Startup CFO

Monte Carlo Says We Have 18 Months of Runway — With 94% Confidence

94% confidence on 18-month runwayStartup Finance6 min read

Key Takeaway

10,000 Monte Carlo simulations gave us probability-weighted runway projections for PyratzLabs — 94% confidence on 18 months, 73% on 24 months — replacing gut-feel estimates with board-ready data.

The Problem

"How much runway do we have?"

Every founder gets this question. From investors, from board members, from their own anxiety at 3 AM. And almost every founder answers it the same way: take current cash, divide by monthly burn, get a number.

That number is wrong.

It's wrong because it assumes burn is constant. It isn't — it grows with headcount, infrastructure, and unexpected costs. It's wrong because it ignores revenue variance. Some months Billy does 800K€ GMV, some months 1.2M€. It's wrong because it doesn't account for the probability of a client churning, or a funding round closing late, or a hiring plan accelerating.

A single-point runway estimate is a lie you tell yourself because probability distributions are harder to explain in a board meeting.

At PyratzLabs, I manage a portfolio of companies — Zama (FHE encryption), Artificial-Lab (AI studio), and Billy (5.5M€ GMV in Q1 2026). Each has different burn profiles, revenue curves, and risk factors. I needed runway projections that accounted for uncertainty, not pretended it didn't exist.

The Solution

The Monte Carlo Financial Simulator on Mr.Chief. I input current financials, growth assumptions with ranges (not single points), and variance parameters. The agent runs 10,000 simulations, each with randomized inputs drawn from the specified distributions, and outputs a probability distribution of runway scenarios.

The result isn't "we have 18 months." It's "we have 94% probability of 18+ months, 73% probability of 24+ months, and 12% probability of running out in 12 months." That's a sentence a board can make decisions with.

The Process

yamlShow code

# Monte Carlo Runway Simulation — On demand
name: monte-carlo-runway
trigger: "simulate runway"
channel: telegram

task: |
  Run Monte Carlo simulation for runway projections.

  INPUT PARAMETERS:
  Current cash: €2.8M
  Monthly burn (base): €145K
  Burn growth rate: 3-5% per month (uniform distribution)
  Monthly revenue (current): €85K
  Revenue growth: 8-15% per month (normal distribution, μ=11%, σ=3%)

  RISK EVENTS (Bernoulli trials per month):
  - Major client churn: 5% probability, -€25K/month revenue impact
  - Hiring acceleration: 10% probability, +€30K/month burn increase
  - Unexpected cost (legal, infra): 3% probability, €50-150K one-time
  - Follow-on funding: 8% probability after month 12, €1-3M injection

  SIMULATIONS: 10,000
  HORIZON: 36 months

  OUTPUT:
  - Probability distribution of runway (months)
  - Confidence levels: 50%, 75%, 90%, 95%
  - Sensitivity analysis: which variable impacts runway most
  - Scenario comparison: base vs optimistic vs pessimistic
  - Board-ready summary

The simulation engine:

pythonShow code

import numpy as np

def monte_carlo_runway(params, n_simulations=10000, horizon_months=36):
    results = []

    for sim in range(n_simulations):
        cash = params["current_cash"]
        monthly_revenue = params["current_revenue"]
        monthly_burn = params["current_burn"]
        runway_months = 0

        for month in range(1, horizon_months + 1):
            # Revenue growth (normal distribution)
            rev_growth = np.random.normal(
                params["rev_growth_mean"],
                params["rev_growth_std"]
            )
            monthly_revenue *= (1 + rev_growth)

            # Burn growth (uniform distribution)
            burn_growth = np.random.uniform(
                params["burn_growth_min"],
                params["burn_growth_max"]
            )
            monthly_burn *= (1 + burn_growth)

            # Risk events
            if np.random.random() < params["churn_probability"]:
                monthly_revenue -= params["churn_impact"]

            if np.random.random() < params["hiring_accel_probability"]:
                monthly_burn += params["hiring_accel_impact"]

            if np.random.random() < params["unexpected_cost_probability"]:
                one_time = np.random.uniform(
                    params["unexpected_cost_min"],
                    params["unexpected_cost_max"]
                )
                cash -= one_time

            if month > 12 and np.random.random() < params["funding_probability"]:
                funding = np.random.uniform(
                    params["funding_min"],
                    params["funding_max"]
                )
                cash += funding

            # Net cash flow
            net = monthly_revenue - monthly_burn
            cash += net

            if cash <= 0:
                runway_months = month
                break
        else:
            runway_months = horizon_months  # Survived full horizon

        results.append(runway_months)

    return np.array(results)

