Financial Analyst

Comprehensive financial analysis for investment decisions, budgeting, and business evaluation.

No API Required

Download Skill Package (.skill) View Source on GitHub Workflow

Table of Contents

Overview

Financial Analyst provides a systematic toolkit for evaluating financial health, investment opportunities, and business performance. It combines quantitative analysis with professional reporting across four core workflows:

Workflow Purpose
Financial Statement Analysis Calculate profitability, liquidity, and efficiency ratios with benchmarks
Investment Evaluation (DCF) NPV, IRR, WACC, and payback period calculations
Budget Variance Analysis Actual vs. budget comparison with root-cause decomposition
Sensitivity & Scenario Analysis Risk assessment through variable stress testing

Outputs include structured Markdown reports, visualization charts (PNG via matplotlib/seaborn), and summary tables with interpretations.

When to Use

  • Analyze financial statements – calculate key ratios and assess company health from balance sheet and income statement data
  • Evaluate investment opportunities – run DCF valuations, compute NPV/IRR, and compare against hurdle rates
  • Review budget performance – identify material variances and decompose them into price, volume, and mix components
  • Assess risk – perform single-variable sensitivity analysis, multi-scenario modeling, and break-even calculations

Prerequisites

  • Claude Code installed and running
  • The financial-analyst skill copied to ~/.claude/skills/
  • Financial data in JSON or CSV format (balance sheet, income statement, cash flow projections, or budget files)

No external API keys are required. The bundled Python script (financial_analyzer.py) uses standard libraries plus matplotlib/seaborn for charts.

How It Works

1. Data Ingestion

Provide financial data as JSON or CSV. Claude validates required fields (balance sheet, income statement) and checks for inconsistencies such as negative values, mismatched totals, or missing periods. If fields are missing, Claude reports which ones are absent before proceeding.

2. Ratio Calculation and Analysis

Depending on the workflow, Claude calculates one or more of the following:

  • Financial ratios – profitability (ROE, ROA, margins), safety/liquidity (current ratio, quick ratio, D/E, interest coverage), and efficiency (asset turnover, inventory turnover, receivables turnover).
  • Discounted cash flows – NPV, IRR, and payback period using projected cash flows and a discount rate.
  • Budget variances – line-by-line actual vs. budget comparison, decomposed into price, volume, and mix components.
  • Sensitivity ranges – single-variable and multi-variable stress tests on key assumptions.

3. Interpretation and Classification

Each metric is classified into one of three levels:

Level Color Meaning
Strong Green Exceeds benchmark; no action required
Acceptable Yellow Within normal range; monitor
Weak Red Below threshold; corrective action recommended

Industry context is applied when the user specifies a sector. Without a sector, general cross-industry benchmarks are used.

4. Visualization

Claude generates charts using matplotlib/seaborn:

  • Radar chart for ratio analysis (shows strengths and weaknesses at a glance)
  • Waterfall chart for budget variance (cumulative contribution of each line item)
  • Tornado chart for sensitivity analysis (variables ranked by NPV impact)
  • Bar/line charts for trend comparisons across periods

5. Report Generation

Results are formatted into a professional Markdown report using bundled templates (financial_report_template_en.md or financial_report_template_ja.md). Each report includes an executive summary, detailed tables, embedded charts, and actionable recommendations.

Key Formulas

Metric Formula
ROE Net Income / Total Equity
ROA Net Income / Total Assets
Current Ratio Current Assets / Current Liabilities
Quick Ratio (Current Assets - Inventory) / Current Liabilities
D/E Ratio Total Liabilities / Total Equity
NPV -Investment + Sum(CF_t / (1+r)^t) + Terminal Value
IRR Rate where NPV = 0 (Newton-Raphson iteration)
WACC (E/V x Re) + (D/V x Rd x (1 - Tc))
Variance % (Actual - Budget) / Budget x 100

Industry Benchmark Reference

Ratio thresholds vary significantly by sector. The table below provides representative benchmarks for common industries.

Profitability Benchmarks

Sector Gross Margin Operating Margin Net Margin ROE
Technology / SaaS 60–80% 15–30% 10–25% 15–30%
Manufacturing 25–40% 8–15% 5–10% 10–18%
Retail 25–35% 3–8% 2–5% 12–20%
Financial Services 40–60% 25–40% 15–30% 10–15%
Healthcare / Pharma 50–70% 15–25% 10–20% 12–22%
Utilities 30–50% 15–25% 8–15% 8–12%

Liquidity and Leverage Benchmarks

Sector Current Ratio Quick Ratio D/E Ratio Interest Coverage
Technology / SaaS 2.0–4.0 1.5–3.0 0.2–0.8 10+
Manufacturing 1.5–2.5 0.8–1.5 0.5–1.5 4–8
Retail 1.2–2.0 0.5–1.0 0.8–2.0 3–6
Financial Services N/A N/A 2.0–10.0 2–5
Healthcare / Pharma 1.5–3.0 1.0–2.0 0.3–1.0 6–12
Utilities 0.8–1.2 0.5–0.8 1.0–2.5 2–4

Mention the sector in your prompt (e.g., “This is a SaaS company”) to receive sector-specific interpretations.

