Patterns · persuasion

Social Proof

Displaying evidence that others have made the same decision — reviews, ratings, testimonials, user counts, or 'X people bought this recently.' Social proof reduces perceived risk by borrowing credibility from the crowd. Cialdini's Consensus principle: when uncertain, people look to others' behavior as a guide. Most effective when the proof comes from people the target user identifies with.

Typical CVR lift
+2.0–15.0%
per published studies
Category
persuasion
Source
VWO A/B Testing Benchmarks 2024

When it works

High-consideration purchases where risk is perceived. Products with large user bases (high review count). Contexts where the target audience identifies with existing users. Above the fold on product and landing pages. Most effective when reviews are specific, recent, and verified.

When it backfires

Negative reviews displayed prominently. Very low review counts (3 reviews hurts more than 3,000 helps). When audience doesn't identify with the reviewers cited. When reviews look generic or templated. When social proof contradicts the page's positioning (e.g., 'enterprise-grade' with only SMB testimonials).

Ethical notes

Only display genuine, unfiltered reviews. Do not suppress negative reviews — this is a deceptive practice flagged by the FTC and EU Digital Services Act. Never manufacture fake social proof ('Join 10,000 happy customers' when you have 47 users). Star ratings must reflect real distribution, not cherry-picked 5-star reviews.

Examples in the wild

Amazon star ratings on search results

Review count and star rating shown on every product card — highest-studied social proof implementation in ecommerce

Basecamp user count in hero

Specific large user count in hero section; specific numbers (e.g., 75,000 companies) are more credible than round estimates

Booking.com 'Viewed 18 times today'

Recency-specific social proof that blends into urgency territory; most effective on hotel/travel booking flows