Platform

The ReviewSignal Platform

Institutional-grade review intelligence that combines neural anomaly detection with sentiment propagation modeling to deliver actionable trading signals from consumer data.

ReviewSignal platform overview
53,642+
Locations Tracked
100,024+
Reviews Analyzed
205
Chains Monitored
Real-time
Continuous Monitoring
Neural Core

AI-powered anomaly detection at zero API cost

Our proprietary Neural Core engine transforms raw review data into quantitative intelligence using a three-layer architecture designed for institutional-grade precision.

  • ✓ MiniLM Embeddings — 384-dimensional sentence embeddings capture semantic meaning beyond simple keyword matching, detecting subtle shifts in customer language and tone.
  • ✓ Isolation Forest — Adaptive anomaly detection trained on 8,700+ real samples identifies statistically significant rating deviations, volume spikes, and sentiment shifts across locations.
  • ✓ Welford's Algorithm — Incremental online statistics track running means, variances, and z-scores per location without ever needing to reprocess historical data.
  • ✓ Zero external API cost — The entire pipeline runs locally on our infrastructure. No OpenAI, no cloud NLP. Every embedding, every inference, computed in-house.
384
Dimensions
<1s
Latency
€0
API Cost
Neural Core AI anomaly detection
Echo Engine

Sentiment propagation across geographic and brand networks

The Echo Engine models how consumer sentiment cascades across related locations and competing brands, turning a single signal into a market-wide view.

  • ✓ Geo-propagation — When Starbucks NYC sentiment drops 15%, the engine propagates that signal to Boston, Chicago, and 847 related locations with distance-decay weighting.
  • ✓ Brand coherence — Sentiment shifts at one chain predict movements at competitors. A McDonald's decline often signals broader fast-food headwinds.
  • ✓ Temporal interference — Separates seasonal patterns from genuine shifts. Holiday spikes and summer dips are factored out, leaving only true signal.
  • ✓ Monte Carlo simulation — Confidence scores are computed through thousands of simulations, giving you probability-weighted predictions rather than point estimates.
53,600+
Locations
50+
Countries
0.78
Confidence
Echo Engine sentiment propagation network
Coverage

Massive data coverage across industries

Tracking 205 chains across restaurants, retail, grocery, and drugstores in 50+ countries with reviews from Google Maps, Glassdoor, Trustpilot, and TripAdvisor.

Restaurant chain data coverage
🍔
Fast Food
McDonald's, KFC, Burger King, Subway, Taco Bell
28 chains
☕
Coffee & Bakery
Starbucks, Dunkin', Costa, Tim Hortons
12 chains
🍴
Casual Dining
Chipotle, Panera, Cheesecake Factory, Olive Garden
22 chains
🛒
Retail & Grocery
Whole Foods, CVS, Walgreens, Kroger, Trader Joe's
39 chains
Pipeline

From raw reviews to trading signals

A four-stage pipeline transforms millions of consumer reviews into actionable intelligence for investment teams.

Data pipeline for alternative data
1

Collect

Automated scrapers pull customer reviews from Google Maps, Trustpilot, TripAdvisor, and employee reviews from Glassdoor across 55,000+ locations. Normalized, deduplicated, and quality-scored before storage.

2

Analyze

Neural Core processes each review through 384-dim embeddings, extracting sentiment, topics, and semantic meaning far beyond star ratings.

3

Detect

Isolation Forest flags statistical anomalies: unusual rating drops, sentiment shifts, and review volume spikes that deviate from historical baselines.

4

Signal

Echo Engine propagates detected anomalies across geographic and brand networks, generating confidence-weighted trading signals before earnings reports.

Why ReviewSignal

Built different from every other data vendor

Most alternative data providers sell you raw data and wish you luck. ReviewSignal delivers processed, actionable intelligence that integrates directly into your investment workflow.

  • ✓ Zero API cost architecture — No per-query charges from OpenAI or Google Cloud NLP. Our models run entirely on our infrastructure, keeping your costs predictable.
  • ✓ Real-time monitoring — Reviews are scraped, analyzed, and scored continuously. Sentiment shifts trigger alerts within hours, not days.
  • ✓ Institutional-grade coverage — 205 chains, 53,600+ locations, 50+ countries. The breadth and depth needed for portfolio-level analysis.
  • ✓ Transparent methodology — We publish our approach. MiniLM embeddings, Isolation Forest, Welford's algorithm. No black boxes, no proprietary mystery scores.
  • ✓ REST API integration — Clean JSON endpoints, webhook alerts, CSV/JSON exports. Plugs directly into your existing research stack.
Institutional-grade data intelligence
Get started

Ready to turn reviews into
revenue signals?

Join leading hedge funds using ReviewSignal to detect sentiment shifts before they hit earnings reports.

Start Free Trial View Pricing