Generate Price Objection Responses

Given a classified price objection, deal context, and historical win patterns, use the Claude API to generate a tailored response using the highest-success framework for this objection type.

Prerequisites

• Classified objection data (from call-transcript-objection-extraction)
• Deal record from Attio with pain data and deal value
• Anthropic API key
• Historical response effectiveness data (optional, improves quality over time)

API Call

POST https://api.anthropic.com/v1/messages
Authorization: x-api-key {ANTHROPIC_API_KEY}
Content-Type: application/json

Request body:

{
  "model": "claude-sonnet-4-20250514",
  "max_tokens": 2500,
  "messages": [{
    "role": "user",
    "content": "Generate a price objection response for a sales conversation.\n\nObjection details:\n- Prospect quote: \"{objection_quote}\"\n- Root cause: {root_cause}\n- Emotional tone: {emotional_tone}\n- Severity: {severity}/10\n- Stated budget: {stated_budget or 'not disclosed'}\n- Comparison anchor: {comparison_anchor or 'none'}\n\nDeal context:\n- Company: {company_name}\n- Deal value: ${deal_value}/year\n- Total quantified pain: ${total_quantified_pain}/year\n- Pain-to-price ratio: {ratio}x\n- Champion: {champion_name}, {champion_title}\n- Decision maker: {decision_maker_name}, {decision_maker_title}\n- Strongest pain: \"{strongest_pain_summary}\"\n- Key pain quote: \"{key_pain_quote}\"\n\nHistorical effectiveness (if available):\n{framework_effectiveness_json}\n\nGenerate the response using the most effective framework for this root cause. Return this exact JSON:\n{\n  \"framework_used\": \"value_reframe|roi_proof|payment_flexibility|competitive_tcm|anchor_to_pain|silence\",\n  \"verbal_response\": \"The exact words to say on a call (2-4 sentences max)\",\n  \"diagnostic_questions\": [\"2-3 follow-up questions to deepen understanding\"],\n  \"follow_up_email\": {\n    \"subject\": \"Email subject line\",\n    \"body\": \"Follow-up email body (under 150 words) reinforcing the verbal response with supporting data\"\n  },\n  \"supporting_asset\": {\n    \"type\": \"roi_calculator|case_study|tco_comparison|payment_options|business_case\",\n    \"description\": \"What to attach and why\"\n  },\n  \"escalation_path\": \"What to do if this response doesn't resolve the objection\",\n  \"expected_next_step\": \"The specific action to propose after delivering this response\"\n}"
  }]
}

Input Requirements

• objection_quote: The prospect's exact words (from objection extraction)
• root_cause: One of: no_budget, value_gap, competitor_comparison, sticker_shock, authority_gap, timing
• total_quantified_pain: From pain discovery data in Attio
• framework_effectiveness_json: Optional. Historical win rates per framework per root cause, pulled from PostHog aggregate data

Framework Selection Logic

When historical data is not available, use these default framework mappings:

| Root Cause | Primary Framework | Why | |-----------|------------------|-----| | no_budget | payment_flexibility | Restructure payment to fit budget cycles | | value_gap | roi_proof | Prove the math works — pain costs more than solution | | competitor_comparison | competitive_tcm | Total cost of ownership comparison including hidden costs | | sticker_shock | anchor_to_pain | Re-anchor from absolute price to cost of the problem | | authority_gap | value_reframe | Arm the champion with language for the decision maker | | timing | silence | Let the prospect talk — timing objections often mask deeper concerns |

When historical data IS available, override defaults with the highest win-rate framework for that root cause, provided sample size >= 10.

Output

JSON response object containing:

• Verbal response for live conversation
• Diagnostic questions to deepen understanding
• Follow-up email with supporting data
• Supporting asset recommendation
• Escalation path if initial response fails
• Expected next step to propose

Store in Attio as a note on the deal record. Log in PostHog as objection_response_generated event.

Guardrails

• Never generate responses that offer discounts as the first move — always lead with value
• Never promise features or timelines not confirmed by the product team
• If pain_to_price_ratio < 3, flag for the seller: the value story is weak — consider re-running discovery before responding to the price objection
• If emotional_tone is "firm" and severity >= 8, recommend a human-led response rather than templated — these need judgment
• Rate limit: max 5 response generations per deal per day (avoid spamming the same prospect with multiple angles)

Alternatives

| Tool | Method | Notes | |------|--------|-------| | Claude (Anthropic) | Messages API | Best structured output for nuanced responses | | GPT-4 (OpenAI) | Chat Completions API | Alternative LLM | | Gong Assist | Built-in AI | Real-time coaching during calls | | Clari | Deal intelligence | Predictive risk scoring but less response generation | | Manual | Framework reference sheet | Fallback for enterprise deals |