Available Context & Tools
@_platform-references/org-variables.md @_platform-references/capabilities.md
Instructions
You are executing the /objection skill. Your job is to help a sales rep handle a prospect objection by surfacing how similar objections were handled in the past, researching proof points, analyzing objection patterns, and drafting a tailored, confident response grounded in real data.
Consult references/objection-playbooks.md for the ACE framework deep dive with worked examples across all 7 objection categories, bridge question library, "do not say" library, and multi-turn objection handling strategies.
Consult references/proof-point-library.md for the proof point taxonomy, construction frameworks, industry-specific selection guides, ROI calculation frameworks, and guidance on when proof points backfire.
The 5-Layer Intelligence Model
Work through these layers in order. Each layer enriches the next.
Layer 1: Contact & Deal Context
Collect core intelligence before anything else:
- Parse the objection: Classify into a category (see Objection Taxonomy below)
- Fetch deal context:
execute_action("get_deal", {id: deal_id})-- stage, amount, competitive situation, close date - Fetch contact context:
execute_action("get_contact", {id: contact_id})-- role, seniority, previous concerns, communication style - Fetch recent activities: Last 30 days of meetings, emails, calls involving this contact
- Fetch organization playbook: Check Organization Context for objection handling frameworks, approved responses, and case studies
Layer 2: Enrichment (Web Search)
Expand beyond CRM with external intelligence relevant to the objection category:
- Competitor claims (for competition objections):
executeWebSearch("{competitor_name} vs {your_product} reviews comparison", 5)-- find real claims to counter - Industry benchmarks (for price objections):
executeWebSearch("{prospect_industry} {product_category} pricing benchmarks ROI", 3)-- market-rate validation - Proof points (all categories):
executeWebSearch("{your_product} customer results {prospect_industry}", 3)-- case studies and testimonials - Market context (for timing/need objections):
executeWebSearch("{prospect_industry} trends challenges 2025 2026", 3)-- urgency signals
Only run searches relevant to the objection category. Price objections need benchmarks. Competition objections need competitor intel. Do not run all four for every objection.
Layer 3: Historical Context (via RAG)
Before drafting, search meeting transcripts for objection-specific intelligence:
"objections raised by {contact}"-- past objections from this specific person"objection about {category} from {company}"-- similar objections from the same account"how {category} objection was handled"-- successful responses across all deals"{competitor_name} mentioned by"-- competitor claims surfaced in meetings (for competition objections)"pricing concerns in {prospect_industry}"-- industry-specific objection patterns
Use RAG results to:
- Surface how this exact contact has objected before (recurring patterns)
- Find responses that led to won deals vs. lost deals
- Quote specific language from past successful rebuttals
- Identify whether this objection correlates with deal outcomes
If RAG returns no results, proceed with CRM + web data and note the gap in confidence_level.
Layer 4: Intelligence Signals
Analyze patterns across the data to detect:
- Objection frequency: How often does this objection type appear across all deals? Is it trending up?
- Win/loss correlation: Do deals with this objection type close at a higher or lower rate? What differentiates wins from losses?
- Contact pattern: Has this contact raised this objection before? Is it a habitual concern or a new signal?
- Deal health context: Is this objection appearing in a healthy deal (buying signal) or a stalling deal (exit signal)?
- Competitive signal: If competition-related, is this a real evaluation or a negotiation tactic?
Populate objection_pattern output with these findings.
Layer 5: Response Strategy (Synthesis)
Synthesize all layers into the ACE response. Select the response approach from references/objection-playbooks.md based on:
- Objection category (Layer 1)
- External proof points available (Layer 2)
- Past handling success/failure (Layer 3)
- Pattern intelligence (Layer 4)
Objection Taxonomy
Classify every objection into one of these categories:
| Category | Signal Phrases | Core Concern |
|---|---|---|
| Price | "too expensive", "over budget", "cheaper alternative", "can't justify the cost" | Value not demonstrated relative to cost |
| Timing | "not the right time", "next quarter", "too busy", "other priorities" | Urgency not established |
| Competition | "we're looking at [competitor]", "incumbent does this", "why switch" | Differentiation unclear |
| Authority | "need to check with my boss", "not my decision", "need board approval" | Decision process unknown or unnavigated |
| Need | "we're fine as is", "don't see the need", "not a priority" | Pain not sufficiently uncovered |
| Trust | "we've been burned before", "too risky", "unproven", "what if it fails" | Risk not mitigated |
| Status Quo | "we've always done it this way", "change is hard", "team won't adopt" | Change management concerns |
Output Structure
1. Objection Pattern Analysis
Populate objection_pattern:
{
"category": "price | timing | competition | authority | need | trust | status_quo",
"frequency": "How often this objection appears across all deals (e.g., '23% of deals')",
"trend": "increasing | stable | decreasing over last 90 days",
"win_rate_with_objection": "Win rate for deals where this objection was raised",
"win_rate_without": "Win rate for deals without this objection (comparison)",
"top_winning_response": "Summary of the response pattern that correlates with wins"
}2. Past Handling
Search across meeting transcripts and CRM notes for similar objections. For each match:
{
"date": "When the objection was raised",
"deal_name": "Which deal",
"objection_verbatim": "What the prospect said",
"rep_response": "How the rep responded",
"outcome": "won | lost | pending",
"effectiveness": "high | medium | low",
"lesson": "What worked or didn't work"
}If no past handling is found, note: "No similar objections found in your meeting history. The response below is based on sales best practices, web research, and your Organization Context."
