Social proof that converts
Last updated:
Proof converts when it is specific, relevant to the buyer on this page, and sitting next to the claim it supports. A logo strip three scrolls below the fold does not rescue a weak hero. Do not invent customers. Move one real metric, named quote, or honest logo into the decision band. See checklist Proof and the sample missing-proof wireframe.
What good looks like
- One specific proof in the decision band — a named quote, a metric with a denominator, or a recognizable logo the buyer would know.
- Proof type matches the segment. Enterprise logos on a $49 self-serve page need a qualifier (“also used by…”), or they feel borrowed.
- Near the claim. If the H1 promises a 60-second grade, the proof is about speed or the artifact — not a vague “trusted by teams.”
- No placeholders. Empty logo slots, lorem quotes, and “Coming soon” on evidence are anti-proof.
- Honest labeling. Illustrative or anonymized quotes are marked; stock photos are not presented as customers.
SiteRevenue’s own sample is labeled illustrative. Public pages should be at least that honest.
Common gotchas
- “Trusted by teams” with no names, numbers, or logos — a claim that asks for trust instead of giving it.
- Proof three scrolls down while the primary CTA sits in an empty band.
- Borrowed parent or partner logos presented as your customers.
- Vanity metrics with no timeframe, population, or definition (“10× pipeline”).
- Stock testimonials or invented names. If you would not put it in a contract, do not put it next to the buy button.
- Autoplay video quote that steals the tap target and the first seconds of a paid click.
- “As seen in” with no link to a real article, or press about a different product.
How SiteRevenue scores it
New full jobs emit a separate Proof dimension (0–10, higher = healthier) — not folded into Trust on the public scorecard. Letter-grade math still uses engine key trust (no 9th engine key). Older jobs stay omit-safe and may fold proof notes into Trust. The sample scores Proof 6/10: logos exist, but first proof sits three scrolls down. Findings may call out thin proof near the CTA or borrowed logos.
The engine only sees public HTML you submit. It cannot verify that a quote is a real customer — you must not invent one. We do not publish named customer case studies we cannot stand behind, and the audit will not fabricate yours.
Teaser = titles only. Full report adds the scorecard, captures, and ranked findings (typically 5–8 — a quality target, not a refund SLA).
Ship this week
| Change | Effort | Why |
|---|---|---|
| Move one real quote + logo under the primary CTA | S (~30 min) | Trust in the decision band |
| Delete placeholder logos and lorem quotes | S | Anti-proof is worse than no proof |
| Add a qualifier on mismatched logos (“Used by agencies and…”) | S | Relevance without implying a fake segment |
| Replace a vanity number with a defined metric or cut it | S | Specific beats impressive-and-empty |
Do not wait on a case-study project to fix the fold.
Related guides
- Conversion leak checklist — Proof items 13–15
- Trust and risk reversal — policies and guarantees
- Hero clarity — the claim proof must support
- Offer and pricing clarity
- What is a website revenue audit?
- All guides
Start with a free grade
Paste the public URL on the home page. Compare your decision band to the sample missing-proof wireframe. Unlock the $19 page file when you want rank + evidence. Pricing.