mcpField
R
returns-analysis

returns-analysis

github.com/anthropicsData & databases40.9k1.1k installsUpdated today
Repository

About returns-analysis

A returns-analysis skill for private-equity deal teams. It structures investment assumptions into base-case returns, sensitivities, scenarios, and an IC-ready summary.

Key features

  1. 01Deal and financing assumptions
  2. 02MOIC and IRR calculations
  3. 03Returns attribution waterfall
  4. 04Two-way sensitivity tables
  5. 05Bull, base, and bear scenarios
  6. 06Excel and IC summary outputs

Use cases

  1. 01Assess a proposed acquisition's returns
  2. 02Compare entry and exit multiple outcomes
  3. 03Test leverage and growth assumptions
  4. 04Prepare an investment committee summary

Install

Copy the command and run it in a terminal.

$npx skills add https://github.com/anthropics/financial-services --skill returns-analysis

After installing, type/returns-analysisin Claude Code to confirm it loaded.

Details

From the project README

Returns Analysis

Workflow

Step 1: Gather Deal Inputs

Ask for (or extract from prior analysis):

Entry:

  • Entry EBITDA (LTM or NTM)
  • Entry multiple (EV / EBITDA)
  • Enterprise value
  • Net debt at close
  • Equity check size
  • Transaction fees & expenses

Financing:

  • Senior debt (x EBITDA, rate, amortization)
  • Subordinated debt / mezzanine (if any)
  • Total leverage at entry (x EBITDA)
  • Equity contribution

Operating Assumptions:

  • Revenue growth rate (annual)
  • EBITDA margin trajectory
  • Capex as % of revenue
  • Working capital changes
  • Debt paydown schedule

Exit:

  • Hold period (years)
  • Exit multiple (EV / EBITDA)
  • Exit EBITDA (calculated from growth assumptions)
Step 2: Base Case Returns

Calculate:

MetricValue
Entry EV
Equity invested
Exit EBITDA
Exit EV
Net debt at exit
Exit equity value
MOIC
IRR
Cash-on-cash

Show the returns waterfall:

  • EBITDA growth contribution
  • Multiple expansion/contraction contribution
  • Debt paydown contribution
  • Fee/expense drag
Step 3: Sensitivity Tables

Build 2-way sensitivity matrices:

Entry Multiple vs. Exit Multiple

Exit 6xExit 7xExit 8xExit 9xExit 10x
Entry 7x
Entry 8x
Entry 9x
Entry 10x

EBITDA Growth vs. Exit Multiple (at fixed entry)

Leverage vs. Exit Multiple (at fixed entry and growth)

Hold Period vs. Exit Multiple

Show both IRR and MOIC in each cell (IRR / MOIC format).

Step 4: Scenario Analysis

Build 3 scenarios:

BullBaseBear
Revenue CAGR
Exit EBITDA margin
Exit multiple
Exit EBITDA
MOIC
IRR
Step 5: Output
  • Excel workbook with:
    • Assumptions tab
    • Returns calculation
    • Sensitivity tables (formatted with conditional coloring)
    • Scenario summary
  • One-page returns summary suitable for IC deck

Key Formulas

  • MOIC = Exit Equity Value / Equity Invested
  • IRR = solve for r: Equity Invested × (1 + r)^n = Exit Equity Value (adjust for interim cash flows)
  • Returns attribution:
    • Growth: (Exit EBITDA - Entry EBITDA) × Exit Multiple / Equity
    • Multiple: (Exit Multiple - Entry Multiple) × Entry EBITDA / Equity
    • Leverage: Debt paydown over hold period / Equity

Important Notes

  • Always show returns both gross and net of fees/carry where applicable
  • Management rollover and co-invest change the equity check — ask if relevant
  • Dividend recaps or interim distributions affect IRR significantly — include if planned
  • Don't forget transaction costs (typically 2-4% of EV) — they reduce Day 1 equity value
  • Tax considerations (asset vs. stock deal, 338(h)(10) election) can materially affect after-tax returns

Comments

Markdown supported