Volume Estimation Methodologies

Bottom-up estimation aggregates partner-level estimates:

BOTTOM-UP METHODOLOGY

Formula:
Total ODL Volume = Σ (Partner Volume Estimates)

1. List all known ODL partners
2. Classify by tier and stage
3. Estimate volume for each partner
4. Sum estimates
5. Apply uncertainty range

Advantages:
├── Based on observable partnership data
├── Can update as new info emerges
├── Identifies concentration (which partners drive volume)
└── Forces engagement with specifics

Disadvantages:
├── Missing partner data (undisclosed partners)
├── Estimation error compounds
├── Requires many assumptions
└── Labor-intensive

For each partner, estimate volume using:

PARTNER VOLUME ESTIMATION FRAMEWORK

Known Data:
├── Partner identity and business
├── Corridors operated
├── Stage (testing/pilot/production/scale)
├── Duration of ODL usage
└── Any disclosed metrics

Estimation Inputs:
├── Corridor market size (public data)
├── Partner market share in corridor (estimate)
├── ODL percentage of partner's volume (estimate)
├── Growth rate (from stage progression)
└── Seasonality factors

Example: SBI Remit
├── Japan→Philippines corridor: ~$2-3B annually
├── SBI Remit market share: 10-20% (estimate)
├── SBI Remit total: $200-600M
├── ODL percentage: 70-90% (estimate)
├── SBI Remit ODL: $150-500M annually
└── Range reflects genuine uncertainty

Build a structured model:

BOTTOM-UP MODEL TEMPLATE

TIER 1 PARTNERS
┌─────────────────┬───────────┬──────────────┬──────────────┐
│ Partner         │ Stage     │ Low Est      │ High Est     │
├─────────────────┼───────────┼──────────────┼──────────────┤
│ SBI Holdings    │ 5         │ $300M        │ $600M        │
│ Tranglo         │ 5         │ $200M        │ $600M        │
├─────────────────┼───────────┼──────────────┼──────────────┤
│ Tier 1 Total    │           │ $500M        │ $1,200M      │
└─────────────────┴───────────┴──────────────┴──────────────┘

TIER 2 PARTNERS
┌─────────────────┬───────────┬──────────────┬──────────────┐
│ Partner         │ Stage     │ Low Est      │ High Est     │
├─────────────────┼───────────┼──────────────┼──────────────┤
│ Pyypl           │ 4-5       │ $50M         │ $150M        │
│ Travelex Brazil │ 4-5       │ $50M         │ $150M        │
│ Novatti         │ 4-5       │ $30M         │ $80M         │
│ Other Tier 2    │ Various   │ $100M        │ $300M        │
├─────────────────┼───────────┼──────────────┼──────────────┤
│ Tier 2 Total    │           │ $230M        │ $680M        │
└─────────────────┴───────────┴──────────────┴──────────────┘

TIER 3 (Early Stage)
├── Various testing/pilot │ $20M │ $100M │
└─────────────────────────┴──────┴───────┘

TOTAL ODL VOLUME
├── Low Estimate: $750M annually
├── High Estimate: $2.0B annually
├── Mid-point: ~$1.4B annually
└── Confidence: Medium (wide range reflects uncertainty)

Apply adjustments for known biases:

ADJUSTMENT FACTORS

Undisclosed Partners Adjustment:
├── Some ODL partners may not be publicly known
├── Add 10-20% to account for undisclosed volume
├── Basis: Ripple mentions undisclosed customers
└── Adjustment: +10-20%

Double-Counting Prevention:
├── Some volume may flow through Tranglo AND partners
├── Tranglo is infrastructure; partners are end users
├── Risk of adding both
└── Adjustment: Review for overlap, typically -5-10%

Stage Uncertainty Adjustment:
├── Announced stage may not reflect current reality
├── Partners may have scaled up or down
├── Last update may be stale
└── Adjustment: Widen confidence interval

Final Adjusted Range:
├── Raw total: $750M - $2.0B
├── Undisclosed adjustment: +$75M - $400M
├── Double-counting adjustment: -$40M - $150M
├── Final range: $800M - $2.3B
└── Round to: $1-2.5B (or $1-3B for simplicity)

