The model your investor asked for — built to survive the room.

What's in this report

Four movements: we mirror your situation, analyze it tab by tab, show you the transformation, then hand you the action plan and the words to use.

• Attention — your situation, mirrored back
• A · The email that sent you here3
• A · Situation snapshot — your five numbers4
• A · The five ways an untied model fails in the room5
• A · How to read this report6
• Interest — the analysis, tab by tab
• I · Assumptions cockpit — the drivers7
• I · Revenue build & customer roll-forward10
• I · The ARR bridge investors ask for14
• I · Income statement17
• I · Balance sheet — the part that must tie20
• I · Cash flow & runway24
• I · SaaS metrics — computed from the build28
• I · Valuation — method, not opinion34
• I · Sensitivity & scenarios38
• Desire — what good looks like
• D · Before / after: your next investor call41
• D · The transformation, metric by metric44
• D · Milestone roadmap to a funded round50
• D · Defend every assumption they'll poke54
• Action — do this before the meeting
• Act · Prioritized checklist: 72h / 30d / 90d61
• Act · Scripts for the diligence questions64
• Act · Advisor / CPA handoff sheet67
• Act · Next step & guarantee69

Educational analysis and a spreadsheet template — not investment, legal, financial, or tax advice, and not a valuation opinion or an offer of securities. All figures are illustrative outputs of a deterministic model driven by the assumptions shown; change an assumption and every figure changes. Consult qualified professionals before making any financing or investment decision.

"Can you send us the model?"

Six words from a partner at the fund you most want on your cap table.

You have a deck. You have a story that got you the meeting. You have a Google Sheet with a revenue tab and a cell that says "runway." What you do not have — yet — is a model: an integrated set of statements where the income statement flows into the balance sheet flows into the cash flow, where the metrics are computed from the same build, and where the balance check reads 0.00 so an analyst can't unravel it in ninety seconds.

That gap is the most common, most fixable reason a promising round stalls. It is not that your business is weak — the numbers below describe a company any seed investor would take a second meeting with. It is that the artifact you're being graded on doesn't exist in defensible form. This report shows you exactly what that artifact looks like, on a fictional company (Cadence Robotics, Inc. (FICTIONAL EXAMPLE)) whose profile was chosen to rhyme with yours — so that when you drop your own numbers into the cockpit, you already know what every page will say.

Read the next page first. It mirrors your situation back to you in five numbers — the same five an investor forms an opinion on before you finish your intro.

Your situation, in five numbers

This is what a partner sees before you say a word. For Cadence Robotics, Inc. (FICTIONAL EXAMPLE), the engine computes:

Those are strong numbers — and that's the point. The business isn't the problem. The problem is that until this model exists, you can't prove any of them without contradiction. An NRR you claim in the deck but can't reconcile to the P&L is worth less than an NRR of 118% that falls out of a model that ties.

Every number on this page is a computed output of the model — see the tab it comes from later in the report. Change an assumption in the cockpit and each of these recomputes.

The whole model on one page

If you read nothing else before the meeting, read this. Every figure is a computed engine output.

All figures are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE). Not a forecast or a recommendation.

The five ways a model fails in the room

Each of these has killed a raise. Each is fixed by structure, not by better numbers.

How to read this report

Four movements, in the order your anxiety shows up.

The four movements

• Interest (pages 7–40) — the analysis, one tab at a time. Every section opens with why this matters to you now, then shows the engine-driven tables, and carries a What you told us → what this section computed box so you can see your own inputs driving your own report.
• Desire (pages 41–60) — the transformation. What "good" looks like, the before/after of your next investor call, the milestone roadmap, and a defense for every assumption.
• Action (pages 61–70) — the plan. A prioritized checklist (next 72 hours / 30 days / 90 days), word-for-word scripts, a handoff sheet for your CPA, and the natural next step.

Everything past this page is computed by the ModelKit engine from the Cadence Robotics, Inc. (FICTIONAL EXAMPLE) assumptions. When you buy the workbook, you replace those assumptions with yours and every page regenerates.

The assumptions cockpit

One tab of drivers. Every other tab is a formula off these. This is the only place you type numbers.

These are the drivers for Cadence Robotics, Inc. (FICTIONAL EXAMPLE). When you buy the kit, this is the one tab you edit; the nine tabs recompute from it.

