SaaS Financial Model: Predict Your Runway

The typical SaaS financial model starts as an honest attempt at planning and ends as an abandoned 12-tab workbook. A founder downloads a template, adds a few hundred assumptions, updates it for two weeks, realises it takes longer to maintain than it takes to actually run the business, and quietly stops touching it. The model lives in Google Drive, increasingly out of date, occasionally guilt-inducing.

The problem is not that financial models are useless. The problem is that most of them are designed for CFOs or fundraising decks — not for the founder who needs to know, every month, whether recurring revenue is growing fast enough, whether costs are creeping too fast, and how many months of runway remain if churn ticks up.

A model only helps if it gets used. A model only gets used if it is small enough to update in 20 minutes.

What a SaaS financial model actually is

A SaaS financial model is a structured way to project how a subscription business will perform over time — specifically: how recurring revenue will move, how costs will compound, how cash will change, and how much runway remains under different assumptions.

The key difference from a generic startup financial model is that SaaS financials are built around recurring revenue mechanics rather than general revenue lines. That means the engine is MRR movement (new subscriptions, upgrades, cancellations) rather than sales pipeline, services revenue, or product units. This narrower scope is what allows the model to stay genuinely minimal — the complexity of a SaaS business is already encoded in the MRR bridge. Y Combinator’s startup financial guidance consistently emphasises this point: build the simplest model that captures your actual business mechanics.

For the broader startup financial model framing that this sits inside, see Startup Financial Model: The Minimalist Guide for Founders.

Every forecast needs a clean MRR baseline. Get yours from Stripe in 90 seconds →

The 8 inputs that are enough

A useful SaaS financial model does not need 50 assumptions. It needs the inputs that actually move the business.

Revenue inputs:

Starting MRR — the recurring revenue base at the beginning of the period
New MRR — expected recurring revenue from new paying customers
Expansion MRR — expected additional revenue from upgrades
Churned MRR — expected recurring revenue lost to cancellations (split voluntary and failed payment where possible)

Cost inputs: 5. Fixed costs — infrastructure, SaaS subscriptions, tools (relatively stable month to month) 6. Variable costs — transaction fees, per-user infra costs, contractor work that scales with revenue or usage

Cash inputs: 7. Cash on hand — current bank balance 8. Scenario toggle — base, upside, or downside

That is the complete input set for most early-stage SaaS products. Every additional input after this set increases maintenance burden without proportionally increasing decision value — unless the specific input reflects a material cost or revenue driver that the existing inputs cannot capture.

The formulas that matter

Ending MRR

Ending MRR = Starting MRR + New MRR + Expansion MRR − Contraction MRR − Churned MRR

This is the revenue backbone of the model. It makes recurring revenue movement legible — which component is driving growth, which is leaking, and whether the dynamics are improving or deteriorating.

For a full breakdown of what belongs in each MRR component, see What Is MRR? The Clean Version.

Total costs

Total Costs = Fixed Costs + Variable Costs

For most early SaaS products, this grouping is sufficient. Fixed costs cover everything with a monthly recurring bill (hosting, software tools, subscriptions). Variable costs cover anything that scales with the business — transaction fees as a percentage of revenue, infra costs that grow with active users, or contractor work tied to specific deliverables.

Net burn

Net Burn = Total Costs − Cash Inflow

In a simplified SaaS model, cash inflow in a given month is approximately equal to recognised recurring revenue (MRR) plus any non-recurring cash received. For founders running on Stripe, the actual cash collected lags slightly from the MRR timing due to payment processing windows — close enough for planning purposes at early scale.

Runway

Runway = Cash on Hand / Monthly Net Burn

When revenue exceeds costs (net burn is negative), the business is self-sustaining in that period and runway calculation shifts from urgency to resilience — how much cash buffer exists in the event of a downturn. When burn is positive, runway is the most operationally important number on the sheet. KeyBanc Capital Markets’ SaaS Survey tracks median runway and burn multiples across private SaaS companies by ARR stage, which can help calibrate whether your burn rate is within normal operating range.

The minimal one-sheet structure

A single sheet with four sections is enough.

Section 1 — Revenue inputs: starting MRR, new MRR, expansion MRR, contraction MRR, churned MRR, ending MRR.

Section 2 — Cost inputs: fixed costs by category (infra, tools, other), variable costs, total costs.

Section 3 — Cash and runway: opening cash, net burn, closing cash, runway months.

Section 4 — Scenario toggle: a single cell or dropdown that adjusts the forecast inputs between base, upside, and downside assumptions without requiring separate sheet tabs.

