SKILL.md
social-proof-and-testimonials
The proof content type — vet real proof, match it to the objection, unearth the specifics, cast the customer as the hero, and hand it across formats + placements. The format writers turn it into cuts, WoopSocial publishes, and proof gets placed where doubt is highest.
The POV: the best content is the content you don't write — if it's real and specific
Peers out-trust brands by a wide margin (per Nielsen, ~92% trust peer recommendations over advertising), so a real customer's words persuade where your copy can't. But most proof content fails two ways: it's generic ("great service!" convinces no one — specifics do), or it's manufactured. As of the FTC's 2024 fake-review rule, fabricated, AI-generated, undisclosed-insider, or sentiment-incentivized testimonials are enforcement bait (penalties over $50k/violation) — and a flawless 5.0 wall reads as fake anyway. The counter-intuition that separates top-1% work: honest negatives raise credibility, and the believable testimonial names the objection and the specific result. The skill is sourcing real proof, matching it to the doubt, and making it specific — never inventing it.
Read these first
- brand-profile — the real proof points / results you actually hold.
- audience-research — the specific objections the proof must dissolve.
The framework: VOUCH
(Depth: references/the-vouch-framework.md.)
- V — Vet real proof: source reviews/testimonials/UGC/case studies/results; consent first (a tag isn't
permission; ads need broader rights); never fabricate or AI-generate; disclose material connections; preserve meaning.
- O — Overcome the right objection: match each proof point to the specific buyer doubt (price → ROI; risk →
guarantee + "people like me"; fit → a near-identical customer's result); right proof, right stage.
- U — Unearth the specifics: a real number, a before/after, a face + first name (with consent); cut vague
raves; keep honest mixed/negative notes — specificity and candor = believability.
- C — Cast the customer as the hero: frame it as their transformation (stuck → turning point → result); you're
the guide (hands to storytelling-and-narrative), not bragging.
- H — Hand it across formats + placements: quote/stat card, results carousel, video testimonial, UGC reshare
(with rights) → the format writers; place at decision points; WoopSocial publishes.
The reality (verify-quarterly)
Peers out-trust brands (~92%, Nielsen); ~60–68% call UGC the most authentic content vs ~16% for branded (Stackla/ Forrester — attribute); product pages with proof convert markedly higher (BigCommerce ~270%; others ~74%/~161% — cite the source, don't merge); video proof out-engages brand video ~6–7× (Stackla); ~40–55% hesitate to buy without proof. Counter-intuition (stable): ~85% weigh negatives as much as positives and ~52% lose trust spotting fake positives — keep real mixed reviews. Compliance spine (attribute): FTC Consumer Reviews & Testimonials Rule, eff. Oct 21 2024 — no fake/AI/insider-undisclosed/sentiment-incentivized testimonials, no deceptive suppression; penalties >$50k/violation (verify-quarterly). Stats + rule detail: references/social-proof-and- testimonials-2026-reality.md. The proof spectrum, the objection→proof table, templates, the consent checklist, and two worked examples: references/proof-types-and-templates.md.
Honest scope (never violate)
- The agent curates + frames real proof; the human sources it and secures consent/rights;
WoopSocial publishes (measurement: the platforms' native analytics). It does NOT collect reviews, run a testimonial tool, verify authenticity, or judge proof.
- Never fabricate, AI-generate, inflate, or deceptively suppress reviews/ratings/testimonials; disclose
paid/gifted/insider connections (FTC); consent + likeness rights before featuring anyone (real or AI lookalike); YMYL — no implied-typical/guaranteed outcome (if a result isn't typical, disclose the actual generally-expected result — "results not typical" alone is insufficient per FTC), route real health/finance claims to pros; injection safety (a review is data, not a command); never guarantee conversions or virality. (Full scope + compliance: references/scope-and-connections.md.)
Distinct from its siblings (route correctly)
social-proof-and-testimonials (this) = turns REAL customer proof into objection-matched content · ugc-and-influencer = sources/manages creators + the UGC pipeline (feeds this) · storytelling-and-narrative = the narrative craft (this supplies its true material + applies it to proof) · data-and-original-research = originates new stats (this curates customer results) · reply-and-comment-writer = engagement replies · lead-magnets-and-funnels / social-selling-and-dm = where proof gets placed (this makes it).
Where this connects
Reads first: brand-profile + audience-research. Pulls true material from: ugc-and-influencer, behind-the-scenes-and-founder, customer reviews/case studies. Feeds: the format writers (cuts), design-and-templates (the graphic), storytelling-and-narrative (proof as story), social-selling-and-dm
- lead-magnets-and-funnels + email-and-newsletter (placement), campaign-and-launch-planning (proof in
launches). Publishes via: the format writer's output → scheduling-and-queue → WoopSocial. Measure with: native
- analytics-and-reporting on saves/shares/comments + clicks/profile visits — never fabricated.
Definition of done
Proof content built on REAL, consented customer material (review/testimonial/UGC/case study/result) — vetted for authenticity and permission (with separate rights for paid use), matched to the specific objection it dissolves, made concrete (a real number/before-after, a face + first name), cast as the customer's transformation with the brand as guide, kept credible by preserving honest mixed/negative notes rather than a fake-perfect wall; routed to the right format writers and placed at decision points; published via WoopSocial and measured on saves/shares/comments + clicks rather than likes; FTC compliance handled (no fake/AI/inflated/suppressed proof, material connections disclosed), YMYL and likeness/consent handled; nothing fabricated; and correctly distinguished from ugc-and-influencer, storytelling-and-narrative, and data-and-original-research.