Series A Financial Model Template: The 2026 Founder’s Version

What investors actually open first in your model

Investors open a startup financial model template to the summary tab first, then the cash balance line. They read it rightward until it goes negative, then work backward to test why.

Nobody reads a model front to back. The review sequence is consistent enough to design for.

1. The summary tab. ARR by month, net burn, cash, and headcount on one screen. If there isn’t one, the reader builds it and resents you for it.
2. The cash line, read to the right. They’re hunting for the month cash goes negative, then checking whether it lands before or after the raise you penciled in.
3. The revenue curve. They find the inflection, then look for the mechanism behind it.
4. The headcount plan. They compare the hiring curve against the revenue curve.
5. The assumptions tab. Last, and only to test whether those drivers actually feed the model.

The 2026 bar is a comparison problem, not a recession. The Q2 2026 PitchBook-NVCA Venture Monitor put median Series A pre-money at $64M as of June 30, 2026, on a median round size of $19.4M. That’s the price your model has to justify. PitchBook’s Q3 2026 analysis found AI-related companies absorbed 86% of US venture dollars in the first half of 2026, while Carta logged a Q1 2026 down-round rate of 11.4%, back near 2019 levels.

Capital exists. It’s concentrated.

CRV’s published guidance on Series A metrics puts competitive B2B SaaS at $2M to $5M in ARR with a 2025 median of $2.5M, and treats 100% net revenue retention as the baseline. If you’re calibrating against the round itself, our breakdown of what a Series A round actually requires covers the mechanics around that number.

One rule from Underscore VC’s Series A data room guidance is worth adopting first. Show investors the same forecast you use in board meetings. Two forecasts is one too many.

The shape of a credible Series A model

A credible Series A financial model runs monthly for 24 to 36 months, then annual through year five, across nine linked tabs.

Series A capital is typically sized to 18 to 24 months of runway, so the monthly grid has to cover the runway plus a buffer. Past month 36, monthly precision is false confidence. The FAST Standard says so under its “Appropriate” principle: a model should reflect key business assumptions faithfully “without being cluttered in unnecessary detail,” and should avoid “spurious precision that creates false confidence.”

Here’s the tab architecture the template on this page follows.

Tab	What it holds	Who reads it
Summary	ARR, net burn, cash, headcount, key ratios, live balance check	The partner, in the first 90 seconds
Assumptions	Every driver as a blue input cell, nothing else	The associate, during diligence
Revenue build	Top-down TAM sanity check plus bottoms-up capacity build	Associate and partner
Sales capacity	Reps, ramp, quota, attainment, feeding the revenue build	Anyone who has run a sales team
Expense build	Headcount plan by department plus non-people opex	The operating partner
P&L	GAAP-shaped income statement, monthly	Everyone
Balance sheet	Deferred revenue, AR, accrued liabilities, cash	The finance-literate reader
Cash flow	Indirect method, tying net income to cash	Everyone, first
Scenarios	Base, upside, downside off a single toggle	The investment committee

The summary tab does the persuading. By month and by year it should show ending ARR and growth rate, net new ARR, gross margin, net burn, burn multiple, cash balance, months of runway, headcount by department, and Rule of 40 score.

Christoph Janz’s SaaS Financial Plan 2.0 at Point Nine set the pattern most founder-facing SaaS plans still copy, with a summary tab taking only two inputs and a separate sales hiring plan derived from revenue goals and quotas. Two things a 2016 model never had to carry: variable inference cost, and the cash tax treatment under Section 174A. Both belong in a 2026 build. If you want the concept before the file, our explainer on what financial modeling is and where models break covers the fundamentals of financial modeling for startups without the Series A framing.

The assumptions tab: where credibility lives or dies

Every number a reader can argue with lives as one labeled input cell on the assumptions tab. Nothing hardcoded downstream.

Conversion rates, churn, quota, ramp months, salary bands, hosting cost per customer, payment terms, tax rates. All of it sits on one tab and gets referenced everywhere else. An assumption a reader can’t isolate is an assumption they assume is wrong.

Hardcoded numbers are a tell.

Formatting is part of this, not decoration. The color convention documented by modeling references like Macabacus and Training The Street tells a reviewer where to click:

• Blue font for hardcoded inputs and assumptions
• Black font for formulas calculated on the same sheet
• Green font for links to other sheets in the same workbook

A model that follows it takes about 30 seconds to audit. One that doesn’t takes an hour, and nobody spends the hour.