Revenue drivers

Cost drivers

Working-capital & financing drivers

Reading the financing plan

The model assumes primary capital comes in during 2027 and 2029 — $14.0M then $30.0M. Those raises are inputs, not conclusions; they let the cash line stay positive so you can see what runway your plan actually requires. Your real rounds are set by your investors, not by this tab.

Why driver-based beats typed-in

A typed forecast ("we'll grow 2.4×") is a single number an investor can only accept or reject. A driver-based forecast decomposes that same growth into pieces you can each defend:

• New logos tie to sales capacity — reps × ramped quota — not to a wish.
• ARPA ties to your price list and observed expansion, not to a round number.
• Churn ties to your cohort data; you can point to the retention curve.
• CAC ties to the S&M line in the P&L — it can't secretly disagree.

The rest of this report walks each of these from driver to output, so when a partner asks "where does this number come from?", the answer is always another number they can see, not "trust me."

Assumptions shown are the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE) inputs. Your workbook ships with these as editable defaults; you overwrite them.

How the nine tabs wire together

The one diagram that turns "a spreadsheet" into "a model." Follow a number from driver to output.

• 1 · CockpitEight-ish drivers you type. Nothing downstream is typed.
• 2 · Revenue build → 3 · ARR bridgeDrivers produce the customer roll-forward and the ARR roll-forward. Revenue flows out of here.
• 4 · Income statementRevenue − COGS − opex (S&M = CAC × logos, straight from Tab 2) = net income.
• 5 · Balance sheet ← 6 · Cash flowNet income drives retained earnings; cash flow computes the cash line; working-capital drivers set AR/AP/deferred. The identity A = L + E holds by construction.
• 7 · MetricsEvery metric is a formula on the tabs above — so they can't contradict the statements.
• 8 · ValuationReads the FCF path and exit ARR; illustrates method, asserts nothing.

Revenue build & customer roll-forward

Where your top line actually comes from: logos in, logos out, price per logo.

Recognized revenue is average customers × ARPA × 12, with new logos added mid-year (they earn half a year of revenue in the year they land). That mid-year convention is why Year-1 revenue of $992K is lower than 60 logos × ARPA would suggest — the model is conservative on the timing, which is exactly the direction you want to be conservative in front of an investor.

The customer roll-forward, in words

In 2026 the company starts with 14 logos, adds 46, and loses 5.1 to churn, ending at 55. New-logo adds grow 55%/year — so 46 becomes 71 the next year — which is what turns $992K of Year-1 revenue into $15.8M by 2030.

The churn you can defend

Logo churn of 1.2%/month compounds to roughly 13% of the base per year on an annualized basis. That is an input — you set it from your cohort data and can show the retention curve behind it. It is not a residual the model backed into.

ARPA and expansion — the quiet compounder

ARPA starts at $2,400/month and grows 6%/year from price and expansion, reaching $3,030/month by 2030. Small ARPA growth is deceptively powerful: on a growing base it contributes materially to the $15.8M exit-year revenue without adding a single logo.

What an investor will test here

All revenue figures are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE). Mid-year add convention and monthly-compounded churn are stated so the method is auditable.

The ARR bridge investors ask for

Beginning + new + expansion − churn = ending. The single most-requested SaaS exhibit.

The bridge closes in every year — beginning plus new plus expansion minus churn lands exactly on ending — which is one of the 217 automated checks. A bridge that doesn't close is the fastest way to signal a hand-built spreadsheet.

Reading the mix

By 2030, expansion ARR of $4.8M more than offsets churned ARR of $1.8M on the base — which is the mechanical reason net revenue retention comes out above 100%. New ARR of $9.6M then stacks on top. That ordering — retain the base, expand it, then add new — is the story investors want to hear, and here it's a table, not a claim.

Gross vs. net, kept honest

Churned ARR is computed from a 1.0%/month gross revenue-churn input compounded over the year, and expansion from a 2.2%/month input on the retained base. Because both are drivers, your gross retention (GRR) and net retention (NRR) can never contradict this bridge — they're computed from the same two numbers.

The questions this exhibit pre-answers

ARR bridge figures are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE). The bridge-closure check is one of the 217 automated assertions.