The board-ready output:

View details

📊 MONTE CARLO RUNWAY ANALYSIS — PyratzLabs
Simulations: 10,000 | Horizon: 36 months

RUNWAY PROBABILITY DISTRIBUTION:
  ≥12 months:  98.2% ✅
  ≥18 months:  94.1% ✅
  ≥24 months:  73.4%
  ≥30 months:  52.8%
  ≥36 months:  41.2%

PERCENTILE ANALYSIS:
  P5  (worst 5%):    14 months
  P25 (conservative): 21 months
  P50 (median):       28 months
  P75 (optimistic):   34 months
  P95 (best case):    36+ months (cash-flow positive)

BREAK-EVEN ANALYSIS:
  Month revenue exceeds burn (median): Month 16
  Probability of reaching profitability within 24mo: 67.3%

SENSITIVITY ANALYSIS (impact on median runway):
  Revenue growth ±5%:     ±6.2 months  ← HIGHEST IMPACT
  Burn growth ±2%:        ±3.8 months
  Client churn rate ±3%:  ±2.4 months
  Funding probability:    ±4.1 months (if removed entirely: -4.1mo)

SCENARIO COMPARISON:
                    P(≥18mo)   P(≥24mo)   Median
  Base case         94.1%      73.4%      28 months
  Revenue +20%      98.8%      91.2%      34 months
  Major client churn 87.2%     58.1%      22 months
  Hiring freeze      97.4%     84.8%      32 months
  No follow-on       91.3%     64.2%      24 months

🎯 BOARD SUMMARY:
  With 94% confidence, we have 18+ months of runway.
  Revenue growth rate is the single biggest lever — a 5%
  increase in monthly growth adds 6 months of runway.

  Key risk: client concentration. One major churn event
  drops 24-month probability from 73% to 58%.

  Recommendation: maintain current burn, accelerate revenue
  diversification, begin fundraising conversations at month 9
  (6 months before median break-even).

The Results

Metric	Single-Point Estimate	Monte Carlo
Runway answer	"19 months"	"94% chance of 18+, 73% chance of 24+"
Accounts for revenue variance	No	Yes (normal distribution)
Accounts for burn growth	No	Yes (uniform distribution)
Models risk events	No	Yes (churn, hiring, unexpected costs)
Includes funding scenarios	No	Yes (Bernoulli trials after month 12)
Sensitivity analysis	Manual "what if"	Automated across all variables
Board confidence in number	Low ("what if…" questions)	High (probabilities answer "what if")
Time to produce	30 min in spreadsheet	45 seconds

The board meeting where I presented this was the first time nobody asked "but what if revenue doesn't grow that fast?" The probability distribution already answered that question. At every confidence level.

Try It Yourself

Gather your current financials: cash, monthly burn, monthly revenue
Define ranges, not single points: revenue growth between 8-15%, not exactly 11%
List your risk events with estimated probabilities and impact
Run 10,000 simulations — anything less and the tail risks are undersampled
Present the probability distribution, not a single number
Update monthly as actuals come in — the simulation gets more accurate over time

A single-point runway estimate is a comforting lie. A probability distribution is an uncomfortable truth. Boards deserve the truth.

Start free on Mr.Chief →

I don't tell my board we have 18 months of runway. I tell them we have 94% confidence in 18 months. The 6% is what keeps us sharp.

Monte Carlorunwaystartup-financeboard-reportingfinancial-modeling

Related case studies

Startup CFO

PyratzLabs Runway Simulator — What the Board Sees Every Quarter

Monte Carlo runway simulation with 10,000 scenarios replaced back-of-napkin estimates. How AI agents give our board probabilistic cash runway forecasts with confidence intervals.

P10/P50/P90 runway ranges vs single-point estimates, 15 min generation5 min read

Startup CFO

6-Month Cash Flow Forecast — Updated Every Monday Morning

AI agent builds rolling 6-month cash flow forecasts with Monte Carlo simulation. 1,000 scenarios, P10/P50/P90 confidence bands, automated weekly updates. Mr.Chief CFO agent.

Single-point forecast → 1,000-scenario probability distribution5 min read

Fund Manager

DCF Valuation of a Portfolio Company — Agent Says It's 40% Overvalued

An AI agent built a 5-year DCF model with sensitivity analysis in 5 minutes, flagging a portfolio company as 40% overvalued. Replacing $15K analyst fees with autonomous valuation.

5 min vs 2-4 weeks, €0 vs €15K4 min read

Want results like these?

Start free with your own AI team. No credit card required.

Start Free →Browse agents