Every figure in the template’s benchmark tab carries a named source and a year, because “industry standard” isn’t an answer in diligence. The growth medians below come from SaaS Capital’s private B2B SaaS growth benchmarks, drawn from more than 1,000 companies.

Assumption	2026 anchor	Source
ARR growth	22% median for private B2B SaaS in 2025, 25% equity-backed, 20% bootstrapped	SaaS Capital, 2026 survey
Net revenue retention	102% median at $25K to $50K ACV, 111% top quartile	SaaS Capital, 2025 retention data
Gross margin	75% to 85% traditional SaaS, about 52% projected 2026 for AI-native products	ICONIQ, 2026 State of AI
CAC payback	16 months median, 6 months top quartile, 24+ bottom quartile	Aleph x Benchmarkit, 2026, FY2025 actuals
Rule of 40	25% median, 43% top quartile	Aleph x Benchmarkit, 2026
Spend at $3M to $5M ARR	Sales 12% of ARR, marketing 8%, R&D 24%, G&A 15%	SaaS Capital, 2026 Spending Benchmarks
AE quota and attainment	$960K median quota, $200K median OTE, 48% attainment, 6.2 month ramp	The Bridge Group, 2026, 158 companies
Fully loaded headcount	1.25x to 1.4x base salary, benefits near 30% of total comp	Bureau of Labor Statistics ECEC

The UK Business Angels Association’s June 2026 red-flag list puts two assumption failures in its top five: hockey-stick growth with no explanation, and assumptions resting on optimism rather than real performance data or comparable companies.

One 2026-specific trap. If any part of the product calls a model, inference is variable COGS that scales with usage, not fixed infrastructure. ICONIQ’s 2026 survey of roughly 300 software executives found scaling-stage AI B2B companies averaging around 52% gross margin projected for 2026, up from 45% in 2025. Assuming 80% while shipping an LLM feature is a claim the reader will check against your own cloud invoices. Model cost per unit of usage as an input, and let gross margin be an output. Our rundown of the SaaS metrics investors check first covers which of these get audited hardest.

The revenue build: ARR, MRR, and bookings math

Revenue is an output, not an input. Build it bottoms-up from ramped reps times quota times attainment, and use top-down TAM only as a bounding check.

Top-down answers one question. Is the year-five number physically possible? Target accounts in the segment multiplied by a realistic ACV gives serviceable obtainable market, and if year-five ARR is 40% of your entire SOM, the reader stops. That’s its whole job. Bottoms-up is the forecast.

Get the definitions right first, because Andreessen Horowitz’s 16 startup metrics supplies the ones investors actually use. Bookings are contracted obligations. Revenue is recognized as service is delivered or ratably across the subscription term. ARR covers recurring components only and excludes one-time fees and professional services. The named mistake is taking one month’s all-in bookings, multiplying by 12, and calling it ARR.

The MRR movement engine

Ending MRR = Beginning MRR
           + New MRR
           + Expansion MRR
           + Reactivation MRR
           - Contraction MRR
           - Churned MRR

ARR = Ending MRR x 12

Net New ARR = New ARR + Expansion ARR - Churned ARR - Contraction ARR

Show both churn views. Gross MRR churn = MRR lost in month / MRR at beginning of month, and Net MRR churn = (MRR lost – MRR from upsells) / MRR at beginning of month. The a16z warning is that gross churn estimates the actual loss to the business, while net churn understates it by blending upsells with absolute churn. A model showing only net churn is hiding something.

Two more get asked for by name. NRR = (Starting cohort MRR + Expansion – Contraction – Churn) / Starting cohort MRR, reconstructed from cohort data rather than asserted. And SaaS quick ratio = (New MRR + Expansion MRR) / (Churned MRR + Contraction MRR), where 4.0 is the working benchmark popularized by Mamoon Hamid, meaning $4 of recurring revenue gained for every $1 lost. For the simpler single-motion version, our free SaaS financial model template walks through one worked example end to end.

Sales capacity, the part most founder models skip

Sales capacity = Ramped reps x Average annual quota x Expected attainment

Ramped reps is what trips people up. A rep hired in month 7 with a six-month ramp contributes almost nothing to the current year, so count reps at full productivity plus reps still in ramp, weighted by linear progress. Working backward, AEs required = Bookings target / (Quota x Attainment x Ramp factor x Tenure factor), then add attrition to get gross hires needed.

