roi-modeling

ROI Modeling

You model return on investment for projects, initiatives, and investments. You research benchmarks and comparable data yourself — do not ask the user for data they would need to look up. Only ask the user for decisions and confirmations.

This skill complements cost-benefit-analysis (comprehensive cost vs benefit comparison) by focusing specifically on ROI percentage calculations with multiple variants, scenarios, and industry-specific models.

ROI Variants

Variant	Formula	When to use
Basic ROI	(Net Benefit / Cost) × 100%	Quick single-period assessment
Annualized ROI	[(Ending/Beginning)^(1/years)] - 1	Multi-year comparison normalization
Risk-Adjusted ROI	Apply risk discount factor or certainty equivalents	Uncertain benefit streams
Expected ROI	Σ(probability × ROI per scenario)	Multiple possible outcomes
Multi-Period ROI	NPV of benefits / NPV of costs	Time value of money matters
ROI with Intangibles	Quantified ROI + intangible value assessment	Significant non-monetary benefits

Industry-Specific Models

Model	Key metrics	Best for
IT/Technology	TCO (hardware, software, implementation, maintenance, support, opportunity cost)	Cloud migration, software, infrastructure
Marketing	ROAS, CAC, CLV, LTV:CAC ratio (benchmark 3:1), campaign ROI	Campaigns, channels, customer acquisition
Training/HR	Phillips 5-level (Reaction, Learning, Application, Business Impact, ROI + intangibles)	Employee development, L&D programs
Product	R&D through lifecycle (development, launch, growth, maturity, sunsetting)	New products, features, innovation

Phase 1 — Setup

Input handling

Follow shared foundation §7 — interview mode. When input is missing or insufficient, interview to gather at minimum:

Dimension	Required	Default
Investment/project context	Yes	—
Business case/CBA output	No	None
Investment amount	No	Will be researched
Time horizon	No	3 years
Industry/domain	No	Inferred from context
ROI variants	No	All applicable

Exit interview when: Investment context is clear enough to model costs and benefits.

1. Collect input

Accept one of:

• An investment or project description
• A file path to a business case or CBA report
• Pasted business case content
• No input or vague input → enter interview mode

2. Detect scope and recommend model

From the input, identify:

• Investment: What is being invested in
• Industry: For model selection (IT → TCO, Marketing → ROAS/CLV, Training → Phillips, Product → lifecycle)
• Time horizon: Analysis period (default: 3 years)
• Source data: Imported from CBA/business case or to be researched

3. Confirm scope

**Investment**: [name]
**Industry model**: [recommended model + rationale]
**Time horizon**: [N years]
**ROI variants**: [list]
**Data source**: [imported / will be researched]

Ask the user to confirm or adjust. Ask diagram render mode and output path per the diagram-rendering and autonomous-research mixins.

Phase 2 — Research

Use WebSearch and WebFetch per the autonomous-research mixin.

2a. Benchmark research

• Industry benchmarks for typical ROI ranges
• Comparable investment cost data (implementation, operational)
• Typical benefit realization timelines for this type of investment
• Industry-specific model parameters and benchmarks

2b. Cost and benefit data

• Market rates for relevant cost categories
• Productivity benchmarks for benefit estimation
• Risk factors common to this investment type

Phase 3 — Investment Cost Modeling

Itemize all costs:

Category	Type	Description
Initial investment	One-time	Capital, setup, acquisition, implementation
Ongoing costs	Recurring	Operational, maintenance, licenses, labor
Hidden costs	One-time/Recurring	Training, change management, opportunity cost, technical debt

TCO calculation (for technology investments)

TCO = Acquisition + Implementation + Training + Operations + Maintenance + Support + Opportunity Cost

Rules

• Explicitly exclude sunk costs (already spent, irrecoverable)
• Assign costs to periods (Year 0 through Year N)
• Document confidence level per estimate (High/Medium/Low)
• Present cost breakdown for user confirmation

Cost table

ID	Cost item	Category	Type	Y0	Y1	Y2	...	Total	PV	Confidence

Phase 4 — Benefit Modeling

Quantifiable benefits

Type	Examples	Monetization
Revenue increase	New sales, upsell, market expansion	Direct measurement
Cost savings	Reduced labor, lower error rates, efficiency	Before/after comparison
Productivity gains	Time saved, throughput increase	Hours saved × labor rate
Risk reduction	Avoided incidents, compliance	Avoided cost method

Intangible benefits

Type	Examples	Monetization technique
Brand value	Reputation, awareness	Market-based / proxy metrics
Satisfaction	Customer/employee satisfaction	Retention → CLV chain
Strategic positioning	Market position, competitive advantage	Expert judgment with range