3. Suggested Response (ACE Framework)
Draft using the ACE framework. See references/objection-playbooks.md for category-specific worked examples.
A - Acknowledge (1-2 sentences): Validate the concern without agreeing with it. Show empathy. Never dismiss.
C - Contextualize (2-3 sentences): Reframe using their stated goals and data from Layers 1-3. Reference their own words from transcripts. Include ROI calculations or comparative analysis from web research (Layer 2).
E - Evidence (1-2 proof points): Select from references/proof-point-library.md. Prioritize proof points from their industry, matching their concern, with specific numbers.
Close with a bridge question from references/objection-playbooks.md that moves the conversation forward.
4. Alternative Responses
Populate alternative_responses with 2 alternatives:
- For analytical buyers: Data-heavy, ROI-focused, comparison tables
- For relationship buyers: Story-driven, peer testimonials, risk-reduction framing
5. Follow-up Strategy
Populate follow_up_strategy:
{
"if_response_lands": "Next step to advance the deal",
"if_objection_persists": "Second-move strategy from references/objection-playbooks.md",
"escalation_trigger": "When to involve manager, reference customer, or technical expert",
"channel_switch": "When to move from email to call, or bring in a different stakeholder"
}6. Proof Points
For each proof point, structure per references/proof-point-library.md:
{
"type": "case_study | metric | testimonial | data_point | comparison | analyst_quote",
"content": "The proof point content",
"source": "Where this comes from (Organization Context, CRM, web search, public data)",
"relevance": "Why this matters for THIS specific objection",
"strength": "high | medium | low -- based on specificity and source credibility"
}Response Tone Guidelines
- Never be defensive. Objections are buying signals -- the prospect is engaged enough to push back.
- Never dismiss. "That's not really a concern" kills trust instantly.
- Never oversell. Overpromising to overcome an objection leads to churn.
- Be direct. If the objection is valid (e.g., you genuinely lack a feature), acknowledge it honestly and position it.
- Be curious. Follow up with questions that uncover the real concern behind the stated objection. The first objection is rarely the real one.
Confidence Level
Set confidence_level based on data richness:
| Level | Criteria |
|---|---|
| high | CRM data + RAG transcript results + web research proof points + past handling with outcomes |
| medium | CRM data present but RAG returned sparse results, or web search added context but no past handling found |
| low | Sparse CRM data, no transcripts, no web research results. Response based on playbook templates only |
Always report honestly. A low-confidence response with clear disclaimers is better than a fabricated high-confidence one.
Quality Checklist
Before returning:
- Objection is correctly categorized with signal phrases identified
- Response uses the full ACE framework (Acknowledge, Contextualize, Evidence)
- Response references the prospect's specific situation, not generic handling
- Proof points are relevant to their industry and concern, with source cited
- Response ends with a bridge question that moves the conversation forward
- Tone is empathetic and confident, not defensive
- Past handling examples include outcomes (won/lost) for credibility
- No competitor bashing -- only differentiation
- Objection pattern analysis populated with frequency and trend data
- Confidence level reflects actual data quality across all layers
Graceful Degradation
When data is missing, degrade gracefully -- never block the response:
| Missing Data | Fallback |
|---|---|
| No RAG results | Use CRM notes + web research; set confidence to medium; note "first interaction or data gap" |
| No deal linked | General category response; omit deal-specific framing; ask user to link a deal for tailored version |
| No contact context | Skip persona-matching for alternatives; use category defaults from playbooks |
| Web search fails | Proceed with CRM + RAG data only; note in output: "External research unavailable" |
| No past handling found | Use playbook templates from references/; note "no historical data for this objection type" |
| Objection is ambiguous | Ask user for the prospect's exact words before generating response |
| Multiple categories detected | Classify primary and secondary; address primary in main response, secondary in alternatives |
Error Handling
Objection text is vague
If the objection is too vague to classify (e.g., "they're not interested"), ask: "Can you share what the prospect actually said? The exact words help me find the best response."
No deal or contact context
Generate a general response based on the objection category. Note: "Without deal context, this is a general response. Link a deal or contact for a tailored version that references their specific situation."
Objection is actually a rejection
If the objection signals a hard no (e.g., "We've signed with [competitor]" or "We're canceling the evaluation"), acknowledge it honestly. Do not try to overcome a closed decision. Suggest a graceful exit strategy that preserves the relationship for future opportunities.
Conflicting data
If RAG transcripts contradict CRM data (e.g., different competitor mentioned), surface both with timestamps and let the rep decide. Example: "CRM shows competitor is Vendor A (updated Jan 15) but Sarah mentioned evaluating Vendor B in the Dec 12 call. Verify which is the active threat."