You have the social proof - you're just showing it in the wrong place.

Placement beats volume.

Across the 1,200+ pages Levri has analysed, 74% have at least one form of social proof - a testimonial, a logo strip, a star rating. Only 19% place it where it actually influences the purchase. That gap is the opportunity: moving the same testimonial from below the fold to above it has lifted conversion 63% in a controlled test, and a proof widget placed directly under the primary CTA (Call to action) lifted it 68% - more than adding video, more than doubling review count.

Most CRO advice says "add more reviews." The data says placement is the bigger lever - same proof, different position, a 60%+ swing. (What proof to show, and the trust psychology behind it, is the companion guide trust signals that convert; this one is purely about where it goes.)

The six placement patterns that move conversion.

1. Reviews below the fold - the proof exists, but buyers leave before they see it.
2. Proof far from the CTA - it answers "should I trust this?" three scrolls from "should I click?"
3. Vague instead of specific - "trusted by thousands" converts like a blank page.
4. Wrong format for the page - video where text belongs, and vice versa.
5. Perfect ratings - a flawless 5.0 reads as fake.
6. No real-time proof - static "someone liked this once" instead of "someone's buying now."

#1 - Reviews buried below the fold.

Products with five or more reviews convert 270% better than products with none - 380% for items over $100. But most sites push the review section to the page foot, below three scrolls of specs nobody reads. The problem isn't the reviews; it's that buyers decide to leave before seeing them. This is the placement layer of the proof work in product page conversion.

What we see on scans

• Star rating and review count not in the first viewport.
• The review section after description, specs, and related products.
• No compact rating summary near the title or price.
• A review widget that lazy-loads, adding 1-2 seconds before it renders.

What to ship

• Put a compact star rating with review count directly below the product title - above the fold.
• Make the rating clickable to scroll to the full reviews.
• Add a volume badge ("1,200+ sold") next to it.
• Preload the review widget - lazy-loading your most persuasive element defeats the point.

Typical lift: +12% to +63% depending on current placement. Impact: high - the single highest-leverage move for pages that already have reviews.

#2 - Social proof too far from the CTA.

Position matters more than type. A proof widget adjacent to the primary CTA lifted conversion 68% in one study - more than video testimonials, more than real-time notifications. The mechanism is simple: proof answers "should I trust this?" at the exact moment the buyer decides "should I click?" Separate those by three scrolls and it stops working.

What we see on scans

• Testimonials in a section 800px+ from the nearest CTA.
• A logo strip in the header but nothing near the signup button.
• Trust badges grouped in the footer instead of by the checkout CTA.
• Proof only in a carousel halfway down the page.

What to ship

• Move your strongest testimonial or logo strip within 150px of the primary CTA.
• On landing pages, put a one-line micro-testimonial right under the button - name, role, specific result.
• On product pages, show "verified buyer" + a compact rating beside Add to Cart.
• At checkout, place trust seals next to the payment form, not the footer.

Typical lift: +24% to +68% by page type. Impact: high - the largest single-variable lift in social-proof research.

#3 - Vague proof instead of specific claims.

"Trusted by thousands of businesses worldwide" converts about the same as no proof at all - your visitors have read it on fifty other sites. Named, quantified claims ("Used by 8 of the Fortune 50") lift conversion ~22%; a single specific testimonial ~14%; a generic logo strip ~8%. Specificity is the variable.

What we see on scans

• "Trusted by thousands" with no number, names, or logos.
• Testimonials with first names only and no company, role, or photo.
• Logo strips of five unrecognisable brands.
• Case-study links that say "Read more" instead of stating the result.

What to ship

• Replace "trusted by thousands" with an exact number ("Used by 2,847 ecommerce teams").
• Add company, role, and a real photo to every testimonial - stock photos cut trust ~35%.
• Lead with the result: "Cut checkout abandonment 31% in two weeks" beats "Great product."
• Name marquee customers if you have them - it can outperform a generic logo grid severalfold.

Typical lift: +14% to +22% over vague proof. Impact: medium-high - compounds with placement.

#4 - Wrong format for the page.

Video testimonials can lift conversion 80% over text - but an autoplay video hero can cut conversion 7% by slowing the page. Format and placement have to match: compact text at the decision point, video mid-page after the value proposition, a logo strip for a fast credibility check up top, UGC (User-Generated Content — customer photos and videos) on product pages.

What we see on scans

• Text-only testimonials where video would convert better.
• An autoplay video above the fold tanking page speed.
• One testimonial format reused across product, landing, and checkout pages.
• No UGC or customer photos on product pages.

What to ship

• Use compact text testimonials next to CTAs (decision moments).
• Place video mid-page, after the value prop - never as the hero.
• Add UGC photos to product pages - they beat studio shots for trust.
• On B2B (business-to-business) pages, combine logos + text testimonials.

Typical lift: +35% to +84% when format matches context. Impact: medium - needs content production, compounds with positioning.

#5 - Perfect ratings that kill credibility.

Analysis of 20M+ product pages found products with a flawless 5.0 average convert about the same as ones rated 3.0-3.5. Buyers don't trust perfection; purchase likelihood peaks at 4.2-4.5 - the range that says "real people reviewed this and mostly loved it."

What we see on scans

• A 5.0 shown prominently on fewer than 10 reviews.
• No negative reviews visible - filtered or hidden.
• A star rating with no review count.
• No rating distribution bar.

What to ship

• Show the review count beside the rating ("4.3 stars, 847 reviews" beats "5.0 stars").
• Display the distribution bar - a natural curve builds trust.
• Don't hide negatives; a 1-star with a thoughtful reply converts better than none.
• Add "verified buyer" badges - verification beats perfection.

Typical lift: +8% to +15% moving from a thin 5.0 to a supported 4.2-4.5 with volume. Impact: medium - higher on high-ticket items.

#6 - No real-time proof at the decision moment.

Static testimonials say "someone liked this once." Real-time proof says "someone's buying this now." Purchase notifications lift conversion 10-15%; live viewer counts cut bounce ~21%. The combination of urgency and validation is one of the highest-ROI (Return on investment) elements you can add - if it's tied to real data.

What we see on scans

• No real-time proof on product or landing pages.
• Static carousels that auto-rotate and get ignored.
• "Popular" labels with no supporting data.

What to ship

• Add small, real purchase notifications on product pages.
• Show live viewer counts on high-traffic pages ("47 viewing").
• Replace static carousels with a live feed of recent reviews.
• Use low-stock indicators tied to actual inventory - real converts, fake backfires.
• Keep it subtle; aggressive pop-ups every few seconds trigger blindness.

Typical lift: +10% to +15% for notifications, up to +21% bounce reduction for live counters. Impact: medium - quick to add, compounds on traffic.

How Levri spots all six in 60 seconds.

Levri checks review visibility above the fold, proof proximity to the CTA, claim specificity, format match, rating health, and real-time signals - then ranks each by estimated revenue impact.

You paste the URL, and you get your fixes - a ranked list, each issue priced in $/mo, with a written hypothesis, a variant-B suggestion, and an expected lift range. No install, no tracking script, no "connect your analytics."

Fix these first.

In the order we'd ship - most are free and take minutes:

1. Move existing reviews/ratings above the fold.
2. Place your strongest proof beside the CTA.
3. Replace vague copy with specific, named claims.
4. Add real-time notifications.
5. Audit rating health and review volume.
6. Match proof format to the page and position.

What to show (and the trust psychology) is in trust signals that convert; proof at the price decision is in pricing page optimisation. Ship three and measure by Friday; if proof isn't moving conversion, it's in the wrong place, which is exactly what Levri is built to diagnose.