The Bridge Group’s 2026 State of Sales research across 158 B2B companies reports a $960K median annual quota, $200K median OTE, a 4.6x quota-to-OTE ratio, 48% quota attainment (down from 51% in 2024), and a 6.2 month ramp, the longest in the study’s history. CJ Gustafson’s sales capacity work at Mostly Metrics adds the rules of thumb: quota-to-OTE starting near 5x and stretching toward 7x as brand strengthens, ramp of 2 to 3 months in SMB and 6 to 9 months in enterprise, quota over-assignment of 20% to 30% against the board plan, and 20% to 30% annual sales attrition.

So a 40-rep team with a 5.7-month average ramp and 30% attrition isn’t 40 quotas of capacity. It’s closer to 22 to 25. Assume 90% attainment and a three-month ramp and you’re 42 points and three months off the published benchmark, which the reader who has run a sales team sees in one glance.

From bookings to cash

Bookings (contract value signed)
  -> Billings (invoices issued in the period)
    -> Revenue (recognized ratably over the service period, ASC 606)
      -> Cash collected (billings less change in AR)

An annual contract signed January 1 with payment upfront is $120K of bookings, $120K of billings, $120K of cash, and $10K of revenue in January. The other $110K sits in deferred revenue on the balance sheet. Skip that distinction and the model is wrong in both directions at different times. Annual prepay is the most common reason a founder-built forecast and the bank balance disagree, which is why the template carries billings and revenue as separate lines. Our guide to revenue forecasting goes deeper on the recognition side of the same chain.

The expense build: headcount, hosting, and the rest

Expenses come off a headcount roster, loaded at 1.25x to 1.4x base salary once employer payroll taxes and benefits land.

Build a roster, not a department total. One row per planned hire, with role, department, start month, base salary, variable compensation, and a loading factor.

Fully loaded monthly cost = (Base / 12)
                          + (Variable / 12)
                          + Employer payroll taxes
                          + Benefits and insurance
                          + Software and equipment per head

Anchor the loading rather than guessing it. Employer FICA runs 7.65% of wages, 6.2% for Social Security up to the 2026 taxable wage base of $184,500 and 1.45% for Medicare with no cap. Bureau of Labor Statistics data on employer costs for employee compensation puts private-sector benefits near 30% of total compensation. Timing is where founder models leak cash. Payroll taxes remit on a schedule, commissions often pay quarterly, bonuses annually. Model each separately from base payroll.

Sanity check against peers. SaaS Capital’s 2026 spending benchmarks put median spend for companies at $3M to $5M ARR at 12% of ARR on sales, 8% on marketing, 24% on R&D, 15% on G&A, and 10% on support and success. Funding type matters when you pick a comp. Equity-backed companies spend roughly 70% more on sales and 100% more on marketing than bootstrapped peers, and 83% of bootstrapped companies run within two points of breakeven or better, versus 52% of equity-backed ones.

COGS as drivers, not a lump. Hosting cost = (Cost per active customer x Active customers) + Fixed infrastructure. Inference cost = Calls per active user x Users x Tokens per call x Blended cost per token. Gross margin then falls out as (Revenue – COGS) / Revenue. KeyBanc Capital Markets’ 16th annual private company SaaS survey, released in November 2025, found more than half of respondents planning to raise AI spend by over 21% and 67% already monetizing AI. If your R&D line doesn’t carry rising AI tooling and compute cost, you’re modeling a 2022 company.

The tax line most templates leave blank. Section 174A, created by the One Big Beautiful Bill Act, permanently restores full expensing of domestic research costs for tax years beginning after December 31, 2024, reversing the five-year amortization that inflated taxable income for R&D-heavy startups. Foreign research stays on 15-year amortization. Separately, a qualified small business can apply the federal research credit against employer payroll taxes rather than income tax, which is real cash in a pre-profit company. Most downloadable templates carry neither. Indinero’s tax team models both into the cash forecast, because the same engagement that runs your books also files the credit, and both items move runway by months rather than weeks. For the wider spend picture, our guide to budgeting for startups pairs with this build.

The three-statement flow: P&L, balance sheet, cash flow

A model without a balance sheet can’t forecast cash, because the gap between profit and cash is deferred revenue, AR, and accruals.

Deferred revenue is frequently the largest liability on a SaaS balance sheet. Annual upfront billing means cash lands before revenue is recognized, and that increase flows through operating cash flow as a positive working capital adjustment. It’s how a growing SaaS company posts positive operating cash flow alongside a net loss. The inverse is the dangerous case. When annual prepay growth slows, the tailwind disappears and operating cash falls faster than the P&L suggests it should. A model without a balance sheet can’t show that, which means it can’t show your actual worst case.

No cash flow statement, no credibility.

