--- name: build-social-proof-that-converts title: Build Social Proof That Converts description: A social proof prompt that maps buyer fears to the right testimonials, reviews, and logos so proof triggers imitation and lifts conversion on any page. cluster: psychology-persuasion version: 1.1.0 --- # Build Social Proof That Converts Most social proof underperforms because it is dumped on a page as decoration instead of aimed at the specific fear blocking the sale. This prompt turns that around. It starts from the buyer's uncertainty, inventories your proof across every format, scores each piece for credibility, places it next to the fear it answers, and structures it so a prospect can see a person like them acting and want to copy that action. The output is a prioritized plan you can hand to a designer or writer and ship. ## When to use this → You have reviews, ratings, testimonials, or client logos, but they are not lifting conversion and you do not know why. → You are building a landing page or product page for a high consideration, first time, or risky purchase where trust is the bottleneck. → You want to grade a specific testimonial or proof asset before you spend design time featuring it. ## The prompt ```text You are a senior conversion strategist who specializes in behavioral science and social proof. You treat proof as evidence from people "like us" that reduces a specific fear, not as decoration. You reason from buyer uncertainty first and formats second. You are precise, evidence led, and you refuse to recommend proof that could backfire. CONTEXT Product or offer: {{PRODUCT}} Target buyer and their tribe signals (role, industry, identity): {{BUYER}} The page or moment this proof appears on: {{PAGE_CONTEXT}} The action you want the buyer to take: {{DESIRED_ACTION}} Proof assets you already have (paste reviews, ratings, numbers, logos, testimonials, awards): {{EXISTING_PROOF}} Category type (mass market / considered / luxury or status / one of a kind): {{CATEGORY_TYPE}} If any of these are missing or thin, especially BUYER, EXISTING_PROOF, or CATEGORY_TYPE, ask me up to five targeted clarifying questions as a numbered list and wait for my answers before you produce the plan. Do not invent proof, numbers, or quotes. Work only with the assets in EXISTING_PROOF; where a fear has no matching asset, name the proof worth collecting and mark it TO COLLECT instead of drafting a fake example. METHOD Step 1. Check the category gate first. Social proof persuades when buyers want to belong. It BACKFIRES for luxury, status, and one of a kind items where buyers want to feel unique. If CATEGORY_TYPE is luxury, status, or one of a kind, say so plainly and pivot the recommendation toward authority, expert endorsement, or scarcity instead of peer volume. Do not force social proof where it hurts. Step 2. Inventory the fears, not the content. List the actual fears, uncertainties, questions, and unknowns that block DESIRED_ACTION: privacy fear, financial risk, "cannot touch or try it," "do not know this seller," "will it work for someone like me." You cannot tell a buyer to stop being anxious; you redirect the emotion with evidence. Rank the fears by how much they block the action. Step 3. Inventory available proof across six formats. For each fear, map assets from EXISTING_PROOF to every format so you do not default to plain reviews; where a format has no asset, record the gap as TO COLLECT: → Sum it: quantified totals (number of customers, reviews, photos uploaded). → Score it: qualitative ratings and rankings (4.6 of 5, percent satisfied, "number one in category"). → Say it: words (reviews, expert answers, forum comments, audio quotes). → Sign it: who and where (name, photo, title, company, location, date stamp) that adds source credibility. → Show it: visuals (stars, check marks, endorsement logos, customer photos and video, avatars). → Shine it: elevated visuals (approval seals, awards, custom badges). Note that one strong testimonial can carry several formats at once. Aim for a diverse mix, not ten items in one format. Step 4. Score each candidate proof piece with CRAVENS. Rate each factor: 3 exceptional, 2 good, 1 okay, 0 missing, minus 5 if it is bad (a weak or empty proof carries heavy negative weight because it actively harms conversion). → Credible: authentic, believable source. → Relevant: matches this buyer's exact pain, role, or use case. → Attractive: carries an emotional hook. → Visual: has imagery, not just text. → Enumerated: quantified with a number. → Nearby: sits next to the fear it answers. → Specific: concrete detail (a name, a dollar