How to Build a Financial Model in Google Sheets

Every startup needs a financial model. Not because your predictions will be accurate (they won't be), but because the process forces you to think through how your business actually makes money.

You don't need expensive software. Google Sheets is free, shareable, and sufficient for most early-stage startups. The model you build will help you plan hiring, understand your runway, and answer investor questions with real numbers instead of guesses.

This guide walks through building a basic 3-year financial model from scratch, covering the tabs you need, how to structure your assumptions, and what actually matters to investors.

A clean financial model has 5-6 tabs. Don't overcomplicate it.

1. Assumptions All your input variables in one place. Pricing, growth rates, hiring plans. Change assumptions here, and everything else updates automatically.

2. Revenue Your revenue model built from your assumptions. How many customers, at what price, equals how much revenue.

3. Expenses All costs: salaries, software, marketing, office, legal, etc. Monthly for Year 1, quarterly or annually for Years 2-3.

4. P&L (Profit & Loss) Revenue minus expenses, showing whether you're profitable and by how much. Also called an Income Statement.

5. Cash Flow When money actually moves. Revenue recognized isn't always cash received. This shows if you'll run out of money.

Optional: Headcount If you have or will have employees, a separate tab for hiring plan by role and department helps track your biggest expense.

Start simple. You can add complexity when you understand the basics.

The Assumptions tab is the control center. Everything else should reference it.

What goes here:

Pricing assumptions:

• Price per unit/month/seat
• Annual vs monthly pricing mix
• Discount rates (if any)

Customer assumptions:

• Starting customers
• Monthly customer growth rate (or new customers per month)
• Churn rate (percentage of customers who leave)

Cost assumptions:

• Average salary by role
• Software costs
• Marketing spend as % of revenue or fixed amount
• Cost of goods sold (if applicable)

Growth assumptions:

• Month-over-month growth rate by year
• When growth rates change

Formatting tips:

• Use clear labels (not just "20%" but "Monthly growth rate Y1")
• Group related assumptions together
• Highlight cells that are key drivers
• Include units ($/month, %, headcount)

Every number in your model should trace back to this tab. If you can't find where a number comes from, you've built it wrong.

Revenue is built from your assumptions, not typed in manually.

For SaaS businesses:

Month 1 customers = Starting customers from Assumptions
Month 2 customers = Month 1 customers * (1 + growth rate) * (1 - churn rate)
Revenue = Customers * Price per customer

Structure the tab:

• Columns = Months (Month 1, Month 2... through Month 36)
• Rows = Metrics (New customers, churned customers, total customers, revenue)

For different revenue types:

• Monthly recurring: Customers * monthly price
• Annual contracts: Spread over 12 months (for recognized revenue)
• Transaction fees: GMV * take rate
• Usage-based: Users * average usage * price per unit

Multiple revenue streams: If you have tiers (Basic, Pro, Enterprise), model each separately and sum them.

Bottom-up vs top-down: Bottom-up (customers * price) is more credible than top-down (TAM * market share). Investors will challenge hand-wavy market share assumptions.

Expenses fall into a few categories. Be realistic, not optimistic.

People costs (usually 60-80% of startup expenses):

• Salaries by role
• Payroll taxes (~15% on top of salary)
• Benefits (health insurance, etc.)
• Contractors

Software and tools:

• AWS/cloud hosting
• SaaS subscriptions
• Development tools

Marketing and sales:

• Paid advertising
• Content and creative
• Events
• Sales commissions

Operations:

• Office/coworking
• Legal and accounting
• Insurance
• Travel

Cost of goods sold (COGS):

• Payment processing fees
• Hosting costs that scale with usage
• Customer support costs

Typical mistakes:

• Forgetting payroll taxes and benefits
• Underestimating legal costs
• Assuming marketing costs stay flat as you grow
• Not accounting for hiring ramp time (new employees aren't productive immediately)

The P&L (Profit and Loss) pulls together revenue and expenses to show if you're making or losing money.

Standard structure:

Revenue
- Cost of Goods Sold
= Gross Profit
- Operating Expenses (salaries, marketing, etc.)
= Operating Income (EBITDA)
- Depreciation & Amortization
- Interest
- Taxes
= Net Income

For early-stage startups, focus on:

• Revenue (top line)
• Gross margin (Revenue - COGS) / Revenue
• Operating expenses
• Net income (or loss)

What investors look at:

• Gross margin: Should be 60-80%+ for SaaS
• Burn rate: How much you're losing monthly
• Path to profitability: When do you expect to break even?

Formatting:

• Monthly columns for Year 1
• Quarterly or annual for Years 2-3
• Summary row with totals by year
• Color negative numbers red

Cash flow is different from P&L. Revenue recognized isn't the same as cash received. You can be "profitable" and still run out of money.