Income statement

Revenue → gross profit → operating lines → the bottom line. Every line is arithmetic off the tabs before it.

Gross margin holds at 74% — appropriate for a hardware-touched vertical SaaS. Sales & marketing is not a percentage guess; it is $11,000 of fully-loaded CAC × the new logos from Tab 2, so the P&L and the CAC metric can never disagree. R&D and G&A carry floors ($1.1M and $520K) because a team exists before the revenue does.

The path to profitability, stated plainly

The company runs a planned loss while it invests: net income of $-1.4M in 2026 narrows to $-32K by 2030 as revenue scales past the R&D and G&A floors. EBITDA crosses from $-1.4M to $236K over the horizon. This is the shape an investor wants pre-profit: losses that shrink as a share of revenue, driven by operating leverage, not by cutting growth.

Operating leverage, visible

Because R&D and G&A have dollar floors while revenue compounds, their percentage of revenue falls every year. That falling percentage is the entire pre-profit SaaS thesis — you're buying growth now so the fixed base gets cheaper per dollar of revenue later — and here it's legible on three rows instead of asserted in a sentence.

What an investor will probe

Income-statement figures are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE). Arithmetic (GP = rev − COGS; EBIT = EBITDA − dep; NI = EBIT − tax) is checked in every year.

Balance sheet — the part that must tie

The one test every investor runs first. Assets = liabilities + equity, or the model is broken.

How the identity is held (not plugged)

The most common way a founder's balance sheet "ties" is a hidden cash plug — a formula that forces cash to whatever makes the sheet balance, quietly hiding an error. This model does the opposite:

• Cash comes from the cash-flow statement (Tab 6), not from a plug.
• Retained earnings roll forward as opening deficit + cumulative net income — checked every year.
• Paid-in capital is starting capital + cumulative raises — checked every year.
• AR, AP, deferred revenue come from DSO/DPO/deferred drivers on the tabs above.

Because every line has an independent source, the fact that assets still equal liabilities plus equity is evidence the whole model is consistent — not an assumption baked in. That is precisely the signal an investor is looking for.

Working capital is a story, not a footnote

Deferred revenue rises to $8.6M by 2030 because 30% of ARR is annual prepay — customers paying you a year ahead. That's a genuine, defensible source of cash and a sign of pricing power; investors like to see it. Meanwhile DSO of 52 days means $2.2M sits in receivables at exit — cash you've earned but not collected, which the cash-flow statement accounts for honestly.

Balance-sheet figures are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE). Total assets = cash + AR + PP&E and total liabilities = AP + deferred are checked as separate assertions.

Cash flow & runway

The statement that answers the only question that can end your company: when do you run out?

This is the indirect method: start at net income, add back non-cash depreciation, adjust for working-capital movements (AR, AP, deferred revenue), then layer investing (capex) and financing (equity). Ending cash here equals the cash line on the balance sheet in every year — a continuity check, and another of the 217 assertions.

Runway, computed — not guessed

At the 2030 exit, cash-from-operations plus investing is positive — the business is cash-flow positive on operations, so runway is not the binding constraint. Ending cash reaches $51.8M after the planned raises of $14.0M and $30.0M. The point isn't the specific number — it's that runway is derived from the cash line, so it can't quietly disagree with the rest of the model.

Working capital is a real cash lever

Deferred revenue swings $3.8M of cash in 2030 — annual prepay financing your growth. AR growth uses $942K of cash as sales scale. These aren't rounding: they're the difference between a model that looks profitable and one that's actually solvent, and investors know to look for them.

The runway questions you'll be asked

Cash-flow figures are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE). Ending cash = balance-sheet cash and beginning + net change = ending are checked every year.

SaaS metrics — computed from the build

The scorecard an investor already has in their head. Yours had better match — and agree with your statements.

Ending ARR grows from $1.8M to $28.7M. Every other row is derived from the statements you've already seen — which is why they reconcile by construction.

Burn multiple is shown to two decimals. A year that still shows a positive net burn but rounds to <0.01 is burning very little per dollar of new ARR — not literally zero — which is why the burn-multiple and net-burn rows stay consistent.

Retention: NRR & GRR

Net revenue retention (NRR) = (beginning ARR + expansion − churn) ÷ beginning ARR = 118.5%. Above 100% means your existing customers are worth more each year even before you add a single logo — the single most valuable property a SaaS business can have.

Gross revenue retention (GRR) = (beginning ARR − churn) ÷ beginning ARR = 88.6%. GRR strips out expansion, so it's the honest floor on how sticky you are. GRR is always ≤ NRR (checked as an assertion); the gap between them is the expansion you're earning.

Efficiency: CAC, payback, LTV:CAC

Blended CAC = S&M ÷ net-new logos = $11,000 at exit. Because S&M is the same line on the income statement, this can't secretly disagree with the P&L.

CAC payback = CAC ÷ (ARPA × gross margin) = 4.9 months at exit — the number of months of gross profit per customer needed to earn the acquisition cost back. Under ~12 months is widely considered efficient; 4.9 is strong.

LTV : CAC = 20.4× at exit, where LTV = (ARPA × gross margin) ÷ monthly gross churn. The classic benchmark is ≥ 3×; this model sits well above it because churn is low and margin is healthy.

Capital efficiency: magic number, burn multiple, Rule of 40

Magic number = net-new ARR ÷ prior-year S&M = 6.71 at exit. Above ~0.75 says every S&M dollar is buying more than its share of new ARR — you should be spending more, not less.

Burn multiple = net burn ÷ net-new ARR = 0.00 at exit (the business isn't burning at exit, so the multiple is 0). It's the Bessemer/Craft-style "how much cash to add a dollar of ARR" — the cleanest single read on efficiency.

Rule of 40 = ARR growth % + FCF margin % = 98 at exit. Above 40 is the bar for a healthy growth-stage SaaS; early on, growth carries it (139 + -0 = 138 in 2027).

The metrics scorecard, read as an investor would

All metrics are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE); each definition is a standard public SaaS definition and is spot-checked against its formula in the test suite.

Where these metrics sit against public benchmarks

The bars investors carry in their heads — from public frameworks, not invented data.

*Benchmarks are widely-cited public rules of thumb (SaaS-metrics literature, YC/NVCA and Bessemer/Craft public frameworks), not proprietary datasets and not a claim about any specific cohort. "This model" figures are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE). Benchmarks vary by segment, stage and market; treat them as orientation, not targets.

One customer, all the way down

Zoom from the P&L to a single logo — the level a sharp investor drills to.

A single Cadence Robotics, Inc. (FICTIONAL EXAMPLE) customer generates $2,242/month of gross profit and costs $11,000 to acquire — earned back in 4.9 months, then profitable for the rest of a ~100-month expected life. That's the engine of the whole model; everything above is this unit, multiplied and rolled forward.

Valuation — method, not opinion

How to show you understand valuation without pretending your DCF sets your price.

Read the band, not the endpoints. The 4–10× forward-ARR band spans the whole cycle: 10× reflects a frothy, peak-multiple market, 4× a compressed one. Since the 2022–2023 reset, most growth-stage SaaS has priced toward the low end of that band, so a realistic mark for a company like this sits closer to $115M — with quality of growth (NRR, CAC payback, net burn) deciding where inside the band you land. Treat the top of the range as a ceiling from a hotter market, not a target, and confirm the live multiple against current comps at the time you raise. This is a method demonstration, not a valuation opinion.

DCF sensitivity — why the $3.0M isn't the number

A single DCF figure invites the question "what if your discount rate is wrong?" Here's the honest answer on this page instead of buried in a guide: the whole grid, computed by the same engine. Swing the WACC from 25% to 45% and terminal growth from 2% to 4%, and the DCF still lands between $1.6M and $6.3M — an order of magnitude below the comps band. That's the lesson, not a defect: an early-stage DCF is terminal-value dominated and can't reach where rounds actually price.

Every cell is a live output of the same engine, re-running the DCF at that WACC / terminal-growth pair on the Cadence Robotics, Inc. (FICTIONAL EXAMPLE) assumptions. Even the most generous corner ($6.3M) sits far below the $115.0M–$287.4M comps band — which is exactly why investors anchor early rounds on comps, not the DCF.

DCF at a 35% early-stage WACC and 3% terminal growth is shown to demonstrate the METHOD, not to assert a value. Early-stage DCF outputs are dominated by the terminal value and are extremely sensitive to inputs — investors anchor pre-Series-B rounds on comparable multiples and round norms, not a founder's DCF. Both are illustrative.

Why the two methods disagree — and why that's the lesson

The DCF lands at $3.0M enterprise value; the comparable-multiples band (4×–10× exit ARR of $28.7M) lands at $115.0M–$287.4M. That's a huge gap — and it's the whole point.

An early-stage DCF is dominated by its terminal value and is wildly sensitive to the discount rate and growth assumptions; at a 35% WACC, the near-term losses barely register and the answer is essentially "whatever you assume happens after year five." That's exactly why investors don't price pre-Series-B rounds on a founder's DCF — they anchor on comps and round norms. Showing both, and knowing which one the room actually uses, is what makes you sound like an operator instead of a spreadsheet.

How to talk about valuation in the room

Valuation figures are illustrative engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE), demonstrating method only. Not a valuation opinion, not investment advice.

Sensitivity & scenarios

"What if you miss?" is a question, not an attack. Walk in with the answer already modeled.

The stress case in ModelKit's test suite proves this works: with margin cut to 55%, logo churn quadrupled, and expansion near zero, the model still ties in every year — the structure holds even when the story turns ugly. That's the confidence you want walking into a downside question.

The three scenarios to bring to the meeting

How to present sensitivity without undercutting yourself

The goal is to look prepared, not scared. Lead with the base case as your plan. Introduce the conservative case as "and here's what we'd still deliver if we're wrong on growth" — proving the round is sound even off-plan. Keep the downside in your back pocket for the "what could kill you?" question, where a specific, modeled answer beats a vague reassurance every time.

Scenario mechanics are properties of the engine, demonstrated on the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE). The stress case (55% margin, 4%/mo churn) is one of the three assumption sets in the 217-assertion test suite.

Your next investor call — before and after

Same company, same numbers. The only thing that changed is the artifact in your hands.

What "good" actually looks like

Good is not a fancier model. Good is a model an analyst can't break and you can't be caught out on. Concretely, for a company like Cadence Robotics, Inc. (FICTIONAL EXAMPLE), "good" is:

• Ties out. A − L − E = 0.00 in every year, from independent sources, no plug.
• Self-consistent. Metrics computed from the statements, so NRR 118.5%, CAC payback 4.9mo and the rest never contradict the P&L.
• Driver-based. Every output traces to an assumption you can defend with data.
• Scenario-ready. One driver change re-runs the whole model — base, conservative, downside.
• Owned by you. You can answer any question because you understand every wire.

The 90 seconds that decide the meeting

Most first analytical calls turn on a 90-second window: the associate opens your model, checks that it balances, spot-checks one metric against the statements, and forms a view of whether you can be trusted with capital. Everything before was story; this is the first test of substance.

All figures reference the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE) model built in this report.

The transformation, question by question

For each thing an investor tests, here's what changes when you hold this model.

…and the rest of the scorecard

Why this compounds beyond one meeting

The model you build once doesn't expire after the partner meeting. It becomes the operating instrument you run the company on:

• Board decks pull from the same tabs, so your board numbers never contradict your investor numbers.
• The next round starts from an updated cockpit, not a blank page — you re-forecast in an afternoon.
• Hiring and spend decisions get tested in the model before they're made, because changing a driver shows the cash impact instantly.
• Diligence — when a term sheet comes, the data room's model is already the one that ties.

Illustrative of how a tied-out model is used; figures reference the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE).

From dread to command

Trace the arc this report is designed to produce. You arrived because a partner asked for a model you didn't have, and the deadline was days away. By the balance-sheet page you saw the one fear — the plug — disappear. By the metrics page you saw your claimed numbers become computed ones. By the sensitivity page you saw yourself answer the "what if you miss?" question you'd been avoiding.

That's the product. Not a spreadsheet — the difference between walking in hoping and walking in in command of your own numbers. The next two sections make that concrete: the roadmap from where you are to a funded round, and the exact words for every question you'll be asked.

What the partner writes in their memo

After the meeting, the partner writes you up for their team. Here's the memo a tied-out model earns.

Illustrative of the memo a defensible model supports; not a real memo. Figures are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE).

From here to a funded round

The path, keyed to the model's own milestones. Dollars paired with proof.

• Now — 2026: prove the baseEnd 2026 at $1.8M ARR with NRR above 100%. This is the traction that makes the current raise fundable. The model shows the associate you already hit the numbers behind the story.
• 2027: the $14.0M milestoneThe plan's first modeled raise. It buys the scale to reach $4.3M ARR — pair the dollars with that milestone in your ask.
• 2028–2029: efficient scaleRevenue compounds to $9.1M while losses narrow to $-195K. This is where operating leverage becomes visible — the story that unlocks the next round on better terms.
• 2030: the outcomeExit ARR $28.7M, Rule of 40 = 98, cash $51.8M. The company the base-case model describes.

Framing your ask as dilution + milestone

The strongest ask never leads with a valuation. It leads with a milestone and the capital to reach it: "We're raising to get from $1.8M to $4.3M ARR; here's the model that shows what that takes and what it produces." The valuation follows from the traction, and you've kept the conversation on the ground the model actually supports.

What changes in the room when you have this

With the roadmap on the table, the meeting's center of gravity shifts. Instead of the partner probing whether your numbers are real, they start pattern-matching your milestones against companies they've funded — which is the conversation you want, because it's the one that ends in a term sheet.

Roadmap milestones are the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE) model's own outputs. Your milestones will be your model's outputs when you enter your numbers.

What "data-room ready" looks like

The model that wins the meeting is the same one that clears diligence — if you keep it current.

Because the model is driver-based and tied out, "data-room ready" isn't a second project — it's the same workbook, kept current. That's the compounding payoff of building it right once.

Defend every assumption they'll poke

The eight questions that decide most rounds — and the answer to each, in your words. This is the workbook's ninth tab — the "defend your assumptions" sheet — worked in full.

Defenses reference the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE) model's own drivers and outputs. When you enter your numbers, each answer updates to your figures.

Turn the defenses into reflexes

Reading the defenses isn't enough — you want them automatic, so a hard question produces a calm number instead of a flinch. Two exercises before the meeting:

1. The 60-second walkthrough. Practice narrating the model start to finish in one minute: drivers → revenue → statements → it ties → metrics → runway. If you can do that cleanly, you can handle any drill-down.
2. The hostile round. Have a friend fire the eight questions above in random order. Answer each with a number and its source. Do it until there's no gap between the question and your pointer to the tab.

Educational analysis and a spreadsheet template — not investment, legal, financial, or tax advice, and not a valuation opinion or an offer of securities. All figures are illustrative outputs of a deterministic model driven by the assumptions shown; change an assumption and every figure changes. Consult qualified professionals before making any financing or investment decision.

Do this before the meeting

Ordered by deadline. The first list is what saves this week.

Next 72 hours — before the partner call

• Get the workbook and open the cockpit. One tab of drivers; everything else is a formula off it.
• Enter your eight core drivers: starting logos, Y1 new logos, new-logo growth, ARPA, gross margin, monthly churn, blended CAC, starting cash.
• Confirm the balance check reads 0.00 in every year. If it does, your model ties — the single most important thing for the meeting.
• Sanity-check three metrics against your gut: NRR, CAC payback, runway. If any looks wrong, a driver is wrong — fix the input, not the output.
• Export the ARR bridge and the balance sheet as the two exhibits you lead with.
• Rehearse the 60-second walkthrough (page 60) once out loud.

Next 30 days — make it yours

• Replace every default with your real data, including working-capital days (DSO/DPO) and deferred-revenue mix from your accounting.
• Reconcile the model's Year-1 to your actuals. If they don't match, adjust drivers until they do — this is how you earn the forecast.
• Build your three scenarios (base / conservative / downside) by saving three copies of the cockpit.
• Back each key driver with a source: cohort retention curve behind churn, hiring plan behind new logos, price list behind ARPA.
• Rehearse the hostile round (page 60) with a friend until answers are reflexive.

Next 90 days — operate on it

• Update the cockpit monthly with actuals; watch where reality diverges from plan and why.
• Use it for real decisions: test a hire or a spend increase in the model before committing.
• Make it your board exhibit, so board and investor numbers come from one source.
• Keep the data room's model current, so when a term sheet comes, the model that ties is already there.

The one-page pre-meeting checklist

Scripts for the questions you'll get

Word-for-word openers. Adapt to your numbers; keep the structure — number first, source second.

Two rules that make every script land

1. Number first, source second. "118.5%, computed from the bridge" beats "our retention is strong." A number with a source ends a line of questioning; an adjective invites three more.
2. Offer to open the file. "Check any year" signals you have nothing to hide. Founders who invite the audit almost never get audited hard; founders who deflect always do.

Scripts reference the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE) model. Replace the figures with your own after you enter your drivers.

Every metric, with its exact formula

So you compute each the way the investor does — no off-by-a-definition surprises. Keep this open in the meeting.

These are standard public SaaS-metric definitions; each is spot-checked against its formula in the engine's test suite. Exit values are engine outputs for the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE).

The mistakes that sink models

Every one of these is avoidable — and every one has cost a founder a round.

Bring this to your advisor, CPA or lawyer

A one-page brief so a professional can review your model fast — and so you get honest, in-scope advice.

ModelKit is an educational template, not a substitute for your accountant or counsel. Hand them this sheet with your completed workbook.

For your CPA / bookkeeper

• Verify the opening balances — starting cash and paid-in capital — against your books.
• Confirm DSO/DPO and deferred-revenue assumptions match your revenue-recognition policy.
• Check the tax treatment; the template applies tax only to positive pre-tax income and defaults the rate to 0 for pre-profit years — confirm this fits your situation.
• Reconcile Year-1 to your actuals and flag any driver that produces a figure you can't support.

For your lawyer (financing)

• Confirm the raise assumptions in the cockpit reflect your actual or intended round structure.
• Review any figures you plan to put in front of investors for consistency with your disclosures.

What this model is — and isn't

Questions to ask your advisor

• "Do these opening balances match our books?"
• "Is our revenue-recognition and deferred-revenue treatment right here?"
• "Are any of these figures inconsistent with what we've told investors or filed?"
• "What would you change before I put this in a data room?"

Educational analysis and a spreadsheet template — not investment, legal, financial, or tax advice, and not a valuation opinion or an offer of securities. All figures are illustrative outputs of a deterministic model driven by the assumptions shown; change an assumption and every figure changes. Consult qualified professionals before making any financing or investment decision.

Turn the model into a board deck

The model doesn't retire after the round — it becomes how you report. Four slides straight from the tabs.

• Slide 1 — ARR bridge. Beginning + new + expansion − churn. The single exhibit that shows the quarter's story.
• Slide 2 — the scorecard. ARR, NRR 118.5%, CAC payback, Rule of 40 vs. plan.
• Slide 3 — cash & runway. The cash line, burn, and the milestone this runway buys.
• Slide 4 — plan vs. actual. Where reality diverged from the cockpit, and the driver you're changing in response.

Because each slide reads from a tab, updating the board deck is updating the cockpit — a monthly habit, not a monthly project. That's the operating dividend of building the model once, correctly.

Illustrative board structure; figures reference the fictional Cadence Robotics, Inc. (FICTIONAL EXAMPLE).

Your next step

You've seen the whole model. Here's how to be holding your own version by tomorrow.

The natural next step is the workbook: the same nine wired tabs you've just read, unlocked, so you drop your drivers into the cockpit and every page in this report regenerates on your numbers. If you'd like a second pair of eyes on the spreadsheet mechanics — how to wire your numbers in and make your balance sheet tie — the review-call tier adds a 45-minute working session (mechanics only; not round strategy, valuation, or investment advice).

Guarantee & the fine print

How to vet it — without a bio

ModelKit is built on public, checkable frameworks rather than a name you'd have to take on faith: standard three-statement mechanics, public SaaS-metric definitions, Damodaran-style DCF, YC/NVCA materials, and SEC/EDGAR & standard exchange-filing conventions. No client data and no resold source files went into it. Instead of asking you to trust a résumé, everything you'd use to judge it is open before you pay — this full 68-page worked sample, the raw computed JSON, and a runnable engine test that enforces 217 checks, including the balance identity (A − L − E = 0) in every year. If the math holds up under your own diligence, it holds up regardless of whose name is on it.

Educational analysis and a spreadsheet template — not investment, legal, financial, or tax advice, and not a valuation opinion or an offer of securities. All figures are illustrative outputs of a deterministic model driven by the assumptions shown; change an assumption and every figure changes. Consult qualified professionals before making any financing or investment decision.