Top-down starts with market size and estimates ODL penetration:

TOP-DOWN METHODOLOGY

Formula:
ODL Volume = Market Size × ODL-Viable Share × Penetration Rate

1. Define relevant market (cross-border payments)
2. Identify ODL-viable segment
3. Estimate ODL penetration rate
4. Calculate implied volume
5. Compare against bottom-up for triangulation

Advantages:
├── Grounded in market fundamentals
├── Forces realistic penetration thinking
├── Useful for future projections
└── Less dependent on partner-specific data

Disadvantages:
├── Multiple assumption layers
├── "Market size" definitions vary
├── Penetration rates highly uncertain
└── May miss structural barriers

Define the addressable market carefully:

MARKET SIZE HIERARCHY

Total Cross-Border Payments: ~$150-180T annually
├── Includes B2B, B2C, interbank, trade finance
├── Most is between major currency pairs
├── Much is intracompany or related party
└── NOT the relevant number

Corporate Cross-Border: ~$30-40T annually
├── Still mostly major currencies
├── Existing infrastructure efficient
├── ODL benefit limited for most
└── Partially relevant

Consumer/SMB Remittances: ~$1-1.5T annually
├── Often high-cost corridors
├── ODL-relevant segment larger
├── But infrastructure varies
└── More relevant

ODL-Viable Corridors: ~$5-8T annually
├── High-cost traditional corridors
├── Corridors where ODL offers savings
├── Excludes efficiently served routes
├── Regulatory access for ODL
└── THIS is the relevant market

Key Insight:
├── "ODL could capture X% of $150T" is misleading
├── "$150T → 0.01% = $15B" sounds reasonable but isn't
├── Start from ODL-viable market (~$5-8T)
└── Penetration of realistic base

Estimate ODL's share of the addressable market:

PENETRATION RATE ANALYSIS

Current Penetration:
├── ODL-viable market: ~$5-8T
├── Current ODL volume: ~$1-3B (our estimate)
├── Current penetration: 0.01-0.05%
└── Extremely early stage

Penetration Comparisons:
├── SWIFT gpi (tracking): ~85% of SWIFT volume
├── Wise (digital remittance): ~5% of relevant corridors
├── Crypto payments overall: <0.1% of global payments
└── ODL at 0.01-0.05% is consistent with early stage

Penetration Projection Factors:
├── Partner growth rate
├── Corridor expansion
├── Competitive dynamics
├── Regulatory developments
└── Technology adoption curves

5-Year Penetration Scenarios:
├── Bear case: 0.1% ($5-8B)
├── Base case: 0.3% ($15-25B)
├── Bull case: 1.0% ($50-80B)
└── Extreme bull: 3.0% ($150-250B)

Construct the top-down model:

TOP-DOWN MODEL

Market Foundation:
├── Global cross-border: $150T
├── Less: Major currency/efficient corridors: -$120T
├── Less: Regulated-out markets: -$20T
├── ODL-viable market: ~$5-8T
└── Use: $6T as mid-point

Current State:
├── ODL-viable market: $6T
├── Current penetration: 0.02-0.04%
├── Implied current volume: $1.2-2.4B
└── Consistent with bottom-up estimate

Projection (5-year):
├── ODL-viable market growth: 5% annually → $7.5T
├── Penetration growth: 0.02% → 0.3% (15× growth)
├── Implied volume: $20-25B
└── Requires ~50% annual volume growth (aggressive)

XRP Ledger data provides observable information:

ON-CHAIN DATA SOURCES

What's Observable:
├── Payment transaction volume on XRPL
├── XRP movement between addresses
├── DEX trading activity
├── Large transaction patterns
└── Exchange inflows/outflows

What's NOT Observable:
├── Which transactions are ODL vs speculation
├── Institution attribution (addresses anonymous)
├── Exact partner volumes
├── Intent behind transactions
└── Off-ledger netting

Useful Metrics:
├── Payment-type transactions (vs offers/other)
├── Large transaction volume
├── Known exchange address activity
├── Velocity patterns
└── Unusual activity correlation with announcements

Identifying likely ODL transactions:

ODL TRANSACTION CHARACTERISTICS

Typical ODL Pattern:
├── XRP purchased on one exchange
├── Transferred via XRPL (3-5 seconds)
├── Sold on destination exchange
├── Round trip: Under 1 minute
└── Amounts: Consistent with remittance sizes

On-Chain Signatures:
├── Source: Known exchange address
├── Destination: Known exchange address (different)
├── Timing: Rapid (under 1 minute total)
├── Size: $1K-$50K typical (remittance-sized)
├── Pattern: Repeated, consistent timing
└── These suggest ODL activity

Attribution Challenges:
├── Exchange addresses handle many users
├── Similar patterns for arbitrage trading
├── Individual transactions can't be definitively attributed
├── Aggregate patterns more reliable than individual
└── Best used as directional indicator

Build estimates from on-chain data:

ON-CHAIN ESTIMATION APPROACH

Step 1: Identify Payment Transactions
├── Filter XRPL data for Payment-type
├── Exclude offers, escrow, etc.
├── Focus on cross-exchange patterns
└── Result: Candidate ODL transactions

Step 2: Apply Filters
├── Size filter: $1K-$100K (typical ODL range)
├── Timing filter: Rapid settlement pattern
├── Address filter: Exchange-to-exchange
└── Result: High-probability ODL candidates

Step 3: Aggregate and Annualize
├── Sum filtered transaction volumes
├── Adjust for sampling period
├── Annualize with seasonality consideration
└── Result: On-chain volume estimate

Step 4: Confidence Assessment
├── False positives: Arbitrage, trading counted as ODL
├── False negatives: ODL via different patterns missed
├── Net bias: Unknown, possibly upward
└── Treat as directional, not precise

Example Output:
├── On-chain analysis suggests: $1-4B annually
├── Wide range reflects attribution uncertainty
├── Use for triangulation, not primary estimate
└── Cross-check against bottom-up/top-down

Be honest about what on-chain analysis can't do:

ON-CHAIN LIMITATIONS

Cannot Determine:
├── Which institution made a transaction
├── Whether transaction is ODL vs trading
├── Geographic origin/destination
├── Whether fiat conversion actually occurred
└── The "why" behind any transaction

Can Mislead:
├── Arbitrage trading looks like ODL
├── Speculation creates volume not utility
├── Exchange consolidations create false signals
├── Market making activity included
└── Over-attribution to ODL possible

Appropriate Use:
├── Directional indicator (growing/shrinking)
├── Order of magnitude check
├── Correlation with partnership announcements
├── Triangulation against other methods
└── NOT primary estimation method

Combine multiple methods for robust estimates:

TRIANGULATION METHODOLOGY

Principle:
├── No single method is reliable alone
├── Multiple methods converge → higher confidence
├── Methods disagree → investigate discrepancy
└── Final estimate weights all methods

Method Weighting:
├── Bottom-up: 50% weight (most granular)
├── Top-down: 30% weight (market-grounded)
├── On-chain: 20% weight (observable but noisy)
└── Weights reflect relative reliability

1. Generate estimate from each method
2. Compare ranges for overlap
3. Investigate discrepancies
4. Weight-average for central estimate
5. Use widest range for confidence interval

Work through a complete triangulation:

TRIANGULATION EXAMPLE

Bottom-Up Result:
├── Low: $800M
├── High: $2.3B
├── Mid: $1.5B
└── Weight: 50%

Top-Down Result:
├── Low: $1.2B
├── High: $2.4B
├── Mid: $1.8B
└── Weight: 30%

On-Chain Result:
├── Low: $1.0B
├── High: $4.0B
├── Mid: $2.0B
└── Weight: 20%

Weighted Synthesis:
├── Mid-point: (1.5×0.5) + (1.8×0.3) + (2.0×0.2) = $1.7B
├── Low: Lowest of lows = $0.8B
├── High: Highest of highs = $4.0B
├── But: On-chain high may be inflated (arbitrage)
└── Adjusted high: $2.5B (excluding noisy on-chain tail)