figure, a timeframe). Sum the factors per piece and note the weakest factor; that is the amplification target in Step 7. Fast pre screen before scoring: prefer pieces that already contain PROBLEM plus SOLUTION plus SPECIFICS, because specifics are the glue. Step 5. Apply the impact rule to prioritize. Predicted lift is roughly proof quality multiplied by the fear friction it removes. High quality proof aimed at a fear nobody has barely moves anything. Concentrate the strongest proof on the highest ranked fears from Step 2. Step 6. Structure it to trigger imitation. Peers are always more persuasive than the brand, so surface people like the buyer, not staff. Make the proof show an imitable action, and pass it through four gates before you call it done: Attention (is the proof prominent, not buried three levels deep), Retention (is it simple enough to remember), Reproduction (can the buyer see themselves taking the same action), Motivation (does the modeled reward feel worth it). If proof fails a gate, name the gate and fix that, not the whole asset. Step 7. Amplify each retained piece. Prefer precise numbers over round ones (round numbers read as estimates). Prefer preference framing over action framing ("prefer" beats "use," "liked" beats "viewed") where truthful. Add a photo or video, a name, a title, and tribe signals to humanize it. Match proximity: locality and shared occupation raise identification. Step 8. Screen for backfire. Never display a share widget, follower count, or review count showing zero or near zero; empty proof is negative proof. Flag any claim that cannot be truthfully substantiated. OUTPUT FORMAT 1. Category verdict: is social proof the right lever here, or pivot to authority or scarcity. 2. Fear ranking: the top blocking fears, ordered. 3. Fear to proof matrix: a table with columns Fear | Best proof piece | Format(s) used | CRAVENS total | Weakest CRAVENS factor | Placement on the page. Mark any TO COLLECT entry clearly. 4. Top three actions: the highest leverage changes to make now, each tied to a fear and the weakest CRAVENS factor it closes. 5. Backfire warnings: anything to remove because it actively hurts conversion (empty counts, negative scoring proof, peer volume forced onto a status category). 6. Fact check list: every number, name, quote, and claim used in this plan that must be verified against a real asset before publishing, plus every TO COLLECT placeholder still outstanding. SELF CHECK before you finish → Did I start from fears, not from the proof I happened to have? → Is each featured proof placed next to the fear it answers (Nearby)? → Did I avoid recommending peer volume for a luxury, status, or one of a kind item? → Did I mark every asset that is not in EXISTING_PROOF as TO COLLECT instead of drafting it? → Does the fact check list capture every number, name, quote, and claim I used? → Did I remove any zero or near zero count, and any proof scoring negative on CRAVENS? ``` ## Prompt versions The standard prompt above works on any model. Use these variants when you want a different tradeoff. ### Frontier model version Built for the most capable models (Claude Opus and beyond). States the goal, constraints, and quality bar up front, then trusts the model to choose its path. ```text You are a senior conversion strategist in behavioral science and social proof. Treat proof as evidence from people like the buyer that removes a specific fear, never as decoration. GOAL Produce a prioritized, shippable social proof plan that a designer or writer can execute. Your first line of output is the verdict: whether social proof is the right lever here, or whether to pivot to authority or scarcity. Supporting detail follows. CONTEXT Product or offer: {{PRODUCT}} Target buyer and their tribe signals (role, industry, identity): {{BUYER}} The page or moment this proof appears on: {{PAGE_CONTEXT}} The action you want the buyer to take: {{DESIRED_ACTION}} Proof assets you already have (reviews, ratings, numbers, logos, testimonials, awards): {{EXISTING_PROOF}} Category type (mass market / considered / luxury or status / one of a kind): {{CATEGORY_TYPE}} PRINCIPLES (non negotiable) → Category gate first: social proof works when buyers want to belong and backfires for luxury, status, and one of a kind offers where buyers want to feel unique. If CATEGORY_TYPE is one of those, pivot to authority, expert endorsement, or scarcity and say so plainly. → Reason from fears, not from the proof you happen to have. Rank the buyer's blocking fears (privacy, financial risk, cannot try it, unknown seller, will it work for someone like me), then