Moat Evaluation — How jini Measures Competitive Durability

Two companies can earn the same profit. One will keep earning it for years. The other will watch competitors copy it and crush the margin. The difference is the moat — and it's the difference between a bargain and a trap. jini's AI rates moat strength and feeds it into both valuation and verdict.

What Is a Moat?

A moat is Warren Buffett's metaphor for competitive durability — the structural advantages that keep competitors from stealing market share.

Examples:

• Brand: Coca-Cola vs. a generic cola. Same molecule, but the name commands a price premium
• Switching cost: Moving your company's accounting to new software means migrating 10 years of data, retraining staff, and risking downtime — so you don't
• Network effect: A social network is more valuable the more users it has → competitors have a harder time bootstrapping a rival
• Cost advantage: Scale, proprietary supply chains, or technology mastery lets you undercut competitors
• Intangible assets: Patents, licenses, or exclusive know-how competitors can't easily replicate

Why Moat Matters for Valuation

Profit without moat is temporary:

Scenario A: Company with strong moat
  Annual profit: $100M (expected to grow to $120M in 5 years)
  → Fair P/E: 10x (safe to assume durable earnings)

Scenario B: Company in competitive industry
  Annual profit: $100M (but competitors undercut you; profit drops to $60M in 5 years)
  → Fair P/E: 6x (earnings are fragile)

Same profit, half the valuation multiple. This is why "cheap for a reason" is such a common investor trap.

How jini Evaluates Moat

① The Score (0–100)

jini's AI analyzes financial metrics, industry structure, and competitive position to assign a moat score:

• 80–100: Very strong moat (Amazon, Microsoft, Coca-Cola tier)
• 60–79: Solid moat
• 40–59: Weak moat
• 0–39: Negligible moat (intensely competitive industry)

The score always carries an AI_ESTIMATE label so you know it's an AI judgment, not raw data.

② Drivers (Reasoning)

A score without reasoning is useless. The AI also lists the moat's pillars:

Company: Microsoft
Moat Score: 85
Drivers:
  1. Network effect — Windows/Office ecosystem lock-in
  2. Switching cost — Enterprise deployment and integration costs are enormous
  3. Cost advantage — Cloud infrastructure scale and technology leadership
  4. Brand — Enterprise trust and security reputation

You see which moat sources the AI identified, so you can validate or contest the judgment.

③ Feeds Into Quality Score

jini's Quality Score (one of 8 company dimensions) incorporates moat assessment:

• Ideal: You provide moat input directly (USER_INPUT)
• Reality: Most companies lack direct moat data → AI fills a 15% slot with AI_ESTIMATE
• Result: Quality Score might be 70, of which 12 points come from estimated moat

The system transparently shows whether each component is USER_INPUT or AI_ESTIMATE.

How Moat Affects Your Verdict

Direct Path: Moat → Quality → Downgrade

Low moat weakens Quality Score. Quality <45 triggers automatic verdict downgrade:

Case: Undervalued (+40% upside) but weak moat (moat score 20)
  → Quality Score drops to 42
  → Quality gate triggered (min 45)
  → Verdict downgraded to "Fairly Valued"

Indirect Path: Moat → Multiple Adjustment → Fair Value

In the peer valuation method, strong moat earns a multiple premium:

Your company's moat > peer median moat
  → Apply +5% to +8% multiple uplift
  → Fair value rises
  → Upside shrinks

Weak moat gets a discount. Same earnings, different multiple, different valuation.

USER_INPUT Overrides AI

Your judgment beats the algorithm. If you input a moat score, it replaces the AI estimate:

Default:
  Moat Score: 45 (AI_ESTIMATE)

You enter:
  Moat Score: 80 (USER_INPUT) — "This company has a strong brand"
  
Result:
  Moat Score: 80 (USER_INPUT)
  AI estimate is removed

This is intentional design. You own the final call.

Moat × Company Type Interaction

Moat strength transforms the investment profile of each company type:

Type	Weak Moat	Strong Moat
A (Fear)	High risk of permanent share loss after recovery	Fear reverses, then earnings rebound with pricing power intact
B (Cyclical)	Recovery uncertain; competitors may steal share	Recovery likely to restore profitability
C (Growth)	New growth is fragile; competitors can leapfrog	New growth sticks; durable margin expansion
D (Trap)	Worst case: profits falling in competitive market → Avoid	Less risky: baseline earnings hold via moat → Maybe recoverable
E (Cyclical Peak)	After recovery, market share at risk	After recovery, returns to historic profitability levels

Example:

• Type A (market panic) + strong moat: When sentiment reverses, the moat protects earnings recovery. High upside
• Type A + weak moat: Even after panic lifts, competitors steal market share. Limited recovery

When Moat Is "Insufficient Data"

Sometimes moat can't be assessed:

• Industry is too new (e.g., AI chip startups with unclear competitive landscape)
• Company is too small for meaningful peer comparison
• Business model is too complex to categorize

When this happens, the screen shows "Moat: Insufficient Data" and no AI_ESTIMATE is generated. Your options:

1. Research and input your own moat score (USER_INPUT)
2. Treat moat as unknown and demand higher safety margin (more conservative verdict)

Final Principle: Moat Is Earnings Durability

Moat strength = How long can this company defend its earnings power against competition?

• Strong moat: Earnings are durable → Higher P/E justified → Undervalued verdict is sound
• Weak moat: Earnings are temporary → Lower P/E required → Low price is still risky
• Unknown moat: Uncertainty → Demand wider margin of safety

jini evaluates moat because valuation must account for competitive reality. The same stock price of $10 can be a bargain if the moat is fortress-like, or a trap if moat is crumbling.

Trust but Verify

If jini's AI moat assessment feels wrong to you, override it. You know the business; the AI sees patterns in data. When your knowledge and the algorithm disagree, your judgment is the tiebreaker.

---

You now understand the full valuation chain: fair value methods feed into verdict logic, shaped by company type and moat durability. Use these together to move from data to judgment.