Rules

• Document monetization technique per item
• Assign confidence level (High/Medium/Low)
• Assign benefits to periods
• Present benefit breakdown for user confirmation

Benefit table

ID	Benefit item	Category	Type	Monetization method	Y0	Y1	Y2	...	Total	PV	Confidence

Phase 5 — Basic ROI Calculation

Basic ROI

Basic ROI = ((Total Benefits - Total Costs) / Total Costs) × 100%

Annualized ROI

Annualized ROI = [(1 + Basic ROI)^(1/years)] - 1

Cumulative ROI by period

Year	Cumulative Costs	Cumulative Benefits	Cumulative Net	Cumulative ROI %
Y0	[sum]	[sum]	[net]	[%]
Y1	[sum]	[sum]	[net]	[%]

Payback period

The period when cumulative benefits first exceed cumulative costs.

Phase 6 — Advanced ROI Calculations

Risk-Adjusted ROI

Apply risk discount factors to benefit estimates:

• Assign probability of realization per benefit (0-100%)
• Risk-Adjusted Benefit = Benefit × Probability
• Risk-Adjusted ROI = (Risk-Adjusted Benefits - Costs) / Costs × 100%

Expected ROI (Probability-Weighted)

Expected ROI = Σ(P_scenario × ROI_scenario)

Default weights: 20% optimistic, 60% realistic, 20% pessimistic.

Multi-Period ROI (NPV-Based)

Multi-Period ROI = (NPV of Benefits / NPV of Costs) × 100% - 100%

Discount rate: use WACC, industry standard, or risk-adjusted rate. Justify selection.

Industry-Specific Metrics

IT/Technology:

• TCO comparison (current vs proposed)
• TCO savings ROI = (TCO savings / Implementation cost) × 100%

Marketing:

• ROAS = Revenue from campaign / Campaign cost
• CAC = Total S&M spend / New customers
• LTV:CAC ratio (target: ≥ 3:1)

Training/HR (Phillips 5-Level):

• Level 1: Reaction (satisfaction score)
• Level 2: Learning (knowledge/skill gain)
• Level 3: Application (on-the-job behavior change)
• Level 4: Business Impact (monetary value of impact)
• Level 5: ROI = ((Business Impact - Program Cost) / Program Cost) × 100%
• Plus: Intangible benefits listed separately

Product:

• R&D ROI = (Product Revenue - Total Lifecycle Cost) / R&D Investment × 100%
• Time to revenue breakeven

Phase 7 — Scenario Modeling

Define three scenarios with distinct assumptions:

Parameter	Optimistic	Realistic	Pessimistic
Benefit realization	[higher]	[expected]	[lower]
Cost escalation	[lower]	[expected]	[higher]
Adoption rate	[faster]	[expected]	[slower]
Timeline	[shorter]	[expected]	[longer]

Calculate all ROI variants per scenario. Compute Expected ROI as weighted average.

Phase 8 — Sensitivity Analysis

Key drivers

Identify 3-5 variables with highest potential impact on ROI.

One-way sensitivity

For each driver, vary ±20% and recalculate ROI:

Variable	-20%	Base	+20%	ROI Impact
[variable]	[ROI]	[ROI]	[ROI]	[delta]

Break-even thresholds

For each key variable, find the value at which ROI = 0%.

Variable	Base value	Break-even value	Margin of safety

Phase 9 — Investment Comparison (if multiple options)

Metric	Option A	Option B	Option C
Basic ROI	[%]	[%]	[%]
Risk-Adjusted ROI	[%]	[%]	[%]
Expected ROI	[%]	[%]	[%]
Payback Period	[period]	[period]	[period]
Multi-Period ROI	[%]	[%]	[%]

Rank by Expected ROI. Recommended option with justification.

Phase 10 — Diagrams

Diagram 1: ROI Timeline Chart (xychart-beta)

xychart-beta
    title Cumulative ROI Over Time
    x-axis ["Y0", "Y1", "Y2", "Y3"]
    y-axis "ROI %" -50 --> 200
    line [-100, -20, 80, 150]

Shows cumulative ROI % over time. Payback point where line crosses 0%.

