| name | vc-fund |
| description | VC fund construction, portfolio math, and LP reporting. Activate when the user mentions
fund construction, portfolio construction for VC, reserve strategy, follow-on allocation,
power law, fund returns, TVPI, DPI, RVPI, J-curve, net IRR, management fee, carried
interest, GP commitment, fund economics, LP reporting, capital call schedule, vintage
benchmarking, PME, or asks about building or managing a venture fund.
|
VC Fund Construction & Portfolio Math
I analyze venture capital at the fund level — portfolio construction, reserve strategy, fund economics, and LP reporting. VC is a power-law business: fund returns are driven by a small number of outlier outcomes. Every portfolio decision should be evaluated through that lens — the question is not "will this company succeed?" but "can this company generate a fund-returning outcome?"
Scope & Boundaries
What this skill DOES:
- Design fund portfolio construction (number of investments, check sizes, reserves)
- Model fund economics (management fee, carry, GP commitment, fund lifecycle)
- Calculate fund metrics (gross/net IRR, TVPI, DPI, RVPI, PME)
- Analyze the J-curve and cash flow timing for LPs
- Design reserve and follow-on strategies
- Build LP reporting packages (quarterly reports, capital account statements)
- Benchmark fund performance against vintage peers
- Model power-law return distributions and their implications
Use a different skill when:
- Evaluating a specific early-stage deal →
/vc-early
- Analyzing a growth-stage company →
/vc-growth
- PE fund analytics →
/pe-buyout or /secondaries
- LP portfolio allocation →
/portfolio or /wealth
Available Tools
| Tool | Command | When to Use |
|---|
| VC Returns | python3 tools/vc_returns.py | Fund metrics (TVPI, DPI, RVPI, IRR) |
| IRR / NPV | python3 tools/irr.py | Cash flow-based fund returns |
| Monte Carlo | python3 tools/monte_carlo.py | Simulating fund return distributions |
Pre-Flight Checks
- Fund parameters: Target size, vintage, strategy (seed, early, multi-stage)
- Portfolio design: Target number of investments, initial check size, reserve ratio
- Fund economics: Management fee, carry, hurdle, GP commit, fund life
- Performance data (if existing fund): Investments, marks, realizations, cash flows
- Benchmarking context: Vintage year, strategy peer group
Phase 1: Portfolio Construction
Goal: Design the portfolio to maximize the probability of capturing power-law winners.
Portfolio Parameters
Fund size: $[X]M
Management fee (total): ~[X]% of committed over fund life = $[X]M
Investable capital: $[X]M (fund size - cumulative fees)
Target investments: [X] companies
Initial check size: $[X]M (avg)
Reserve ratio: [X]% of investable capital for follow-ons
Initial deployment: $[X]M ([X]% of investable)
Follow-on reserves: $[X]M ([X]% of investable)
Deployment period: [X] years
Portfolio Construction Models
Spray and pray (high count):
- 40-60+ investments, $500K-2M checks
- Maximize shots on goal, minimal follow-on
- Works at seed; requires high hit rate or massive outlier
Concentrated (low count):
- 15-25 investments, larger checks
- Deep diligence, significant follow-on reserves (40-50%)
- Works at Series A+; requires better picking ability
Barbell:
- 30-40 initial investments + concentrated follow-on into top 5-10
- Broad initial funnel, then double down on winners
- Most common multi-stage approach
Reserve Strategy
Follow-on allocation decision framework:
| Signal | Action |
|--------|--------|
| Strong performer, inside round, pro-rata available | Follow on at full pro-rata |
| Strong performer, competitive round at step-up | Follow on at half pro-rata |
| Moderate performer, flat/small step-up | Selective — only if new info is positive |
| Underperformer, needs bridge | Generally pass unless strategic value |
Reserve per company:
Initial check: $[X]M
Series A follow-on: $[X]M (target [X]% pro-rata)
Series B follow-on: $[X]M (target [X]% pro-rata)
Total reserved per co: $[X]M ([X]x initial check)
Phase 2: Fund Economics
Goal: Model the fee structure and GP/LP economics over the fund lifecycle.
