Originally published August 2011. Updated June 14, 2026.
Online reviews are the user-generated trust signal that now drives two buyer decisions at once: the human picking a hotel, a doctor, a fintech app, or a crypto exchange — and the answer engine (ChatGPT, Claude, Perplexity, Gemini, Google AI Overviews) that summarizes those choices into a single recommendation. Since the U.S. Federal Trade Commission’s Consumer Review Rule (16 CFR Part 465) took effect on October 21, 2024, fake reviews carry civil penalties of up to $51,744 per violation, and the agency opened its first enforcement sweep against 10 companies on December 22, 2025.
Two things changed simultaneously in 2024–2026. The legal floor under reviews moved up sharply. And the strategic weight of reviews moved up even faster, because answer engines now ingest review corpora as a primary trust signal — the kind of signal that decides which brand a model names in a paragraph and which one it leaves out.
This piece is the operator’s guide. What changed, what counts, and what to build now.
The Online Reviews Landscape in 2026
The review economy spans roughly seven surfaces that matter, each with its own weight in both human decision-making and AI retrieval.
- Google Reviews / Google Business Profile — the largest local-trust surface. Reviewed in Google’s AI Overviews and weighted in Gemini answers for local-intent queries.
- Yelp — still dominant for U.S. restaurant, hospitality, and local service. Yelp signed a content licensing agreement with OpenAI in 2024, giving ChatGPT direct access to its review corpus.
- Trustpilot — the consumer-facing trust surface most cited in answer-engine responses for SaaS, ecommerce, fintech, and crypto. More than 300 million cumulative reviews as of 2025.
- G2 and Capterra — the B2B SaaS surfaces. G2 has more than 2.4 million verified reviews and explicit deals with multiple AI engines.
- Amazon, Booking.com, TripAdvisor — transactional review surfaces with closed ecosystems but enormous retrieval weight inside their categories.
- App Store and Google Play — mobile app trust signal; weighted heavily for any “best app for X” query.
- Reddit, forum threads, and YouTube comments — the unstructured review corpus. The engines treat these as primary trust signal because they are harder to manipulate.
None of these surfaces are interchangeable. Each is a separate decision about where a brand earns trust and where it gets cited.
What the FTC Consumer Review Rule Actually Bans
The FTC Consumer Review Rule, finalized August 14, 2024 on a 5–0 commission vote and effective October 21, 2024, codifies six categories of prohibited practice.
- Fake reviews and testimonials — reviews from people who do not exist (including AI-generated fakes) or who never used the product.
- Buying or selling reviews — paying for positive or negative reviews, in money or in kind.
- Insider reviews without disclosure — reviews by employees, officers, or relatives of company personnel without clear disclosure of the relationship.
- Company-controlled review websites that present as independent.
- Review suppression — using unfounded legal threats, intimidation, or other coercive tactics to remove legitimate negative reviews.
- Fake indicators of social media influence — buying followers, likes, views, or engagement that misrepresent reach.
Civil penalties run up to $51,744 per violation. The first enforcement sweep, December 22, 2025, sent warning letters to 10 unidentified companies and signaled that the FTC intends to actually use the rule, not just announce it.
The rule explicitly names AI-generated fake reviews as a prohibited category — the first piece of U.S. federal consumer-protection regulation to do so.
Why AI Engines Weight Reviews Differently Than Search Engines Do
Traditional search ranked brands. Answer engines synthesize them. The difference matters for reviews.
When Google ranked “best CRM for small business,” the brand with the best on-page SEO and link graph won. When ChatGPT answers the same query, it summarizes the consensus across review corpora and editorial coverage and names two or three brands by name in a paragraph.
For that synthesis, the engines lean on review surfaces in three ways. They use review volume to weight which brands are even candidates for citation. They use review sentiment to color the language they use about each brand. And they use specific review excerpts — sometimes paraphrased, sometimes quoted — to anchor their recommendations.
A brand with 4,200 verified G2 reviews appears in ChatGPT’s “best CRM” answer. A brand with 38 does not. A brand with 4.6-star sentiment gets named with positive language. A brand with 3.1-star sentiment, if cited at all, gets named with reservation.
This is the structural reason review strategy is now part of communications, not just customer success.
