence
Strategic Intent
Enable analysts to access relevant sentiment and user-generated insights without leaving Comps, reducing context switching and improving decision confidence.
Context
Reviews in Comps product, helps analysts compare properties by analyzing key features, pricing, and market position. However, analysts often needed an additional layer of qualitative insight, particularly sentiment from platforms like Google Reviews, Yelp and Apartments.com to contextualize a property’s history, neighborhood quality, or tenant experience.
Understanding the Need
Through user interviews, prototype feedback, and observation of day-to-day workflows, we identified a clear tension:
analysts wanted speed, but only if paired with control and transparency.
Usage patterns showed heavy context switching, inconsistent document handling, and difficulty maintaining a single source of truth for analysis.
Pain Points
Manual reading and extraction slowed analysis dramatically.
Insights lived across scattered tools (notes, spreadsheets, PDFs).
Workflows varied widely across the organization, creating inconsistency.
Lack of structure led to duplicated work and errors.
Tool-switching broke focus and delayed decision-making.
AI outputs felt opaque, reducing trust in automation.
Approach
We designed Docs as a modular system where each capability like extraction, workflows, projects, integrations, solves a specific part of the analytical process while contributing to a unified mental model.
The design emphasizes clarity, predictability, and transparent AI interactions, enabling analysts to validate information at every step without losing efficiency.
Core Capabilities
Sentiment Overview: High-level summary of user feedback for each property.
Rating History: Month-over-month rating trends to understand stability or decline.
Topic Breakdown: Clear organization of the most relevant themes mentioned by users.
Detailed Review Insights: Modal view showing complete review details per topic.
Multi-Property Comparison: Ability to compare sentiment across all properties in the CompSet.
Projects →
A structured space to organize documents, workflows, tables, and insights with flexible privacy (private, user-shared, org-wide).
Integrations →
Direct connections to Google Drive, Dropbox, and OneDrive for seamless import and flexible connection/disconnection.
Data Extraction →
Turn unstructured documents into structured, comparable datasets with editable AI prompts and side-by-side validation.