Key concepts:

Collections vs revenue: If customers pay annually upfront, you have cash now but recognize revenue monthly. If customers pay net-30, you have revenue now but cash in 30 days.

Working capital: Money tied up in operations. Accounts receivable (money owed to you) and accounts payable (money you owe) affect cash timing.

Simple cash flow structure:

Starting cash
+ Cash from operations (adjusted for timing)
+ Cash from financing (investment)
- Cash out (expenses, actually paid)
= Ending cash

Runway calculation: Ending cash / monthly burn = months of runway

If runway drops below 6 months, you need to raise money or cut costs.

For early-stage: Keep cash flow simple. The main insight is "do we have enough money?" Complex working capital modeling can wait.

Hockey stick projections with no basis Showing flat growth then sudden explosion looks like wishful thinking. Growth should connect to specific actions.

Forgetting about churn New SaaS founders often model customer acquisition without customer loss. If you're not including churn, your projections are fantasy.

Underestimating hiring costs Salaries aren't the whole picture. Add payroll taxes, benefits, equipment, and ramp time. A $100K salary costs you $120-130K.

Ignoring seasonality Some businesses have slower months. B2B slows in December. Consumer spikes in Q4. Flat monthly projections might not reflect reality.

Not stress-testing What if growth is 50% of your projection? Does the model still work? Test pessimistic scenarios.

Too much precision Projecting to the dollar three years out is false precision. Round numbers are fine. The point is directional understanding, not exact prediction.

If you're raising money, your model will be reviewed. Here's what makes it credible.

Clearly labeled assumptions: Investors will want to tweak your assumptions and see what happens. Make this easy.

Conservative growth scenarios: Show a base case, optimistic case, and conservative case. Investors will mentally discount your numbers anyway.

Cohort analysis (for SaaS): Show how customers acquired in Month 1 behave over time. This reveals true retention.

Unit economics:

• CAC: Customer Acquisition Cost
• LTV: Lifetime Value of a customer
• LTV:CAC ratio (3:1 or better is healthy)
• Payback period (how long to recoup CAC)

Clean formatting:

• Consistent fonts and colors
• Clear section headers
• Summary sheet that highlights key metrics
• Print-friendly layout

What you'll be asked:

• "What happens if growth is 50% slower?"
• "When do you need to raise again?"
• "What's your path to profitability?"
• "How did you get these assumptions?"

Be ready to defend every number.

Google Sheets works until it doesn't. Here's when to consider alternatives.

Stay with Sheets if:

• Pre-seed or seed stage
• Simple business model
• You want maximum flexibility
• Budget is limited

Consider upgrading if:

• Multiple people need to edit simultaneously (Sheets can handle this, but specialized tools do it better)
• You need automatic integrations with accounting software
• Your model has become unwieldy (100+ tabs, thousands of rows)
• Investors expect more sophisticated reporting

Alternatives:

• Causal: Visual modeling, good for scenarios
• Runway: Built for startup financial planning
• Jirav: Connects to QuickBooks/Xero
• Pry: Built specifically for startup finance

Cost reality: These tools cost $100-500+/month. Sheets is free. The upgrade is worth it when the time saved exceeds the cost.

Foundra has templates for basic financial planning if you want structure without complexity.

• A financial model forces you to think through how your business makes money. That's its real value.
• Use 5 tabs: Assumptions, Revenue, Expenses, P&L, Cash Flow. Start simple.
• All numbers should trace back to your Assumptions tab. Nothing should be hardcoded.
• Build revenue bottom-up (customers * price), not top-down (market share).
• Include churn, payroll taxes, and realistic ramp times. These are commonly underestimated.
• Test pessimistic scenarios. What if growth is 50% slower?
• Google Sheets is sufficient for most early-stage startups. Upgrade when complexity demands it.

How far out should my financial model project?

3 years is standard. Year 1 should be monthly, Years 2-3 can be quarterly or annual. Anything beyond 3 years is speculative fiction.

How accurate do my projections need to be?

They don't need to be accurate. They need to be reasonable and defensible. Investors know you're guessing. They want to see that you've thought through the drivers.

What growth rate should I assume?

Depends on your stage and market. 5-10% month-over-month is aggressive but achievable for early-stage SaaS. Higher than 10% sustained requires explanation. Base it on evidence if you have any traction.

Should I hire someone to build my financial model?

For Series A and beyond, maybe. For pre-seed/seed, build it yourself. You need to understand every number because you'll be questioned on them. CFO-for-hire services can help if you're not numbers-oriented.

How often should I update my model?

Monthly, comparing actuals to projections. Quarterly, revisit your assumptions. Major updates when you have new information that changes the trajectory.