Phase 4: Trial/Hand-Raise

ExpertCloneTrial Component (React): <ExpertCloneTrial expertId="brad-himel" expertName="Brad Himel" playbookId="newsletter-growth" userId={user.id} freeQuestions={3} trialContext={{ completedSteps: 6, emailCaptured: true, userStage: "scaling", userRevenue: "$10K-$50K" }} /> Component Features: - AI chat interface styled like expert (avatar, voice, examples) - Question counter: "3 free questions remaining" - Contextual prompt injection (playbook methodology + user data) - Usage tracking on every Q&A interaction - Paywall after 3-5 questions: "Want more? Watch 5-min demo"

Backend API Endpoint (FastAPI): @app.post("/api/trial/ask") async def trial_ask_question( user_id: str, expert_id: str, playbook_id: str, question: str, trial_context: dict ): # Check trial limit trial = await get_or_create_trial(user_id, playbook_id) if trial.questions_used >= trial.questions_allowed: return {"error": "Trial limit reached", "hand_raise": True} # Build context-aware prompt prompt = f"""You are {expert_id}, expert in {playbook_id}. User context: - Completed 6 steps of your methodology - Stage: {trial_context['userStage']} - Revenue: {trial_context['userRevenue']} Answer their question using your methodology from the playbook: {question}""" # Call Claude API response = await anthropic_api_call(prompt) # Log usage await supabase.table('trials').update({ 'questions_used': trial.questions_used + 1, 'last_question_at': datetime.now() }).eq('id', trial.id).execute() await supabase.table('trial_interactions').insert({ 'trial_id': trial.id, 'question': question, 'response': response, 'question_number': trial.questions_used + 1 }).execute() return { "response": response, "questions_remaining": trial.questions_allowed - trial.questions_used - 1 }

Paywall Copy Template: 🔒 Trial Complete — Want Unlimited Access? You've used your 3 free questions with [Expert Name]. Ready to see how [Expert] AI can transform your [specific use case]? 📹 Watch this 5-minute demo showing: • How [Expert] AI applies methodology to YOUR business • Real examples from businesses like yours • Pricing: $[X]/mo for unlimited questions [BUTTON: Watch 5-Min Demo → ] No credit card required for demo.

Trials Table (Phase 4 Extensions): CREATE TABLE trials ( id UUID PRIMARY KEY, user_id UUID REFERENCES leads(id), playbook_id UUID REFERENCES playbooks(id), expert_id UUID REFERENCES experts(id), -- Trial limits questions_allowed INT DEFAULT 3, questions_used INT DEFAULT 0, -- Timing started_at TIMESTAMP DEFAULT NOW(), last_question_at TIMESTAMP, completed_at TIMESTAMP, -- Hand-raise tracking hand_raise_triggered BOOLEAN DEFAULT FALSE, hand_raise_trigger_type VARCHAR(50), -- 'limit_reached', 'high_engagement', 'explicit_request' hand_raise_at TIMESTAMP, -- Conversion tracking loom_sent BOOLEAN DEFAULT FALSE, loom_sent_at TIMESTAMP, loom_watched BOOLEAN DEFAULT FALSE, loom_watched_at TIMESTAMP, -- Metadata utm_source VARCHAR(100), utm_campaign VARCHAR(100) ); Trial Interactions Table: CREATE TABLE trial_interactions ( id UUID PRIMARY KEY, trial_id UUID REFERENCES trials(id), question_number INT, question TEXT, response TEXT, response_length INT, -- Quality signals user_rating INT, -- 1-5 stars (optional) follow_up_question BOOLEAN DEFAULT FALSE, created_at TIMESTAMP DEFAULT NOW() );

Hand-Raise Detection (FastAPI Background Task): async def detect_hand_raise(trial_id: str): trial = await supabase.table('trials').select('*').eq('id', trial_id).single() if trial.hand_raise_triggered: return # Already triggered # Trigger 1: Limit Reached if trial.questions_used >= trial.questions_allowed: await trigger_hand_raise(trial_id, 'limit_reached') return # Trigger 2: High Engagement interactions = await supabase.table('trial_interactions')\ .select('*').eq('trial_id', trial_id).execute() time_spent = (datetime.now() - trial.started_at).total_seconds() / 60 follow_ups = sum(1 for i in interactions if i.follow_up_question) if (trial.questions_used >= 3 and time_spent >= 15 and follow_ups >= 2): await trigger_hand_raise(trial_id, 'high_engagement') return # Trigger 3: Explicit Request (detected in question content) last_question = interactions[-1].question.lower() trigger_phrases = ['how do i get more', 'can i book', 'pricing', 'want to buy', 'interested in', 'sign up'] if any(phrase in last_question for phrase in trigger_phrases): await trigger_hand_raise(trial_id, 'explicit_request') async def trigger_hand_raise(trial_id: str, trigger_type: str): await supabase.table('trials').update({ 'hand_raise_triggered': True, 'hand_raise_trigger_type': trigger_type, 'hand_raise_at': datetime.now() }).eq('id', trial_id).execute() # Trigger Phase 5: Loom generation await generate_personalized_loom(trial_id)

Conversion Readiness Score Calculation: Base Score (Trial Behavior): - Questions used: 10 points per question (max 50) - Time spent: 1 point per minute (max 20) - Follow-up questions: 10 points each (max 20) Lead Quality Multiplier (from Phase 3): - Scaling stage + $10K+ revenue: 1.5x - Building stage + any revenue: 1.2x - Just starting: 1.0x Engagement Bonus: - Completed all playbook steps: +15 points - Watched Loom demo: +25 points - Replied to nurture email: +10 points - High engagement hand-raise: +15 points Example Calculations: User A: 3 questions (30) + 18 min (18) + 2 follow-ups (20) = 68 × Scaling + $10K+ (1.5x) = 102 points + High engagement (15) = 117 points → HOT LEAD User B: 2 questions (20) + 8 min (8) = 28 × Just starting (1.0x) = 28 points → COLD User C: 5 questions (50) + 25 min (20) + 3 follow-ups (20) = 90 × Building + revenue (1.2x) = 108 points + Completed steps (15) = 123 points → URGENT Score Tiers: - 0-40: Cold (nurture sequence) - 41-70: Warm (standard Loom) - 71-90: Hot (personalized Loom) - 91-100: Urgent (immediate outreach + custom Loom)

Loom Trigger Workflow (n8n): 1. Trigger: Hand-raise detected (Supabase webhook) Event: trials table hand_raise_triggered = TRUE 2. Fetch Context: - Trial interactions (questions + responses) - Lead data (stage, revenue, challenge) - Playbook completion status - UTM attribution data 3. Calculate Conversion Score: - Run scoring algorithm - Determine Loom type (standard/personalized/custom) 4. Branch by Score: IF score >= 91: → Custom Loom (manual, notify Jason) ELSE IF score >= 71: → Personalized Loom (AI-generated, high quality) ELSE: → Standard Loom (template-based) 5. Generate Loom (Phase 5): - Create script using trial context - Record or auto-generate video - Upload to Loom - Get shareable link 6. Send Loom: - Email to user - SMS notification (optional) - Slack alert to Jason 7. Track Delivery: - Update trials.loom_sent = TRUE - Log loom_id and loom_url - Start watch tracking