Loading...
Loading...
When the user wants to build or improve a sales bot's ability to dynamically insert relevant testimonials and case studies. Also use when the user mentions "social proof," "testimonials," "case studies," "customer stories," or "reference injection."
npx skill4agent add louisblythe/salesskills social-proof-injectionStatic approach:
"Companies like Microsoft and Google use us!"
Prospect thinking:
"I'm not Microsoft. We're a 50-person startup."
Not relevant = not compelling.Matched approach:
"[Similar-size SaaS company] saw 40% improvement.
They had the same challenge you mentioned."
Prospect thinking:
"They're like us. If it worked for them..."
Relevant = believable = compelling.Direct quotes from customers:
"We cut our sales cycle by 30%."
- Sarah Chen, VP Sales at TechCorp
Best for:
- Emotional resonance
- Credibility from real person
- Quick impactDetailed customer stories:
TechCorp: 500-person SaaS company
Challenge: Long sales cycles
Solution: Implemented [product]
Result: 30% faster close, $2M additional revenue
Best for:
- Analytical buyers
- Similar situations
- Proving methodologyAggregate data:
"Our customers average 3.2x ROI"
"92% of users see results in 30 days"
"4.8/5 satisfaction rating"
Best for:
- Building credibility
- Supporting claims
- Analytical prospectsVisual credibility:
"Trusted by 500+ companies including..."
[Logo] [Logo] [Logo]
Best for:
- Establishing legitimacy
- Enterprise sales
- Aspirational alignmentReal-time social proof:
"15 companies signed up this week"
"Sarah from TechCorp is also exploring this"
"Most popular choice for teams your size"
Best for:
- FOMO creation
- Validating popularity
- Reducing risk perceptiondef select_social_proof(prospect):
context = {
"industry": prospect.industry,
"company_size": prospect.employees,
"role": prospect.title,
"use_case": prospect.stated_needs,
"stage": prospect.sales_stage,
"pain_points": prospect.pain_points,
"objections": prospect.objections_raised
}
# Find matching proof
candidates = get_social_proof_library()
scored = []
for proof in candidates:
score = calculate_relevance(proof, context)
scored.append((proof, score))
# Return top matches
return sorted(scored, key=lambda x: -x[1])[:3]def calculate_relevance(proof, context):
score = 0
weights = {
"industry_match": 30,
"size_match": 25,
"role_match": 20,
"use_case_match": 15,
"pain_point_match": 10
}
# Industry match
if proof.customer_industry == context["industry"]:
score += weights["industry_match"]
elif proof.customer_industry in related_industries(context["industry"]):
score += weights["industry_match"] * 0.5
# Size match
size_diff = abs(proof.customer_size - context["company_size"])
size_factor = max(0, 1 - (size_diff / context["company_size"]))
score += weights["size_match"] * size_factor
# Role match
if proof.quote_role_level == get_role_level(context["role"]):
score += weights["role_match"]
# Use case match
use_case_overlap = len(set(proof.use_cases) & set(context["use_case"]))
score += weights["use_case_match"] * min(1, use_case_overlap / 2)
# Pain point match
if any(pain in proof.pain_points_addressed for pain in context["pain_points"]):
score += weights["pain_point_match"]
return scoreEarly in sequence, establish credibility:
"[Similar company] was dealing with the same
[pain point] before they found us. Happy to
share what worked for them."
Inject: Similar customer reference, early.When they reveal a challenge:
"You mentioned [pain point]. [Customer name]
told us the same thing—here's what they did
about it: [brief summary/link]"
Inject: Directly relevant case study.When they raise concerns:
Objection: "We tried something similar before"
Response: "[Customer] had the same concern.
They'd been burned by [competitor]. Here's
what made the difference: [specific detail]"
Inject: Proof that overcomes specific objection.When close to deciding:
"Teams like yours typically see [result].
Here's a 2-minute video from [Customer VP]
explaining their experience."
Inject: High-impact proof for final push.class SocialProofLibrary:
def __init__(self):
self.proofs = []
def add_proof(self, proof):
self.proofs.append({
"id": generate_id(),
"type": proof["type"], # testimonial, case_study, metric
"content": proof["content"],
"customer": {
"name": proof["customer_name"],
"industry": proof["industry"],
"size": proof["company_size"],
"logo": proof["logo_url"]
},
"quote_source": proof.get("quote_source"),
"metrics": proof.get("metrics", {}),
"pain_points": proof.get("pain_points", []),
"use_cases": proof.get("use_cases", []),
"objections_addressed": proof.get("objections", []),
"media": proof.get("media"), # video, pdf, link
"verified": proof.get("verified", False),
"recency": proof.get("date")
})
def find_matches(self, context, limit=3):
scored = []
for proof in self.proofs:
score = calculate_relevance(proof, context)
if score > MINIMUM_RELEVANCE_THRESHOLD:
scored.append((proof, score))
return sorted(scored, key=lambda x: -x[1])[:limit]def inject_social_proof(message, prospect, injection_point):
# Find matching proof
matches = social_proof_library.find_matches(
context=get_prospect_context(prospect),
limit=1
)
if not matches:
return message # No relevant proof available
proof = matches[0][0]
# Format based on type and context
if proof["type"] == "testimonial":
injection = format_testimonial(proof)
elif proof["type"] == "case_study":
injection = format_case_study_teaser(proof)
elif proof["type"] == "metric":
injection = format_metric(proof)
# Insert at appropriate point
return insert_into_message(message, injection, injection_point)
def format_testimonial(proof):
return f'''
"{proof['content']}"
— {proof['quote_source']['name']}, {proof['quote_source']['title']} at {proof['customer']['name']}
'''
def format_case_study_teaser(proof):
return f'''
{proof['customer']['name']} ({proof['customer']['industry']}, {proof['customer']['size']} employees)
saw {proof['metrics']['headline_result']} after implementing our solution.
[See the full story]({proof['media']['link']})
'''def select_proof_for_objection(objection, prospect):
# Map objection to proof need
objection_mapping = {
"too_expensive": "roi_focused_proof",
"bad_past_experience": "similar_skeptic_success",
"competitor_using": "competitor_switch_story",
"implementation_concern": "fast_implementation_proof",
"team_adoption": "high_adoption_story"
}
proof_need = objection_mapping.get(objection.type, "general_success")
# Find proof that addresses this objection
matches = social_proof_library.find_matches(
context={
"objection_type": objection.type,
"industry": prospect.industry,
"size": prospect.employees
}
)
return matches[0][0] if matches else NoneStrong proof:
- Specific metrics ("40% improvement")
- Named customer (with permission)
- Relevant to prospect's situation
- Recent (last 2 years)
- Verifiable
Weak proof:
- Vague ("great results")
- Anonymous ("a customer said")
- Irrelevant industry/size
- Outdated
- UnverifiableDon't overdo it:
- 1 proof per message max
- 2-3 proofs per conversation
- Vary the type
- Make it natural
Too much proof = desperate.
Right amount = credible.BAD:
"Buy our product. Also, Company X likes us.
And Company Y. And here's a case study.
And a testimonial."
GOOD:
"You mentioned [pain]. That's exactly what
[Company] was dealing with. They found that
[solution] made a big difference. [Brief result].
Happy to share more if helpful."
Integrate, don't dump.Track:
- Response rate when proof included
- Conversion rate by proof type
- Which proofs get clicked/engaged
- A/B test proof vs no proof
Optimize:
- Which industries respond to which proof?
- What format works best?
- Which customers make best references?