Final Estimate:
├── Central estimate: ~$1.5-2.0B annually
├── Confidence range: $1-3B annually
├── Communicate as range, not point
└── Note: "High uncertainty; methods agree on order of magnitude"

When methods disagree significantly:

DISCREPANCY ANALYSIS

If Bottom-Up >> Top-Down:
├── Check: Partner estimates too high?
├── Check: Market definition too narrow?
├── Check: Double-counting in bottom-up?
└── Resolution: Revise assumptions until explained

If Top-Down >> Bottom-Up:
├── Check: Missing partners in bottom-up?
├── Check: Market penetration overestimated?
├── Check: Market size inflated?
└── Resolution: Identify missing volume or revise market

If On-Chain >> Both:
├── Check: Non-ODL volume included?
├── Check: Arbitrage/trading misattributed?
├── Check: On-chain filters too loose?
└── Resolution: Tighten on-chain methodology

Key Principle:
├── Discrepancies are information
├── Force investigation of assumptions
├── Resolved discrepancies increase confidence
└── Unresolved discrepancies widen range

How to communicate estimates appropriately:

UNCERTAINTY COMMUNICATION

Bad Practice:
├── "ODL volume is $1.7 billion"
├── False precision suggests certainty
├── Reader may not understand uncertainty
└── Decision-making on false foundation

Good Practice:
├── "ODL volume is estimated at $1-3B annually"
├── "Central estimate ~$1.5-2B with wide uncertainty"
├── "Bottom-up suggests $1-2B; top-down $1-2.5B"
└── Range communicates genuine uncertainty

Best Practice:
├── State range prominently
├── Explain methodology briefly
├── Note key assumptions
├── Identify what would change estimate
└── Update as new information emerges

Example Statement:
"Based on bottom-up partner analysis and top-down market 
penetration, we estimate current ODL volume at $1-3B annually,
with central estimate around $1.5-2B. Key uncertainties include
partner-level volumes (not disclosed) and attribution of
on-chain activity. This estimate would increase if SBI or
Tranglo publicly disclose higher volumes, or decrease if
major partners reduce activity."

Map estimates to confidence categories:

CONFIDENCE LEVEL FRAMEWORK

High Confidence (±20%):
├── Multiple confirming data sources
├── Official disclosures available
├── Methods closely agree
├── Limited assumption dependency
└── Example: SWIFT transaction volume

Medium Confidence (±50%):
├── Some confirming data
├── Partial disclosures
├── Methods somewhat agree
├── Moderate assumption dependency
└── Example: Our ODL estimates ($1-3B range)

Low Confidence (±100% or more):
├── Limited/no confirming data
├── No official disclosures
├── Methods disagree significantly
├── High assumption dependency
└── Example: ODL volume by specific corridor

Very Low Confidence:
├── Essentially speculation
├── Any number is plausible
├── Don't estimate; state "unknown"
└── Example: Future ODL volume in 10 years

How to revise estimates as information changes:

ESTIMATE UPDATE PROTOCOL

Update Triggers:
├── New partner disclosure (add to bottom-up)
├── Market data update (revise top-down)
├── On-chain methodology improvement
├── Partner status change (stage progression/regression)
├── Ripple official statement
└── Material new information

Update Process:
├── Document what changed
├── Revise affected method(s)
├── Re-triangulate
├── Update confidence level
├── Communicate change and reason
└── Archive previous estimate

Stability Principle:
├── Don't update for noise
├── Update for material new information
├── Small changes: Annual update sufficient
├── Large changes: Immediate update
└── Document decision either way

Build a working model:

VOLUME MODEL STRUCTURE

TAB 1: Partner Estimates (Bottom-Up)
├── Partner name
├── Tier classification
├── Stage
├── Corridors
├── Low volume estimate
├── High volume estimate
├── Methodology notes
├── Last updated
└── Sum formulas for totals

TAB 2: Market Analysis (Top-Down)
├── Market segment definitions
├── Size estimates with sources
├── ODL-viable market calculation
├── Penetration rate assumptions
├── Implied volume calculation
└── Sensitivity tables

TAB 3: On-Chain Analysis
├── Data sources
├── Filter criteria
├── Aggregated volumes
├── Attribution methodology
├── Confidence assessment
└── Comparison to other methods