The one-sheet constraint is a feature, not a limitation. When everything fits on one page, the model stays readable, stays used, and stays honest. A 12-tab workbook creates hiding places for stale assumptions; a one-sheet model forces confrontation with all inputs simultaneously.

A worked example: the minimal sheet in practice

A SaaS analytics product, month four. Starting inputs:

Starting MRR: €10,000
New MRR: €1,500
Expansion MRR: €600
Contraction MRR: €200
Churned MRR: €500 (€280 voluntary, €220 failed payment)
Fixed costs: €6,000 (€2,400 Railway + Vercel infrastructure, €1,200 SaaS tools, €1,200 Brevo and other subscriptions, €1,200 other)
Variable costs: €2,000 (Stripe fees ~1.5% of revenue, Claude API usage, contractor)
Cash on hand: €45,000

Step 1: Forecast ending MRR

Ending MRR = 10,000 + 1,500 + 600 − 200 − 500 = 11,400

Step 2: Calculate total costs

Total Costs = 6,000 + 2,000 = 8,000

Step 3: Estimate net burn

Simplified cash view (MRR used as revenue proxy):

Net Burn = 8,000 − 11,400 = −3,400

A negative net burn means revenue is covering costs in this simplified model. The business is not burning cash in this scenario — which changes the runway interpretation from urgency to resilience.

Step 4: Read the sheet as a founder

The model is telling a specific story: recurring revenue is growing (ending MRR of €11,400 vs starting €10,000), costs are under control, and the business is operating at approximately break-even. The key risk signals are: churned MRR includes €220 in failed payments that a dunning sequence could partially recover, and the contraction line (€200) suggests some customers are downgrading, which warrants investigation.

The decision the sheet triggers: prioritise the failed payment recovery sequence this week; investigate contraction source before next month.

Month-by-month forecast table

One month of data is orientation. Three months is a trend.

Month	Start MRR	New MRR	Expansion	Contraction	Churn	End MRR	Fixed Costs	Variable	Runway
Apr	10,000	1,500	600	200	500	11,400	6,000	2,000	—
May	11,400	1,400	650	220	520	12,710	6,000	2,100	—
Jun	12,710	1,600	700	250	600	14,160	6,200	2,300	—

This three-month view already reveals a useful pattern: new MRR is reasonably consistent (1,400–1,600), expansion is growing slightly, but churn is also creeping up (500 → 520 → 600). That churn trajectory needs attention before it becomes structural. The model makes this visible without any additional analysis.

Common SaaS financial model mistakes

Too many inputs. A model with 50 assumptions does not produce better decisions — it produces more places to be wrong. Founders who over-parameterise their models spend more time maintaining inputs than interpreting outputs. Fewer inputs, clearly understood, updated with actual data every month, produce more useful signal than comprehensive models updated twice a year.

Forecasting signups instead of MRR movement. The business reality is MRR, not traffic and signups. A model built around traffic → conversion → activation → paid is useful for product thinking but brittle as a financial model — too many conversion rates to estimate, too many places for optimism to creep in. Anchor on MRR movement and let the acquisition analysis live separately. For the right way to track ARR and MRR for recurring revenue forecasting, that guide covers the definitional foundations.

Treating churn as a rounding error. Almost every first-version SaaS financial model is too optimistic about churn because founders underestimate how natural attrition compounds. A 3% monthly revenue churn rate implies losing ~30% of the recurring revenue base annually to cancellations alone, before accounting for contraction. Model churn honestly and in two parts: voluntary (requires product and retention work to reduce) and failed payment (partially recoverable through dunning). Bessemer’s State of the Cloud report publishes median churn benchmarks by ARR stage that help founders calibrate whether their churn assumptions are realistic.

Mixing recurring and non-recurring revenue carelessly. Setup fees, consulting, and one-off project work that flows through Stripe are real revenue — they should not appear in the MRR line. If they do, the model overstates the health of the recurring engine and distorts every downstream metric including ARR and runway.

Building a model nobody updates. This is the most common failure mode. A model that is too complex to update monthly becomes a historical document the moment it is finished. The test is simple: if updating the model for next month takes more than 20 minutes, it is too big.

When to use a template vs build your own

A template is useful when you need a working starting point quickly, when you want the structural logic done for you, or when the act of having a model is more urgent than having the perfect model. Most downloadable SaaS financial model templates (including free XLS versions) have more tabs than necessary — the right move is to use them as inspiration, delete aggressively, and keep only the sections that reflect how the specific business actually works.