DCF Step-by-Step Detail

Calculating WACC

WACC (Weighted Average Cost of Capital) is the discount rate used for DCF valuations. It blends the cost of equity and after-tax cost of debt weighted by their proportion in the capital structure.

WACC = (E/V x Re) + (D/V x Rd x (1 - Tc))

Where:
  E  = Market value of equity
  D  = Market value of debt
  V  = E + D (total firm value)
  Re = Cost of equity (often from CAPM: Rf + Beta x Market Risk Premium)
  Rd = Cost of debt (yield on existing debt or new issuance rate)
  Tc = Corporate tax rate

Typical input example:

Component Value
Equity value (E) $600M
Debt value (D) $400M
Cost of equity (Re) 12%
Cost of debt (Rd) 5%
Tax rate (Tc) 25%
WACC (600/1000 x 12%) + (400/1000 x 5% x 0.75) = 8.7%

Terminal Value Approaches

The DCF model projects explicit cash flows for a finite horizon (typically 5–10 years). Beyond that, a terminal value captures the remaining value. Two methods are supported:

1. Gordon Growth Model (Perpetuity Growth)

Terminal Value = CF_n x (1 + g) / (WACC - g)

Best used when the company is expected to grow at a stable rate indefinitely. The growth rate (g) should not exceed long-term GDP growth (typically 2–3%).

2. Exit Multiple Method

Terminal Value = Final Year EBITDA x Industry EV/EBITDA Multiple

Best used when comparable transactions or trading multiples are available. Common multiples by sector:

Sector Typical EV/EBITDA
Technology 15–25x
Manufacturing 8–12x
Retail 6–10x
Healthcare 12–18x
Utilities 8–12x

NPV Decision Framework

After computing NPV, IRR, and payback period, apply the following decision criteria:

Metric Accept Reject Notes
NPV > 0 < 0 Higher NPV = more value created
IRR > WACC < WACC Margin above WACC indicates safety buffer
Payback < Target period > Target period Simple payback ignores time value; use discounted payback for rigor

Usage Examples

Example 1: Financial ratio analysis 

Analyze these financial statements and calculate key ratios.
Data: financial_data.json
Focus on profitability and liquidity.

Claude reads the JSON, calculates all ratio categories, generates a radar chart, and outputs a report with severity-coded interpretations.

Example 2: DCF valuation 

Perform a DCF valuation for this acquisition target.
Initial investment: $50M
Projected cash flows: $10M, $15M, $20M, $22M, $25M over 5 years
Discount rate: 10%, terminal growth: 2%

Claude calculates NPV, IRR, payback period, and provides a go/no-go recommendation with sensitivity ranges.

Example 3: Budget variance analysis 

Compare actual.csv and budget.csv.
Identify all variances over 5% and provide root cause analysis.

Claude merges the two files, flags material variances, decomposes them into price and volume components, and generates a waterfall chart.

Example 4: Sensitivity analysis on a DCF model 

Run a sensitivity analysis on the factory investment DCF.
Vary revenue growth (+/-20%) and discount rate (+/-3pp).
Show a tornado chart and identify the break-even revenue level.

Claude produces a tornado chart ranking variables by NPV impact, a data table with NPV at each test point, and the break-even revenue where NPV equals zero.

Example 5: Multi-period trend analysis 

Analyze the financial statements for FY2022, FY2023, and FY2024.
Show year-over-year ratio trends and highlight any deteriorating metrics.

Claude calculates ratios for each period, generates trend line charts, and flags metrics that show consistent decline across periods.

Troubleshooting

Incomplete ratio results

Symptom: Some ratios appear as “N/A” or are missing from the report.

Solution: Ensure all required fields are present in the input data. The minimum required fields are: total_assets, current_assets, total_liabilities, current_liabilities, total_equity, revenue, cost_of_goods_sold, operating_income, and net_income. If a field is missing, the ratios that depend on it will be skipped.

NPV calculation seems incorrect

Symptom: NPV value does not match manual calculation.

Solution: Check that (1) the initial investment is entered as a positive number (the script negates it internally), (2) cash flows are in the same currency units as the investment, and (3) the discount rate is expressed as a decimal (e.g., 0.10 for 10%, not 10). Also verify whether terminal value is included – omitting it for a going-concern business will significantly understate value.

Charts not generated

Symptom: The report contains tables but no PNG chart files.

Solution: The financial_analyzer.py script requires matplotlib and seaborn. Install them with pip install matplotlib seaborn (or uv pip install matplotlib seaborn). If running in a headless environment, set the matplotlib backend with export MPLBACKEND=Agg before execution.

Tips & Best Practices

  • Provide complete data – the more fields you include (balance sheet, income statement, cash flow), the richer the analysis.
  • Specify the currency and period – this ensures correct formatting and period-over-period comparisons.
  • Use JSON for structured data – JSON input enables automatic field validation; CSV works but requires consistent column naming.
  • Combine workflows – run a DCF first, then apply sensitivity analysis to stress-test the key assumptions.
  • Check thresholds in context – ratio benchmarks vary by industry. Mention the sector if you want industry-specific interpretation.
  • IT System ROI Analyzer – ROI and TCO analysis specifically for IT investments
  • Project Plan Creator – project planning with budget and resource allocation