Transparency
AI Disclaimer
How KCENAV uses AI — and where the line is between calculated scores and AI-generated insights.
Last updated: April 1, 2026
KCENAV's core value proposition is transparency: We clearly distinguish between outputs that are mathematically calculated (auditable, deterministic) and outputs that are AI-generated (generative, contextual). Both are valuable — but they're different, and you should know which is which.
Type 1
Calculated Outputs
Deterministic. Auditable. Same inputs always produce the same outputs.
- Composite scores (0–100)
- Letter grades (A through F)
- Pillar-level breakdowns
- Percentile rankings vs. peers
- Gap identification
Type 2
AI Insights
Generative. Contextual. Grounded in your scored data — but interpretive.
- Strategic recommendations
- Prioritized action roadmaps
- Narrative diagnostics
- Cross-tool synthesis
- Benchmark commentary
1. How Our Scoring Engines Work
Each of KCENAV's six diagnostic tools uses a weighted input → composite score architecture. Here's the basic model:
- Input collection: You answer 10–20 questions across multiple scoring dimensions (“pillars”)
- Pillar scoring: Each pillar aggregates your responses using predefined weights (e.g., revenue concentration may be weighted at 15% of your HALO score)
- Composite calculation: Pillar scores are aggregated with inter-pillar weights to produce a single 0–100 composite score
- Letter grade assignment: Composite scores map to letter grades based on fixed thresholds (e.g., 85–100 = A, 70–84 = B)
- Benchmarking: Your score is compared against anonymized, aggregated data from other companies on the platform
The scoring logic is deterministic: if you take the same assessment twice with identical answers, you will receive the identical score. There is no randomness in the score calculation layer.
2. Our Six Diagnostic Engines
| Tool |
What It Measures |
Score Type |
| HALO Score |
Overall company health across 5 pillars: market position, operations, leadership, financials, and strategic assets |
Calculated |
| LEAD Score |
Leadership depth and founder dependency risk across team structure, succession readiness, and operational independence |
Calculated |
| Growth Scaling |
Scaling readiness: revenue predictability, team capacity, systems infrastructure, and growth bottleneck identification |
Calculated |
| Valuation Optimizer |
EBITDA multiple drivers and exit value optimization potential across financial, operational, and strategic dimensions |
Calculated |
| M&A Readiness |
Acquisition preparedness for buy or sell side — due diligence readiness, deal structure compatibility, integration risk |
Calculated |
| Exit Readiness |
Exit readiness with EBITDA impact mapping, buyer attractiveness rating, and pre-exit improvement roadmap |
Calculated |
Where AI-generated recommendations appear (roadmaps, narrative summaries, strategic commentary), they are clearly labeled and are derived from — but distinct from — the underlying calculated scores.
3. AI-Generated Insights: What They Are and Aren't
AI insights are a starting point for strategic thinking — not a substitute for professional advisory, legal counsel, financial analysis, or due diligence.
When KCENAV generates written recommendations, action plans, or narrative commentary, these are produced by large language AI models (LLMs) working from your scored data. You should understand:
- Grounded in your scores: AI insights are generated with your specific scores, pillar breakdowns, and gap analysis as context — they're not generic
- Interpretive, not definitive: AI models reason probabilistically. Two companies with similar scores may receive different recommendations based on nuanced input differences
- Not licensed professional advice: AI-generated content does not constitute a financial opinion, legal opinion, audit, valuation, or fairness opinion
- Improving over time: As more companies complete assessments, our benchmark data improves — which improves both scoring calibration and the quality of insights
- Human oversight available: For significant decisions, we offer human-delivered advisory services where our team reviews your diagnostics and provides expert guidance
4. Benchmark Data and Improving Over Time
KCENAV's benchmarks improve as more companies complete assessments. Here's how:
- Your assessment responses contribute (anonymously and in aggregated form) to industry benchmarks by company size, sector, and growth stage
- Benchmark percentiles become more precise as the dataset grows
- Scoring weight calibration is reviewed periodically against real-world outcomes (e.g., actual M&A deal valuations, growth trajectories) where available
- We do not use your personal or company-identifiable data in public benchmarks — only anonymized, aggregated signals
Benchmark data is directional and comparative — not a guarantee of outcomes. A company scoring in the 90th percentile is outperforming peers on the measured dimensions, not guaranteed to achieve any specific business result.
5. What KCENAV Is Not
To be completely clear about scope:
- Not a licensed financial advisor: KCENAV is not registered as an investment advisor, broker-dealer, or financial planner
- Not a law firm: Nothing on this platform constitutes legal advice
- Not an accounting firm: Scores are not audited financial statements
- Not a valuation firm: Valuation Optimizer scores are directional indicators, not formal business valuations for M&A or financing purposes
- Not a substitute for due diligence: M&A Readiness and Exit Readiness scores inform preparation — formal due diligence requires professional advisors
Our advisory services team are experienced operators and strategists — not licensed attorneys, CPAs, or registered investment advisors. Advisory engagements are strategic guidance, not regulated professional services.
6. Accuracy and Limitations
Our scores are only as good as the data you provide. Accuracy limitations to be aware of:
- Self-reported data is subject to individual interpretation — two people at the same company might answer the same question differently
- Benchmarks represent the KCENAV user population, which may differ from your specific industry's broader distribution
- Scoring models are updated periodically — historical scores may not be directly comparable across major model revisions
- AI-generated insights may occasionally misinterpret edge cases or unusual business configurations