Module 21: Reviews, Ratings & FTC Compliance
Review signal patents plus the 2024 FTC Consumer Review Rule — the legal framework that governs all review practices.
Overview
Reviews are both a ranking signal (from Google's patent perspective) and a legal compliance matter (from the FTC's perspective). This module covers both — the patent mechanisms that determine how reviews influence rankings, and the 2024 FTC Consumer Review Rule that makes certain review practices illegal.
Review Signal Patents
US8417713B1 - Sentiment Detection as Ranking Signal
Year: 2007-2013
The Core Review Patent:
Google analyzes sentiment in review TEXT — not just star ratings.
How It Works:
- Review texts identified referencing specific businesses
- Sentiment analysis applied (positive/neutral/negative)
- Sentiment scores generated per review
- Aggregate sentiment combined with other ranking signals
- Business ranking influenced by sentiment score
Key Insight: The actual WORDS in reviews matter as much as the star rating. A 4-star review that says "good but the service was slow and the staff was rude" has negative sentiment even with a positive rating.
US9792330B1 - Identifying Local Experts for Local Search
Year: 2013-2017
Review Weighting by Reviewer Expertise:
Not all reviews are equal. Google weights reviews differently based on reviewer expertise.
Expert Identification:
- Reviewer has multiple reviews in the same business category
- Reviews are specific to the geographic area
- Reviews not flagged as spam
- Expert status in a category = higher review weight
Practical Impact:
- A review from someone who has reviewed 50 restaurants carries more weight than a first-time reviewer
- Local Guides with high review counts are effectively "expert reviewers"
- Fake reviews from new accounts with no history get lower weight
US7996210B2 - Large-Scale Sentiment Analysis
Year: 2008-2011
Sentiment at Scale:
- Sentiment analysis applied across millions of reviews
- Topic-specific sentiment (food quality, service, price, ambiance)
- Comparative sentiment (better/worse than competitor)
- Temporal sentiment trends (improving vs. declining)
Review Signal Hierarchy
Based on patent analysis, reviews are weighted as follows:
| Signal | Impact | Patent |
|---|---|---|
| Sentiment of text content | High | US8417713B1 |
| Reviewer expertise level | High | US9792330B1 |
| Review quantity | Medium | Multiple |
| Star rating | Medium | Multiple |
| Review recency | Medium | US8549014B2 |
| Review detail/length | Medium | US8417713B1 |
| Response engagement | Low | N/A (indirect) |
The 2024 FTC Consumer Review Rule
Effective: October 21, 2024 Authority: Federal Trade Commission (16 CFR Part 465) Penalties: Up to $51,744 per violation
What Is Prohibited
1. Fake Reviews or Testimonials
- Creating fake reviews (company employees, AI-generated, paid without disclosure)
- Soliciting fake reviews from third parties
- Paying for reviews without requiring disclosure
- AI-generated reviews presented as authentic human experiences
2. Review Gating (NOW ILLEGAL) Review gating = pre-screening customers based on expected review sentiment before soliciting a review.
ILLEGAL: "Were you satisfied with your experience?"
→ YES: "Please leave us a review on Google!"
→ NO: "Please contact us directly to resolve."
LEGAL: "Would you like to share your experience?"
→ Sends ALL customers to the same review platform3. Suppressing or Hiding Negative Reviews
- Removing negative reviews from your platform
- Hiding reviews below a certain star threshold
- Not publishing all submitted reviews
4. Buying Positive Reviews Without Disclosure
- Paying customers (cash, discounts, free products) for reviews without requiring them to disclose the incentive
- "Incentivized reviews" must clearly state the incentive
5. Operating Fake Review Websites
- Creating websites that appear to be independent review sites but are controlled by the business
6. Buying Social Media Indicators
- Buying followers, likes, or engagement that misrepresents popularity
Case Study: Fashion Nova Settlement
FTC v. Fashion Nova (January 2022)
- Fashion Nova suppressed negative reviews on their own website
- Only reviews above a certain star threshold were published
- Settlement: $4.2 million fine
- Required to pay customers whose reviews were suppressed
Compliant Review Strategy
The Compliant Approach:
Step 1: Ask ALL customers for reviews equally
(no pre-screening based on satisfaction)
Step 2: Send ALL customers to the SAME review platform
(no separate paths for happy vs. unhappy)
Step 3: Publish ALL reviews you receive
(positive and negative)
Step 4: Respond professionally to negative reviews
(publicly, helpfully, without hostility)
Step 5: If incentivizing reviews, require disclosure
("I received a discount in exchange for this review")What You CAN Do (Legally)
LEGAL: Ask ALL customers to leave a review
LEGAL: Send a follow-up email asking for a review
LEGAL: Make it easy to leave a review (QR code, link)
LEGAL: Respond to both positive and negative reviews
LEGAL: Report fake reviews to Google for removal
LEGAL: Offer an incentive IF disclosure is required
LEGAL: Thank reviewers for their feedbackWhat You CANNOT Do (Illegal)
ILLEGAL: Route only happy customers to Google reviews
ILLEGAL: Ask unhappy customers to contact you instead of reviewing
ILLEGAL: Pay for reviews without requiring disclosure
ILLEGAL: Use AI to write fake reviews
ILLEGAL: Suppress/hide negative reviews
ILLEGAL: Buy followers or fake social proof
ILLEGAL: Create fake "consumer" review sites you controlReview Management Best Practices (Patent + FTC Aligned)
1. Maximize Review Sentiment Quality
Per US8417713B1, sentiment in TEXT matters:
- Ask customers to describe specific experiences in their reviews
- "What did you appreciate most about our service?"
- Specific positive language in reviews = stronger positive sentiment signal
2. Get Reviews from Engaged Local Customers
Per US9792330B1, expert reviewers count more:
- Customers who actively review other businesses have higher expert weight
- Regular customers more likely to be "local experts" in your category
- Focus review requests on engaged customers (not first-timers)
3. Respond to All Reviews
While not directly a ranking signal, response signals:
- Business activity (active businesses rank better)
- Trust (customers see you engage)
- Sentiment management (professional responses soften negative impact)
4. Review Volume Strategy
Consistent review cadence > periodic spikes
SUSPICIOUS: 0 reviews for 6 months → 50 reviews in one week
NATURAL: 2-5 reviews per month, consistentlyGoogle's Review Spam Policies
Beyond FTC compliance, Google has its own review policies:
Google Removes Reviews That:
- Are fake or from accounts that don't exist
- Come from the same device/IP as the business
- Are posted by employees of the business
- Violate content policies (off-topic, spam, adult content)
- Are written by competitors about your business
Key Patents Referenced
| Patent | Title | Year |
|---|---|---|
| US8417713B1 | Sentiment Detection as Ranking Signal | 2007-2013 |
| US9792330B1 | Identifying Local Experts | 2013-2017 |
| US7996210B2 | Large-Scale Sentiment Analysis | 2008-2011 |
Legal References
| Document | Year | Authority |
|---|---|---|
| FTC Consumer Review Rule (16 CFR 465) | 2024 | FTC |
| FTC v. Fashion Nova | 2022 | $4.2M settlement |
| FTC Guides Concerning Endorsements and Testimonials | 2023 Update | FTC |
Next Steps
- CTR & User Behavior Module — Click signals
- Local SEO Module — Review signals in local context
- Social Trust Signals Audit — Apply review signals