Management Fee:
Commitment period (Years 1-5): [X]% on committed capital
Post-commitment (Years 6-10+): [X]% on invested (or declining)
Total fees over fund life: ~$[X]M ([X]% of committed)
Carried Interest:
Carry rate: [X]% (standard: 20%)
Preferred return (hurdle): [X]% (standard: 8%)
Catch-up: [X]% to GP after pref met
Clawback: [Y/N]
Distribution model: Deal-by-deal or whole fund
GP Commitment: $[X]M ([X]% of fund, standard: 1-5%)
Fund lifecycle cash flows:
| Year | Calls | Distributions | Net CF | Cumulative |
|------|-------|--------------|--------|-----------|
| 1 | ($[X]M) | $0 | ($[X]M) | ($[X]M) |
| 2 | ($[X]M) | $0 | ($[X]M) | ($[X]M) |
| 3 | ($[X]M) | $[X]M | ($[X]M) | ($[X]M) |
| ... | | | | J-curve trough |
| 7 | ($[X]M) | $[X]M | $[X]M | ($[X]M) |
| 10 | $0 | $[X]M | $[X]M | $[X]M |
Phase 3: Power Law & Fund Return Math
Goal: Understand the return distribution and what it takes to return the fund.
Power Law Distribution
Typical VC fund outcome distribution (30 investments):
0-1x return: 15-20 companies (50-65% of portfolio)
1-3x return: 5-8 companies (20-25%)
3-10x return: 3-5 companies (10-15%)
10-50x return: 1-2 companies (3-7%)
50x+ return: 0-1 company (0-3%)
Implication: The top 1-2 investments generate 50-80% of fund returns.
The rest of the portfolio provides optionality and learning, not returns.
Fund Return Targets
Fund-returning investment analysis:
Fund size: $[X]M
Ownership at exit needed to return fund: [X]% at $[X]M exit
For a $100M fund:
A $1B exit with 15% ownership = $150M = 1.5x fund
A $2B exit with 10% ownership = $200M = 2.0x fund
Need 2-3 of these to make a top-quartile fund (3x+ net)
Minimum viable outcome per investment:
Initial check $[X]M → need $[X]M back for fund math to work
That requires [X]x gross return on the check
Phase 4: Fund Metrics & Benchmarking
Goal: Calculate and benchmark fund performance.
Fund metrics:
Gross TVPI = (NAV + Distributions) / Called Capital = [X]x
Net TVPI = (NAV + Distributions - Fees - Carry) / Called Capital = [X]x
DPI = Distributions / Called Capital = [X]x (cash-on-cash, "the real number")
RVPI = NAV / Called Capital = [X]x (unrealized, "the hope number")
Net IRR = [X]% (time-weighted, accounts for J-curve)
PME (Public Market Equivalent):
Kaplan-Schoar PME = Fund TVPI / Index TVPI over same period
PME > 1.0: outperformed public markets
PME > 1.3: strong outperformance
Vintage benchmarking:
| Metric | Fund | Top Quartile | Median | Bottom Quartile |
|--------|------|-------------|--------|-----------------|
| Net TVPI | [X]x | >[X]x | [X]x | <[X]x |
| Net IRR | [X]% | >[X]% | [X]% | <[X]% |
| DPI | [X]x | >[X]x | [X]x | <[X]x |
Run: python3 tools/vc_returns.py for fund metrics calculation
Phase 5: LP Reporting
Goal: Build the quarterly LP report.
Quarterly Report Contents:
1. Fund summary (vintage, size, invested, remaining, NAV)
2. Portfolio overview (investments, exits, write-offs)
3. Performance metrics (TVPI, DPI, RVPI, IRR, PME)
4. Capital account statement (calls, distributions, NAV by LP)
5. Portfolio company updates (top 5-10 companies, key metrics)
6. Pipeline and market commentary
7. ESG / impact reporting (if applicable)
Quality Gates
Hard Constraints
- NEVER calculate fund IRR without accounting for management fees and carry
- NEVER benchmark a seed fund against a growth fund — use strategy-matched peers
- ALWAYS show DPI alongside TVPI — unrealized gains are not cash
- ALWAYS test whether the portfolio construction allows for fund-returning outcomes
Common Pitfalls
- Over-reserving for follow-ons — leaving 50% for follow-ons means only 50% for new deals
- Following on into losers — reserves should go to winners, not bridges for struggling companies
- Ignoring the J-curve — LPs need to know when cash comes back, not just how much
- TVPI without DPI — a fund with 3x TVPI and 0.5x DPI is mostly unrealized marks
- Strategy drift — a seed fund doing Series B follow-ons is a different fund than marketed
Related Skills
/vc-early — evaluating individual early-stage deals
/vc-growth — evaluating growth-stage companies
/secondaries — LP secondaries and fund restructuring
/attribution — fund performance attribution