TAB 4: Triangulation
├── Bottom-up summary
├── Top-down summary
├── On-chain summary
├── Weighted average calculation
├── Final range determination
└── Confidence level assessment

TAB 5: Documentation
├── Methodology description
├── Key assumptions listed
├── Update log
├── Source links
└── Revision history

Keep your model current:

MODEL MAINTENANCE SCHEDULE

Weekly:
├── Scan for material partner news
├── Note any updates needed
├── Don't change estimates for minor news
└── Log significant developments

Monthly:
├── Update any partner stage changes
├── Incorporate new partner disclosures
├── Refresh on-chain analysis
├── Re-triangulate if material changes
└── Update confidence levels

Quarterly:
├── Full bottom-up review
├── Market size data update
├── Complete re-triangulation
├── Documentation review
├── Communicate estimate update (if material)
└── Archive previous version

Annually:
├── Complete methodology review
├── Assumption validation
├── Comparison to any disclosed actuals
├── Framework improvement
└── Full documentation update

✅ Multiple estimation methods converge on $1-3B annually — Bottom-up partner analysis, top-down market penetration, and on-chain data all suggest ODL volume in this range for 2024-2025

✅ Triangulation increases estimate reliability — When methods agree on order of magnitude despite different approaches, confidence increases

✅ ODL penetration is ~0.01-0.05% of viable market — Current volume represents early-stage adoption regardless of estimation method

⚠️ Partner-level volumes are not disclosed — All partner volume estimates involve significant assumptions; actual volumes could differ materially

⚠️ On-chain attribution is imprecise — Distinguishing ODL from trading/arbitrage is difficult; on-chain estimates have wide error bars

⚠️ Market size definitions affect conclusions — Whether the relevant market is $5T or $10T changes penetration rate interpretations

🔴 False precision in estimates — Stating "$1.7B" rather than "$1-3B" creates false confidence in uncertain estimates

🔴 Single-method reliance — Any one method has significant weaknesses; triangulation is essential

🔴 Static estimates — Volume changes over time; estimates require regular updating as new information emerges

Rigorous volume estimation produces ranges, not points. Current ODL volume is approximately $1-3B annually based on triangulated analysis—a wide range reflecting genuine uncertainty. This represents 0.01-0.05% penetration of the addressable market, consistent with early-stage adoption. The methodology matters more than the specific number; as new information emerges, estimates should be updated while maintaining appropriate uncertainty acknowledgment.

Assignment: Build your own ODL volume estimation model using the methodologies from this lesson.

Requirements:

Part 1: Bottom-Up Model (35%)

• List all known ODL partners (Tier 1-3)
• Document stage and corridors for each
• Estimate low/mid/high volume for each partner
• Show calculation methodology for key partners
• Sum to aggregate bottom-up estimate
• Apply adjustment factors (undisclosed, double-counting)

Part 2: Top-Down Model (25%)

• Define market size hierarchy (total → viable)
• Document sources for market size data
• Estimate current penetration rate
• Calculate implied current volume
• Create sensitivity table (penetration vs market size)

Part 3: On-Chain Analysis (20%)

• Document data sources used
• Explain attribution methodology
• Produce on-chain volume estimate
• Assess confidence level
• Note limitations

Part 4: Triangulation and Synthesis (20%)

• Present each method's estimate
• Apply weighting
• Investigate any discrepancies
• Produce final estimate with range
• Write uncertainty communication statement

Grading Criteria:

Time investment: 6-8 hours
Value: This model becomes your volume estimation foundation

Market Data Sources:

• McKinsey Global Payments Report
• World Bank Remittance Data
• BIS Cross-Border Payment Statistics
• Industry research on payment corridors

On-Chain Analysis:

• XRPL Mainnet explorers
• ODL tracking tools (community-developed)
• Academic research on blockchain analytics

For Next Lesson:

Lesson 14 covers Growth Trajectory Analysis—projecting how ODL volume might grow over time and what scenarios are realistic versus wishful thinking.

End of Lesson 13

Total words: ~6,100
Estimated completion time: 60 minutes reading + 6-8 hours for deliverable