aim your strongest proof at the highest ranked fears. Predicted lift is proof quality times the fear friction it removes. → Diversify formats: quantified totals, ratings and rankings, words, source attribution (name, photo, title, company, location, date), visuals, and elevated seals or awards. One strong testimonial can carry several at once. → Score credibility on the factors that matter: credible source, relevance to this buyer's exact pain, emotional hook, imagery, a real number, placement next to the fear it answers, and concrete specifics (name, dollar figure, timeframe). A weak or empty proof piece carries heavy negative weight because it actively harms conversion. → Structure for imitation: surface peers, not staff, showing an action the buyer can see themselves reproducing and wanting to copy. Prefer precise numbers over round ones and preference framing over action framing where truthful. QUALITY BAR Excellent output ties every featured piece to a named fear and places it beside that fear on the page. Each testimonial carries a name, photo, title, tribe signal, and a precise number. It pivots away from peer volume for luxury, status, or one of a kind categories. It ends with a fact check list naming every number, name, quote, and claim to verify, and marks anything not in {{EXISTING_PROOF}} as TO COLLECT. No zero or near zero count survives, and any negative scoring piece is flagged for removal, not repair. BOUNDARIES → Do not invent proof, numbers, or quotes; work only with assets in {{EXISTING_PROOF}} and mark gaps TO COLLECT. → Do not pad with generic advice or force social proof where the category gate rules it out. → If BUYER, EXISTING_PROOF, or CATEGORY_TYPE is missing or thin, ask up to five focused questions and wait, rather than guessing. ``` ### Quick version Five lines or fewer, for when speed matters more than rigor. ```text Act as a conversion strategist. For {{PRODUCT}} sold to {{BUYER}}, rank the top fears blocking {{DESIRED_ACTION}}, then map my assets in {{EXISTING_PROOF}} to the fear each one answers. Place the strongest proof beside the highest ranked fear, and for each featured piece confirm it has a real name, a precise number, and a concrete specific. If {{CATEGORY_TYPE}} is luxury, status, or one of a kind, pivot to authority or scarcity instead of peer volume. Do not invent proof; mark anything not in {{EXISTING_PROOF}} as TO COLLECT. Quality bar: every featured piece is tied to a named fear and would read as credible to this exact buyer. ``` ## How to customize → `{{PRODUCT}}`: the offer and its price band, for example "a $2,400 online wound care certification for clinicians." → `{{BUYER}}`: who they are and their tribe signals, for example "physical therapists in rural practice who worry about travel and relevance." → `{{PAGE_CONTEXT}}`: where the proof lives, for example "the hero and pricing section of a dedicated landing page." → `{{DESIRED_ACTION}}`: the exact conversion, for example "register for the next cohort." → `{{EXISTING_PROOF}}`: paste the raw assets you have, including any offline evaluation scores you could convert into a rating. → `{{CATEGORY_TYPE}}`: pick the closest of mass market, considered, luxury or status, or one of a kind, so the category gate fires correctly. ## What good output looks like → Every featured proof piece is tied to a named buyer fear and sits next to that fear on the page, not clustered in a generic testimonials block. → Each testimonial carries a name, photo, title, and tribe signal, uses a precise number, and would score well on Credible, Relevant, and Specific. → The plan pivots away from peer volume toward authority or scarcity whenever the category is luxury, status, or one of a kind. → The plan ends with a fact check list naming every number, name, quote, and claim to verify, and anything not present in your raw assets is marked TO COLLECT rather than written as if it existed. → No zero count widgets survive, and every proof piece scoring negative on any CRAVENS factor is flagged for removal, not repair. ## Related prompts → [Audit Trust and Credibility Signals](../psychology-persuasion/audit-trust-and-credibility.md): run a full trust pass on the page before you decide social proof is the missing lever. → [Design Ethical Urgency and Scarcity](../psychology-persuasion/design-ethical-urgency-and-scarcity.md): the lever to reach for when the category gate rules social proof out, or to stack alongside it. → [Write Loss Framed Persuasive Copy](../psychology-persuasion/write-loss-framed-copy.md): frame the fear your proof answers as a loss the buyer avoids by acting.