Build from scratch when the business has specific constraints that templates handle poorly — unusual billing cadences, multiple revenue streams that need separate treatment, or cost structures that templates do not reflect. Building from scratch also forces clearer thinking about what the model is actually trying to answer.

The best outcome either way: a single sheet, updated monthly, that tells the founder whether the business is on track or needs a change in priorities.

How this connects to the broader forecasting system

The SaaS financial model is the foundation. Once it exists and is being updated monthly, the next layer is comparison — between what the model predicted and what actually happened. That comparison is where the model becomes more useful over time: each month’s actuals calibrate the assumptions, making future forecasts progressively more accurate.

The next two tools in the sequence:

SaaS Forecast Model: Forecast MRR With 3 Inputs (Not 30 Tabs) — the lightweight, recurring revenue-only version of forecasting
Budget vs Actual: The Weekly Loop That Keeps You Alive — the comparison loop that makes the model improve

JSON model for a minimal SaaS financial sheet

{
  "saas_financial_model": {
    "period": "2026-04",
    "currency": "EUR",
    "revenue": {
      "starting_mrr": 10000,
      "new_mrr": 1500,
      "expansion_mrr": 600,
      "contraction_mrr": 200,
      "churned_mrr_voluntary": 280,
      "churned_mrr_failed_payment": 220,
      "ending_mrr": 11400
    },
    "costs": {
      "fixed": {
        "infrastructure": 2400,
        "saas_tools": 1200,
        "brevo_and_comms": 1200,
        "other": 1200,
        "total_fixed": 6000
      },
      "variable": {
        "stripe_fees": 171,
        "claude_api": 200,
        "contractor": 1629,
        "total_variable": 2000
      },
      "total_costs": 8000
    },
    "cash": {
      "opening_balance": 45000,
      "net_burn": -3400,
      "closing_balance": 48400
    },
    "scenario": "base"
  },
  "scenarios": {
    "base": { "new_mrr": 1500, "churned_mrr": 500, "total_costs": 8000 },
    "upside": { "new_mrr": 1800, "churned_mrr": 400, "total_costs": 8000 },
    "downside": { "new_mrr": 1100, "churned_mrr": 750, "total_costs": 8400 }
  }
}

FAQ

What is a SaaS financial model?

A SaaS financial model is a structured way to forecast how a subscription business will perform over time, specifically covering recurring revenue movement (MRR bridge), costs, cash burn, and runway. Unlike a generic startup financial model, it is built around subscription mechanics — new MRR, expansion, churn, and contraction — rather than general revenue forecasting.

What should be in a SaaS financial model?

The minimal set: starting MRR, new MRR, expansion MRR, churned MRR (split voluntary and failed payment), fixed costs, variable costs, cash on hand, and a scenario toggle for base/upside/downside. Those 8 inputs produce ending MRR, total costs, net burn, and runway — the four outputs that support most founder-level decisions.

What is the difference between a startup financial model and a SaaS financial model?

A startup financial model covers any business type and may include multiple revenue streams, hardware, services, and marketplace dynamics. A SaaS financial model is narrower and more specific: it centres on recurring revenue mechanics, MRR movement, and subscription-based cost structures. This narrower scope allows it to stay genuinely simple.

How detailed should a SaaS financial model be?

Only as detailed as needed to support real decisions. For most early founders, a single sheet with 8 inputs updated monthly is enough. Complexity should be earned by specific business needs — unusual billing structures, multiple revenue streams, or material cost categories that cannot be grouped — not added preemptively.

Do I need a template for a SaaS financial model?

Not necessarily. Templates (including free XLS versions) provide a useful starting point, but the best outcome is a simplified version of any template that matches how the specific business works and is short enough to update in 20 minutes. The danger with templates is keeping too many tabs and assumptions that do not apply.

How do I keep a SaaS financial model useful long-term?

Update it with actual numbers every month before adding the next month’s forecast. This comparison between forecast and actual is what calibrates the model’s assumptions over time. A model that is never compared to reality is just an optimistic spreadsheet.

What is the SaaS financial model template vs SaaS startup financial model template difference?

The terms are largely interchangeable in common usage. A SaaS financial model template typically emphasises recurring revenue mechanics (MRR bridge, churn, expansion) while a startup financial model template may be more generic. For subscription businesses, a SaaS-specific template is more directly useful because the structure is already oriented toward the metrics that matter.

Forecasting from dirty MRR is forecasting wrong. Start with numbers you can trust →