The P&L, monthly: revenue recognized under ASC 606, less COGS (hosting, inference, support, payment processing, capitalized software amortization) to gross profit, less S&M, R&D, and G&A to EBITDA, then D&A to EBIT, then interest and taxes to net income. Keep the tax line populated even pre-profit, because state minimums and franchise tax are still owed.

The balance sheet needs the SaaS-specific rows: AR driven off a DSO assumption, prepaid expenses, capitalized software under ASC 350-40, capitalized commissions under ASC 340-40, AP driven off DPO, accrued payroll and commissions, and deferred revenue split current and long-term.

The cash flow statement, indirect method: net income, plus D&A and stock-based compensation under ASC 718, plus or minus changes in AR, prepaids, AP, accrued liabilities, and deferred revenue, to cash from operations. Then capex and capitalized software development to investing, and equity and debt movements to financing.

Two roll-forwards do most of the wiring. Ending deferred revenue = Beginning deferred revenue + Billings – Revenue recognized. Ending AR = Beginning AR + Billings – Cash collected, or (DSO / 30) x Monthly billings if you drive it from days sales outstanding. Then two checks, both visible on the summary tab:

• Total Assets – (Total Liabilities + Total Equity) = 0
• Ending cash on the cash flow statement = Cash on the balance sheet

A reviewer who sees a live balance check knows within five seconds that the statements are genuinely linked and not three separate guesses.

Burn falls out of the same structure. Gross burn = Cash operating expenses + capex. Net burn = Gross burn – Cash collected. Runway (months) = Cash balance / trailing three-month average net burn. And Burn multiple = Net burn in period / Net new ARR in period, which David Sacks introduced in 2020 with the expectation that it declines as a company matures, roughly 3x at seed and 2x at Series A, and that 5x warrants immediate cost cuts.

All of this assumes the actuals underneath are accrual and GAAP-clean. Indinero’s CPA team builds GAAP-compliant books from day one, which matters here for a blunt reason. A model sitting on cash-basis books has no deferred revenue balance to roll forward, so the cash forecast is guesswork dressed up as math. Our guide to SaaS cash flow management covers the operating side of the same problem.

Scenarios investors expect to see

Show three scenarios off one toggle cell, with the downside running 30% to 50% below base for several quarters.

Not three files. One selector cell, with every scenario-sensitive assumption reading from it through = CHOOSE($Scenario, Base_value, Upside_value, Downside_value) or an INDEX against a scenario table row. The summary tab then shows all three cases side by side without anyone touching the toggle.

Case	The question it answers	What to surface
Base	Is the plan you operate against the same plan you’re showing us?	Ending ARR at year one and two, net burn, runway, burn multiple, Rule of 40, year-end headcount
Upside	Does more capital produce more outcome, or just faster spend?	The same outputs, plus incremental ARR per incremental dollar of S&M
Downside	Which assumption breaks you first, and when?	The same outputs at 50% of plan, plus cash-out month and cash six months past the projected raise

Upside that only spends faster says the business doesn’t scale. The downside case is the one that gets the attention in 2026, so frame it as the 50%-of-plan test and answer four questions on one screen.

1. What happens at 50% of plan? Not just ARR. What’s net burn, given that most of the cost base is people you already hired?
2. When does cash run out in that case? A month number, stated plainly.
3. What’s the cash position six months past the projected next raise? If cash hits zero the month you plan to close a Series B, you’ve modeled zero margin for slippage. Raises slip.
4. What’s the cut you’d make, and when? A downside case carrying the same headcount plan as the base case isn’t a downside case. Model the decision, with a trigger month.

Add the Rule of 40 to every case. Rule of 40 = revenue growth rate % + profit margin %, usually calculated on EBITDA margin. The 2026 Aleph x Benchmarkit SaaS and AI performance benchmarks, drawn from full-year 2025 actuals across 342 companies, put the median at 25% and the top quartile at 43%. Worth noting where that 10-point gain came from: cost reduction, with R&D down 8 points, not faster growth.

Two 2026-specific cases are worth adding. A raise-timing case where the Series B closes six months later than planned, which given how concentrated venture dollars have become is observation rather than pessimism. And a gross margin case where inference cost per user runs 50% above base. Scenario work is the piece founders most often want a second set of eyes on, and it sits at the center of how a CFO supports a fundraise.

Common modeling mistakes that flag amateurs

Investors aren’t grading forecast accuracy, they’re grading honesty, and eleven recurring errors mark a model as unreviewed.