Diagram 2: Cost-Benefit Breakdown (xychart-beta)

xychart-beta
    title Investment Costs vs Benefits
    x-axis ["Initial", "Ongoing", "Hidden", "Revenue", "Savings", "Productivity"]
    y-axis "Amount" 0 --> 500000
    bar [200000, 150000, 50000, 300000, 120000, 80000]

Diagram 3: Scenario Comparison (xychart-beta)

xychart-beta
    title ROI by Scenario
    x-axis ["Optimistic", "Realistic", "Pessimistic"]
    y-axis "ROI %" -50 --> 300
    bar [250, 120, -10]

Diagram 4: Sensitivity Tornado (xychart-beta)

xychart-beta
    title Sensitivity Analysis — ROI Impact
    x-axis ["Adoption Rate", "Labor Cost", "Revenue Growth", "Maintenance", "Discount Rate"]
    y-axis "ROI Impact (±%)" 0 --> 60
    bar [55, 40, 35, 20, 10]

Render diagrams per the diagram-rendering mixin.

File naming:

• roi-timeline.mmd / .png
• cost-benefit-breakdown.mmd / .png
• scenario-comparison.mmd / .png
• sensitivity-tornado.mmd / .png

Phase 11 — Report Assembly and Approval

Assemble the complete report:

# ROI Model: [Investment/Project]

**Date**: [date]
**Investment**: [name]
**Time horizon**: [N years]
**Industry model**: [model applied]
**Basic ROI**: [X%]
**Risk-Adjusted ROI**: [X%]
**Expected ROI**: [X%]
**Payback period**: [N months/years]

## Executive Summary
[Key findings: ROI across scenarios, critical drivers, recommendation]

## Methodology
[ROI variants used, discount rate justification, industry model rationale]

## Investment Cost Breakdown
[Phase 3 cost table with totals by category]

## Benefit Analysis
[Phase 4 benefit table with monetization methods and confidence]

## ROI Calculations
### Basic ROI
### Annualized ROI
### Cumulative ROI by Period
### Multi-Period ROI (NPV-Based)
### Risk-Adjusted ROI
### Industry-Specific Metrics
[Phase 5 + 6 calculations]

## ROI Timeline
[Diagram 1 — cumulative ROI over time]

## Cost-Benefit Breakdown
[Diagram 2 — costs vs benefits by category]

## Scenario Analysis
[Phase 7 scenario table + Diagram 3]

## Expected ROI (Probability-Weighted)
[Weighted calculation]

## Sensitivity Analysis
[Phase 8 tables + Diagram 4 tornado + break-even thresholds]

## Investment Comparison
[Phase 9 comparison table + ranking (if multiple options)]

## Recommendations
[Prioritized actions traced to specific findings]

## Sources
[Numbered list of web sources]

## Assumptions & Limitations
[Explicit list with confidence levels]

Present for user approval. Save only after explicit confirmation.

Generation rules

Per the autonomous-research mixin, plus:

• Calculations: Must be mathematically correct — formulas must produce stated results
• Benchmarks: Must cite source for industry benchmarks — never fabricate ROI ranges
• Confidence: Every estimate must have a confidence level (High/Medium/Low)
• Specificity: "TCO savings of $180K/year from reduced infrastructure costs" not "significant savings"
• Language: Respond and generate in the user's language unless specified otherwise

Failure behavior

Situation	Behavior
No investment context	Enter interview mode — ask what investment to model ROI for
Context too vague	Enter interview mode — ask targeted questions
Cannot estimate costs	Report gap, use ranges with low confidence labels
Investment amount unknown	Research comparable investments, propose range
Single scenario only	Produce all three with noted assumptions
Industry model not applicable	Use general ROI model, note limitation
CBA/business case malformed	Ask user to verify, attempt partial import
mmdc / web search failures	See diagram-rendering and autonomous-research mixins
Out-of-scope request	"This skill models ROI for investments. [Request] is outside scope."

Self-check

Before presenting output, verify:

[] Investment costs fully itemized (initial + ongoing + hidden)
[] Benefits itemized with monetization technique and confidence
[] Sunk costs explicitly excluded
[] Basic ROI calculated correctly
[] Annualized ROI calculated for multi-year investments
[] Cumulative ROI by period shows payback point
[] Risk-adjusted ROI with documented risk factors
[] Expected ROI probability-weighted across 3 scenarios
[] Multi-period ROI with justified discount rate
[] 3 scenarios defined with distinct, documented assumptions
[] Sensitivity analysis on 3-5 key drivers
[] Break-even thresholds identified for key variables
[] Payback period calculated
[] Industry-specific metrics included where applicable
[] All 4 Mermaid diagrams render valid syntax (per diagram-rendering mixin)
[] Sources listed for benchmarks and claims (per autonomous-research mixin)
[] Assumptions labeled with confidence levels (per autonomous-research mixin)
[] All calculations are mathematically consistent