anthropics--financial-services/investment-banking/skills/pitch-deck/reference/calculation-standards.md

Calculation Verification Reference

This file provides formulas and guidelines for verifying pre-calculated values in source data before populating templates. Source data should already contain calculated figures—use these formulas to verify accuracy.

Contents

• Key Verification Formulas
• Consensus Methodology
• Rounding Guidelines
• Verification Checklist
• Red Flags to Investigate

---

Key Verification Formulas

CAGR Projection

Formula:

Future Value = Present Value × (1 + CAGR)^n

Variables:

• Present Value: Current/base year market size
• CAGR: Compound Annual Growth Rate (as decimal, e.g., 16.4% = 0.164)
• n: Number of years between base and target year

Verification example:

Source claims: $22.1bn (2024) at 16.4% CAGR = $55.0bn (2030)

Verify: 22.1 × (1.164)^6 = 22.1 × 2.488 = 55.0 ✓

Calculating n (years): Count years between base and target year. Examples: 2024→2030 = 6 years, 2025→2030 = 5 years.

Valuation Multiples

EV/Revenue:

EV/Revenue Multiple = Enterprise Value ÷ Revenue
Implied EV = Revenue × Multiple

EV/EBITDA:

EV/EBITDA Multiple = Enterprise Value ÷ EBITDA
Implied EV = EBITDA × Multiple

Verification example:

Source claims: $436m deal at 9.7x revenue multiple on $45m revenue

Verify: 436 ÷ 45 = 9.69 ≈ 9.7x ✓

Market Share

Formula:

Market Share = (Segment Size ÷ Total Market Size) × 100

Verification example:

Source claims: Online segment ($18bn) is 28% of total market ($65bn)

Verify: 18 ÷ 65 = 0.277 = 27.7% ≈ 28% ✓

Growth Rate

Year-over-Year:

YoY Growth = (Current Year - Prior Year) ÷ Prior Year × 100

CAGR from endpoints:

CAGR = (End Value ÷ Start Value)^(1/n) - 1

---

Consensus Methodology

When source data contains multiple estimates, verify consensus calculations:

Size Consensus (Range)

Method: Full min-max range across all sources

Example:

Sources: $14.9bn, $18.3bn, $21.1bn, $21.2bn, $22.1bn
Consensus: $15-22bn (rounded to nearest $1bn)

CAGR Consensus (Central Cluster)

Method: Exclude outliers (highest and lowest), use central cluster range

Example:

Sources: 10.6%, 16.4%, 17.2%, 19.0%, 22.7%
Exclude outliers: 10.6% (low), 22.7% (high)
Central cluster: 16.4%, 17.2%, 19.0%
Consensus: 16-19% or 16-17% (conservative)

Projection Consensus

Method: Apply consensus CAGR to midpoint of size range

Example:

Size range: $15-22bn → Midpoint: $18.5bn
CAGR consensus: 16-17%
At 16%: 18.5 × (1.16)^6 = $45.1bn
At 17%: 18.5 × (1.17)^6 = $47.5bn
Consensus projection: $45-48bn

---

Rounding Guidelines

These are typical conventions — adjust based on the magnitude of values and template style:

Value Type	Typical Rounding	Example
Large market sizes ($10bn+)	Nearest $1bn	18.47 → $18bn
Smaller market sizes (<$10bn)	Nearest $0.5bn or $0.1bn	2.3 → $2.5bn
Size ranges	Match precision of sources	14.9-22.1 → $15-22bn
CAGR	Whole % or 0.5%	16.4% → 16% or 16.5%
Market share	Nearest 5% or match source	27.7% → 25% or 30%
Revenue ($m)	1 decimal	18.47 → $18.5m
Multiples	1 decimal	9.688 → 9.7x

Rounding principles:

• Rounding should not materially change the figure — for smaller values, use finer precision
• Consistency matters more than precision — use same rounding across similar figures
• When creating ranges, round down for low end, round up for high end
• For summary statistics (mean, median), match precision of input data

---

Verification Checklist

Before using any calculated value from source data:

Formula Verification

• Projection uses correct CAGR formula: PV × (1 + r)^n
• Multiples calculated as EV ÷ Metric (not reversed)
• Growth rates use correct base year in denominator
• Percentage shares sum to ~100% where applicable

Input Verification

• Base year figures match source documents
• CAGR/growth rates match stated source methodology
• Time periods (n) calculated correctly
• Currency and units consistent ($bn vs $m)

Output Verification

• Calculated result matches source's stated figure
• If mismatch, investigate methodology difference
• Rounding applied consistently
• Results are plausible (no order-of-magnitude errors)

Consensus Verification

• All sources included in range calculations
• Outlier exclusion methodology documented
• Midpoint calculations use correct averaging
• Range bounds represent actual min/max or documented subset

---

Red Flags to Investigate

Projection mismatches:

• Calculated projection differs from source by >5%
• Likely cause: Different base year, different CAGR, or rounding

Multiple mismatches:

• Calculated multiple differs from source
• Likely cause: Different metric definition (LTM vs. NTM, Revenue vs. Net Revenue)

Consensus mismatches:

• Your consensus differs from source's consensus
• Likely cause: Source excluded certain data points, different outlier treatment

When in doubt: Note the discrepancy in a footnote and show your calculation methodology.