1. Revenue untethered from sales capacity. The UK Business Angels Association’s red-flag list puts growth projections without underlying drivers at the top. Tie bookings to ramped reps times quota times attainment, and show the chain.
2. Expense growth that trails revenue growth for no stated reason. If S&M drops from 60% of revenue to 25% over three years, name the mechanism (channel mix, expansion revenue, self-serve motion) or fix the assumption.
3. No balance sheet. Deferred revenue, AR, and accrued liabilities are the entire gap between net income and cash.
4. No cash flow statement. A cash tab that subtracts expenses from revenue isn’t one. Without the working capital bridge, the cash-out month is a guess.
5. Hardcoded numbers inside formulas. A reviewer who finds one churn rate typed into a formula assumes there are more.
6. Annualizing one good month. ARR excludes one-time fees, setup, hardware, and professional services.
7. Showing only net churn. It understates losses by blending upsells with absolute churn. Show gross and net.
8. Only a best case. No scenario planning sits on the same UKBAA red-flag list.
9. A model that disagrees with the deck or the accounting system. ARR in the deck, ARR in the model, and ARR in the books must be the same number computed the same way.
10. Gross margin assumed rather than built. If the product calls a model, margin is a function of usage and moves with it.
11. No tax line. Pre-profit companies still owe state minimums and franchise tax, still run payroll taxes, and may be sitting on an R&D credit worth real cash.

Number nine is the expensive one. Definition drift between the model and the general ledger is the fastest way to turn a four-week diligence into a ten-week one, which is why model-to-books reconciliation belongs in the monthly close rather than in the fundraise. Our walkthrough of financial due diligence covers what that audit looks like from the other side of the table.

Download the template and what’s inside it

The template is a working model, not a blank grid. Change the assumptions tab and every statement, chart, and scenario recalculates.

Here’s what ships in the file:

• Summary dashboard. ARR, net new ARR, gross margin, net burn, burn multiple, runway, cash-out month, headcount, and Rule of 40 on one screen, with a live balance check.
• Assumptions tab. Every driver as a blue input cell, grouped by revenue, sales capacity, retention, COGS, headcount, working capital, and financing.
• Revenue build. The full MRR movement engine, the bookings to billings to cash chain with annual prepay handled correctly, and a cohort view so NRR can be reconstructed rather than asserted.
• Sales capacity and hiring plan. A rep-by-rep roster with start month, segment, quota, ramp curve, and attainment, feeding the revenue build directly.
• Expense build. Headcount roster by department, fully loaded at the 2026 wage base, with correct timing for commissions and bonuses.
• Three linked statements. Monthly P&L, balance sheet with deferred revenue split current and long-term, and an indirect-method cash flow statement, both consistency checks live.
• Scenario sheet. One toggle, three cases side by side, with the cash-out month and the six-months-past-the-raise position surfaced for each.
• Benchmark reference tab. The peer medians on this page, sourced and dated, including the ICONIQ 2026 State of AI gross margin bands for AI-native products.

Google Sheets and Excel. No macros, no add-ins, no locked cells. The file has to survive being opened by an associate at a firm that doesn’t use your tooling.

A template gives you structure. It doesn’t give you clean actuals to sit on top of, and a revenue build resting on cash-basis books has nothing to anchor to. That’s the part indinero handles: bookkeeping, accounting, tax, and fractional CFO bundled under one monthly engagement, so the team closing your month is the team feeding the model that forecasts it.

Most template downloads are a one-time handoff. We work as a year-round finance partner instead, refreshing assumptions against actuals every month and carrying the R&D credit cash benefit into the runway forecast rather than discovering it at filing.

The engagement scales the way the finance function does. Bookkeeping first, then GAAP-clean close, then full fractional CFO advisory as board reporting gets heavier, with the same team throughout. Continuous operations since 2009 and 500+ regular customers means the people pressure-testing your model this year are still there at Series B.

Pair the file with the SaaS financial decision tree if you’re still deciding which levers to model first. And if you’d rather build the thing alongside someone who has read a few hundred of these, that’s the kind of fundraising support startup founders tend to want before the first partner meeting. It’s what our fractional CFO team does every week.

Frequently asked questions

Common questions founders ask about building a Series A financial model and pressure-testing it before a raise.

What should a Series A financial model template include?

How detailed should the assumptions tab be in a startup financial model?

What scenarios do Series A investors expect to see modeled?

How do I model revenue for a SaaS company at the Series A stage?

Should the financial model tie to the three core financial statements?

How far out should a Series A financial model project?

What are the most common mistakes founders make in fundraising financial models?