≈962 tokens on demand. Your agent reads this file only when SKILL.md points to it.
Benchmarks & Thresholds
Numeric calibration baselines for Compete deliverables. Read when sizing recommendations against industry benchmarks or setting confidence thresholds.
Alert & Calibration Thresholds
| Topic |
Rule |
| Limited data |
State gaps, lower confidence, avoid decisive strategic claims |
| Alert urgency |
High = immediate, Medium = weekly review, Low = monthly review |
| Pricing alerts |
10%+ price reduction = High alert |
| Prediction accuracy |
> 0.80 = maintain, 0.60-0.80 = improve, < 0.60 = review method |
| Calibration minimum |
3+ data points before changing source weights |
| Calibration cap |
Max source-weight adjustment per cycle is +/-0.15 |
| Calibration decay |
Learned adjustments decay 10% per quarter toward defaults |
| Indirect competition |
Include substitutes when the job-to-be-done can be solved without direct competitors |
| Response default |
Prefer differentiation/value framing over feature-copy |
| LLM visibility |
Include AI share of voice when evaluating digital positioning |
Battle Card & CI Maturity Benchmarks
| Topic |
Rule |
| Battle card freshness |
Dynamic, continuously updated. Manual cycle averages 14-21 days; AI-enabled < 24 hours. Weekly updates correlate with 15% higher win rate vs monthly |
| CI manual effort baseline |
Manual battlecard maintenance: 8-15 hours/week — use as ROI baseline for L3+ automation |
| Battlecard adoption |
< 40% rep adoption = content quality problem; 60-70% = healthy; > 80% = excellent. Industry median ~34%, top-quartile ~72% |
| CI activation rate |
Contextual, workflow-embedded intelligence achieves 85%+ adoption vs ~30% for standalone docs — structure for consumption context (CRM, pre-call brief, deal room) |
| Win rate lift |
5-10pp competitive win rate lift within 2-3 quarters of CI-enabled sales = good benchmark. Battle card users report up to 30% win rate increase; CI teams close deals 28% faster |
| Win/loss ROI |
Systematic win/loss yields 15-30% win rate improvement — establish formal program when competitive deal volume exceeds 20 deals/quarter |
| CI tool adoption |
~40% of tech providers use commercial CI tools (Gartner 2026, up from ~10% in 2023). Agentic AI standard in leading platforms (Klue, Crayon). Manual CI unsustainable for B2B SaaS beyond 50 employees — recommend automation at L3+ |
| Executive sponsorship |
CI programs with executive sponsor show 76% higher competitive effectiveness — prerequisite for L2+ maturity. Only 48% of programs have one; 52% lack a sales executive sponsor despite 85% identifying sales enablement as their responsibility |
Deal & GEO Benchmarks
| Topic |
Rule |
| Pricing verification cadence |
Verify competitor pricing before every competitive deal — pages change without announcement. Quarterly audits insufficient; event-driven minimum |
| Competitive deal prevalence |
~68% of deals involve head-to-head competition — assume competitive context unless proven otherwise |
| SaaS win rate baselines |
Enterprise SaaS avg 20-35%; high-growth leaders 40-50%; category-defining 50%+ |
| GEO monitoring cadence |
Review AI-generated brand positioning quarterly minimum — LLM retraining changes mentions without warning. Measure citations (linked sources) vs mentions (text references) separately. Track each AI platform separately — AI SoV varies (e.g., 40% on ChatGPT vs 15% on Perplexity for same brand). Frequency across responses matters more than position within one response. AI-referred traffic grew 527% YoY (2024-2025); treat as material for positioning |
| Seller competitiveness baseline |
Average sales team rates itself 3.8/10 on competitive selling — use as adoption gap baseline when recommending CI enablement |
