Most financial models founders bring to fundraising are decorative. They look like spreadsheets but they're really hope expressed in pivot tables. An investor who knows what they're looking at will pick that up in the first three questions, and the meeting will go cold.
A defensible model is structured differently. It treats forecasting as a logical chain — drivers → assumptions → outputs — where every output can be traced back to a driver, and every driver has a defensible justification.
Drivers, not cells
The most common failure mode: founders edit forecast cells directly. They want ARR to hit $4M by month 24, so they type $4M into the month-24 cell. The connections to underlying drivers break, and the model becomes impossible to defend under pushback.
The right structure: every output cell is a formula over a set of named drivers. When an investor asks "why does ARR hit $4M by month 24," you walk through the drivers:
- 8 AEs hired by month 18.
- 90 days ramp to full productivity.
- $200K average quota at full ramp.
- 30% close rate on $400K of pipeline per AE per quarter.
Math: 8 × $200K × 4 quarters × 30% close rate = $4M in new ARR over the year. Add starting ARR, subtract churn, get to $4M by month 24.
Now any one of those drivers can be pushed back on, and you have a coherent response. "What if close rate is 20%?" — you flex the cell, model updates everywhere, and the new number is your honest answer.
The five questions every partner asks
Investors who actually read models ask roughly the same five questions:
1. "How do you get from where you are to your year-1 revenue target?"
Answer: pipeline math. Number of reps (or sources of leads), quota or conversion per source, close rate, ACV. Walk the model from current pipeline to projected closed revenue.
2. "What does churn assume?"
If your model has 0% churn, you have a problem. Honest assumptions:
- Logo churn: 5–10%/year early, 3–5%/year mature.
- Revenue churn (net of expansion): -20% to +50%/year depending on category and stage.
Show your assumption. Cite a current cohort or comparable benchmark.
3. "When does the business get to profitability?"
For most VC-scale businesses, the answer is "we don't optimize for profitability — we optimize for growth." That's fine. Be ready to answer:
- What's your burn rate at peak?
- At what ARR do you hit Rule of 40?
- At what point does net burn turn (assuming you stop hiring)?
These three questions are what investors actually want when they ask about profitability.
4. "What's the most aggressive driver in your model?"
Investors are testing whether you can identify your own optimism. Honest answer: "Our ramp time assumes 90 days, which is faster than the industry standard 4–6 months for our category. We've achieved 90 days with our first 3 reps because [reason], and we believe that holds with the next 5 hires, but it's the assumption we'd watch most carefully."
A founder who can't name their most aggressive assumption is a founder who hasn't stress-tested their own model.
5. "What does the model look like if your aggressive assumption breaks?"
This is the killer. Be ready to flex the most-aggressive driver and show the result. If your ARR target moves from $4M to $2.8M, what does that mean for runway, hiring, and the next raise?
If you can't answer this on the spot, you don't really understand your model.
What to include and what to leave out
Include
- Monthly P&L: revenue, COGS, gross profit, opex by function, EBITDA, cash.
- Driver sheet: every assumption that propagates into the model — sales metrics, retention, hiring plan, pricing.
- Hiring plan: month-by-month headcount by function, with start dates and fully-loaded cost.
- Scenario toggle: base case + at least one downside case where the most-aggressive assumption misses by 30%.
Leave out
- 5-year projections. Anything past month 24 is fiction. Have the math, but don't lead with it.
- Synergy calculations. "If our cross-sell hits 20% expansion..." Either model it as a driver or don't include it.
- Aspirational pricing changes. "We're moving to enterprise pricing in year 2" is a hypothesis, not a forecast.
- Pre-revenue companies showing 100% YoY growth from zero. Just show absolute numbers; ratios from a tiny base are misleading.
Two slides, not a workbook
In the room, you have two slides on the financials:
Slide 1: Headline trajectory. ARR by quarter, plotted against your plan, plus a downside scenario. Caption explains the most-aggressive driver: "Plan assumes 90-day ramp; downside flexes to 5-month ramp."
Slide 2: Hiring plan. Month-by-month headcount by function for the next 18 months, with start dates and the GTM/non-GTM split. This is where investors see whether your spend matches your revenue plan.
Everything else — the workbook itself — sits in the data room and is sent to investors who actually want it. Most won't.
The model as a fundraising weapon
A well-built model serves three purposes during a raise:
The cleanest demonstration of competence in a partner meeting isn't the size of your model. It's the speed and clarity with which you can answer "what if" questions. Most founders fumble those questions. The ones who close fast don't.
A note on tools
Excel/Google Sheets remain the lingua franca. If your model is in a more sophisticated tool, have an Excel export ready — investors will ask. Tools that let you flex drivers in real time during the meeting (Arx Forecast, Causal, Mosaic) are useful if you're comfortable demonstrating them; if not, stick with what you know cold.
Whatever tool you use, the test is the same: can you flex any driver in 10 seconds and trace the impact through to runway? If yes, you're in good shape. If not, build a better model.