Reviews in Healthcare, Financial Services, and Crypto
Three categories operate under tighter rules than general consumer review work and deserve separate treatment.
Healthcare reviews sit at the intersection of HIPAA and provider trust. Healthcare providers cannot respond to reviews in ways that confirm a patient relationship without authorization. Platforms including Healthgrades, Zocdoc, and Vitals layer their own verification on top of Google and Yelp. AI engines now answer “best cardiologist near me” or “most reviewed dermatologist in Austin” by pulling from this corpus — making patient-review programs a direct input into healthcare communications outcomes.
Financial services reviews trigger FINRA and SEC scrutiny. FINRA Rule 2210 governs communications with the public; testimonials and endorsements that constitute “retail communications” require disclosure of compensation, material conflicts of interest, and the fact that the testimonial may not represent typical experience. The SEC’s amended Marketing Rule (Rule 206(4)-1), effective November 2022, opened registered investment advisers to testimonials and endorsements for the first time in decades — and immediately raised compliance expectations.
Crypto and Web3 reviews sit in the highest-risk category. The SEC has brought multiple enforcement actions against celebrities and influencers for undisclosed promotion of digital assets. App store reviews, exchange reviews on platforms like CoinGecko and Trustpilot, and community discussion on Reddit and Discord drive the bulk of buyer trust. The FTC’s Consumer Review Rule applies. So does Section 17(b) of the Securities Act for any review tied to an investment opportunity.
What Brands Should Do Now
Six moves, in order.
One. Audit the review surfaces that matter for your category. Not all of them. The two or three where buyers actually decide. For B2B SaaS, that is G2 and Capterra. For local services, Google and Yelp. For fintech and crypto, Trustpilot plus the relevant app store. Map the volume, sentiment, and recency on each.
Two. Build a compliant review-acquisition program. Solicit reviews from real customers, in the ordinary course, without incentivizing sentiment. Document the process. Train customer-facing teams on the new rule.
Three. Kill any insider reviews still on your surfaces. Employee reviews without clear disclosure are now prohibited. Even legacy reviews from years ago carry exposure.
Four. Respond to negative reviews on the record. The act of responding is itself a trust signal. Engines weight responsiveness. So do buyers reading the thread.
Five. Build a measurement layer. Track review volume, sentiment, and citation share inside the major answer engines for your category’s top 20 buyer prompts. Run that audit quarterly.
Six. Treat reviews as a comms surface, not a customer-success leftover. The team that owns reviews now sits adjacent to the team that owns earned media, because both feed the same retrieval graph.
Frequently Asked Questions About Online Reviews in 2026
Are AI-generated fake reviews illegal?
Yes. The FTC Consumer Review Rule, effective October 21, 2024, explicitly names AI-generated reviews from people who do not exist as a prohibited category. Civil penalties run up to $51,744 per violation.
Can a business pay for positive reviews if the reviewers actually used the product?
No. The rule prohibits buying positive or negative reviews regardless of whether the reviewer used the product. The exchange of money or value for review sentiment is itself the violation.
How do AI engines weight online reviews when generating answers?
Engines use review volume to weight which brands are candidates for citation, review sentiment to shape the language they use about each brand, and specific review excerpts to anchor their recommendations. A brand with 4,000 verified reviews at 4.6 stars consistently outperforms a brand with 40 reviews at 3.0 stars in answer-engine citation share.
What review surfaces matter most for B2B SaaS brands?
G2 and Capterra are the dominant B2B review surfaces and are weighted heavily across ChatGPT, Claude, Perplexity, and Gemini for software-category queries. G2 alone has more than 2.4 million verified reviews and licensing relationships with multiple AI engines.
Do healthcare providers face additional rules around online reviews?
Yes. HIPAA restrictions prevent providers from responding to reviews in ways that confirm a patient relationship without authorization. Providers must train staff on compliant response practices and rely on platform-level verification rather than direct rebuttal.
What is review suppression and why is it now illegal?
Review suppression is the use of unfounded legal threats, intimidation, or coercion to remove legitimate negative reviews. The FTC Consumer Review Rule explicitly prohibits these practices and treats them as deceptive trade practice subject to civil penalty.
Related coverage on Everything-PR:
