Venture Capital Modeling Test: Practice Case and Answers

A venture capital modeling test is a spreadsheet exercise used in VC interviews to check that you can build a cap table, model dilution across rounds, calculate exit proceeds for each shareholder, and turn the numbers into an investment view. Unlike a private equity test, it is light on accounting and has almost no discounted cash flow work: a pre-revenue company has no cash flows to discount.

The math is rarely the hard part. In our view, much of what gets graded is the judgment behind your assumptions, and whether the model still holds together when the grader changes them.

This guide covers what the test includes, what graders score, a worked practice case with every number shown, and a four-week prep plan. For the written, qualitative exercise, see our venture capital case study interview guide; for the same math done out loud without a spreadsheet, see the venture capital cap table interview guide.

What a venture capital modeling test usually covers

Mergers & Inquisitions notes that VC case studies tend to be more qualitative than in other finance roles, but it also describes cap table exercises in which candidates calculate proceeds for each investor group across a range of exit values. In practice, tests fall into four types:

Cap table and dilution. Given founder shares, an option pool, SAFEs, and priced rounds, build the fully diluted cap table after each round.
Exit waterfall. Given liquidation preferences and share classes, calculate what each holder receives at several exit values.
Returns analysis. Calculate multiple on invested capital (MOIC) and IRR for the fund's position, and whether the deal can return a meaningful share of the fund.
Operating model or unit economics. For growth-stage roles, project revenue, burn, and runway from customer data, then flag what you would question.

Formats vary by firm. Some run the exercise in the office at a laptop; others send it as a take-home. There's no market-standard time limit, so the instructions are your best guide. Confirm the window, the file format, and whether a written recommendation is expected alongside the spreadsheet. It often is.

What the modeling test checks: assumptions as much as mechanics

At most firms, these are calculations junior investors do regularly: checking a founder's cap table, modeling a SAFE conversion, estimating ownership at exit. The survey of 885 VCs at 681 firms by Paul Gompers, Will Gornall, Steven Kaplan, and Ilya Strebulaev (NBER Working Paper 22587) found that cash-on-cash multiples are the most commonly used metric, cited by 63 percent of the sample, with IRR next at 42 percent, and that the median required multiple is 5x and the average required IRR 31 percent.

The arithmetic behind those numbers is a few lines. What separates candidates is which assumptions they choose, and whether they say so out loud.

Graders typically look for:

Accuracy. Share counts and ownership percentages that tie out, and proceeds that sum to the exit value.
Structure. Clear inputs, calculations, and outputs, with hard-coded numbers in one place.
Correct mechanics. Pre-money vs post-money, option pool placement, SAFE conversion, and preference treatment.
Sensitivity. Results at several exit values, not one.
Judgment. A short recommendation that uses the numbers: is the price right, and what has to be true?
Stated assumptions. Future dilution, exit timing, and the exit multiple are your choices, not the firm's. Put them in labelled cells and defend each in one line.

Worked practice case (illustrative): a two-round exit waterfall

Try this before reading the answer.

Setup. After a Series A, the fully diluted cap table is: founders 55 percent, option pool 10 percent, seed investors 15 percent, Series A investors 20 percent. Seed invested $3M; Series A invested $10M at a $50M post-money. Both hold 1x non-participating preferred, pari passu. Calculate proceeds at exits of $12M, $40M, and $100M.

Method. Each preferred class takes the greater of its preference or its as-converted share. Test conversion for each class, starting with the one that converts last (here, Series A, which needs the highest exit to beat its preference).

$12M exit. Total preferences are $13M, more than the exit. Both classes take preferences and split pro rata: Series A gets about $9.23M, seed about $2.77M, common gets nothing.

$40M exit. Series A as-converted would get 20 percent, or $8M, less than its $10M preference, so it takes $10M. The remaining $30M goes to common and converting seed. Seed's share of that pool is 15 divided by 80, or 18.75 percent, which is $5.625M, more than its $3M preference, so seed converts. Founders get $20.625M and the option pool $3.75M. Total: $40M. Series A earns 1.0x; seed earns about 1.9x.

$100M exit. Series A as-converted gets $20M, more than $10M, so everyone converts. Series A gets $20M (2x), seed $15M (5x), founders $55M, pool $10M.

Breakpoints. Seed converts once 18.75 percent of (exit minus $10M) exceeds $3M, at a $26M exit. Series A converts at a $50M exit.

Notice what happened at $40M. The later investor got its money back. The earlier one nearly doubled. Same company, same exit, very different outcomes.

Recommendation line. "At a $50M post, the Series A needs roughly a $250M exit before dilution to reach 5x; the question for diligence is whether this market supports that outcome." For a deeper treatment of the mechanics, Hustle Fund's 2026 guide to exit waterfall modeling walks through a seven-step process, and our liquidation preference guide explains the terms.

"But I should show I can build a full model"

The instinct makes sense, especially if you come from banking. A bigger model looks like more work, and more work feels safer.

But in venture, the extra tabs tend to hide the thing graders want to see: the three or four assumptions that drive the answer. A tight model with a clear recommendation reads as an investor. A sprawling one reads as someone who hasn't decided yet.

Where the modeling test sits in a VC hiring process

In common practice the modeling test comes late, after a screening call and at least one investor conversation, because it is expensive for the firm to grade. Timelines are set firm by firm and no market-wide data on them exists, so ask the recruiter where the exercise sits and what follows it. Whatever follows, a short note to each grader naming the assumption you'd revisit first can help; the venture capital interview thank you email guide has the post-exercise version.

Knowing the pay band helps you decide how much prep a role justifies. Venture5's 2025 Venture Capital Salary Survey of more than 700 US venture professionals at more than 50 firms reports median base salaries of about $80K for analysts, $130K for associates, and $150K for senior associates. Those are base figures only; the public page gives no bonus amounts, and its detailed carry data is reserved for survey respondents.

A four-week prep plan for the venture capital modeling test

Week 1: mechanics. Build a cap table from scratch with founders, a pool, and two priced rounds. Model a post-money SAFE conversion using the ownership math in Y Combinator's 2018 post-money safe primer, and read the current safe user guide (v1.2) for how the forms themselves work, including the point that a new or increased option pool adopted as part of the Series A does dilute safe holders.
Week 2: waterfalls. Model 1x non-participating, participating, and capped participating preferences at five exit values. Check that every column sums to the exit.
Week 3: returns. Calculate MOIC and IRR for a position, including dilution from future rounds. Carta's State of Private Markets: 2025 in Review reports that median dilution on its platform across rounds from seed through Series C fell over the year from about 18 percent to about 16 percent, so roughly 16 percent per round is a defensible default to write in your assumptions box.
Week 4: timed practice. Do two full mock tests under a 90-minute limit, then write a one-paragraph recommendation for each.

Checklist before you submit:

Inputs are color-coded and in one place.
Ownership sums to 100 percent after every round.
Proceeds sum to the exit value in every scenario.
The recommendation states a view, not just a number.

Where candidates lose points on a VC modeling test

Building a DCF. Discounting cash flows a pre-revenue company doesn't have is one of the clearest signs you brought a banking template to a venture problem.
Confusing pre-money and post-money. Ownership equals investment divided by post-money. See post-money valuation.
Putting the option pool in the wrong place. Pools created before the round dilute only existing holders.
Forgetting to test conversion. Under standard terms, non-participating preferred gets either its preference or its as-converted share, not both.
Hard-coding outputs. Graders often change inputs to test whether your model updates.
Unstated assumptions. Burying a 20 percent dilution figure or a 10x exit multiple inside a formula makes the output hard to test.
No written takeaway. A correct model with no view reads as a banking answer, not an investing one.

Getting real reps before the interview

Good assumptions come more from watching how real rounds get built than from rebuilding the same template. 1752vc's Venture Fellow program spends eight weeks in live virtual sessions on real pitch materials and diligence for live companies, which is where the inputs to a model like the one above come from: the round a founder is really raising, the pool a board really demanded. Fellows leave with a certification. More interview guides live on the venture capital hub.

The bottom line

A VC modeling test is short on math and long on judgment. Get the mechanics right, make every assumption visible, and end with a sentence that takes a position.

Bankers get graded on the model.

Investors get graded on the call.

Key takeaways

A venture capital modeling test checks cap table, dilution, exit waterfall, and returns skills, usually in a short timed format.
VC tests are lighter on accounting than private equity tests; the recommendation you attach often matters as much as the math.
For waterfalls, it helps to test each preferred class's choice between preference and conversion and confirm proceeds sum to the exit.
Labelling assumptions helps: Carta's 2025 year in review says median dilution across rounds from seed through Series C fell over the year from about 18 percent to about 16 percent, which is a defensible default to state and defend.
Four weeks of structured practice on mechanics, waterfalls, returns, and timed mocks is, in our view, enough for most candidates.

Frequently asked questions

Do venture capital interviews have modeling tests?

Some do. Growth-stage and multi-stage firms are more likely to give a spreadsheet test, while many seed funds use a qualitative case study instead. Mergers & Inquisitions notes that VC cases tend to be more qualitative than in other finance roles.

What is on a VC modeling test?

Common tasks are building a cap table across rounds, converting SAFEs, calculating an exit waterfall with liquidation preferences, and computing MOIC and IRR. Some tests add an operating model or customer data analysis, followed by a short written recommendation that uses the numbers to take a view on price and risk.

How long is a venture capital modeling test?

There is no standard length, and no public data on it. Firms set their own window: some run the exercise inside an office visit, others send it as a take-home. It is worth asking for the time limit, the file format, and whether a written recommendation is expected, and treating the answer as part of the brief.

How do I prepare for a VC modeling test without finance experience?

A common approach is to build a cap table and an exit waterfall from a blank sheet, then practice returns math with dilution applied round by round. Y Combinator's post-money safe primer is a good reference for conversion math, and timed mocks help until your proceeds consistently tie back to the exit value.

How is a VC modeling test different from a private equity modeling test?

A private equity test usually centers on a leveraged buyout model with debt schedules and three linked statements. A VC test drops almost all of that, including the DCF, and focuses on ownership, dilution, preferences, and exit scenarios, with far more weight on the assumptions you choose and the investment view you draw from them.

Sources

Disclaimer: This guide is for general education only and is not legal, tax or investment advice. Laws, market data and program terms change, so it may not reflect the latest developments or fit your situation. Treat it as a starting point, not a source of truth, and talk to a qualified lawyer, accountant or financial adviser before you make decisions.