Venture & Private Capital · Capstone brief
INV-02 · Who gets paid at $40M, $120M, $400M?
If it is sold for $40 million, $120 million, or $400 million, who gets paid what?
The question
Acme Technologies has raised three rounds. If it is sold for $40 million, $120 million, or $400 million, who gets paid what? Model all three exits, find every point where an investor's choice changes, and write the one-page explanation a common-stock employee would actually understand.
A student who finishes this brief should be able to tell an Acme engineer, in plain words, why a $40 million sale of "her" company might be worth much less to her than she thinks.
Data room / materials
CASE-ACME only. Read data-room-index.md first. The student needs:
- The current cap table —
cap-table-series-b.csv: two founders, two option-pool lines, Seed, Series A, and Series B. - The term-sheet summaries —
term-sheet-series-a.mdandterm-sheet-series-b.md. The Series B is participating with a 3x cap; Series A and Seed are non-participating. Series B is senior to Series A, which is senior to Seed. - The exit scenarios —
exit-scenarios.md: $40M, $120M, and $400M, each already net of debt and fees. - The CapStack waterfall calculator (see Tools). It runs in the browser; nothing typed into it leaves the tab.
- A spreadsheet to rebuild at least the $40M case by hand.
The calculator is the source of truth for the numbers. The student's work is to enter the terms correctly, prove the results add up, find the kinks, and explain them.
Do not use any AI tool to produce the payout table. If a student wants to see what a model does with this cap table, that is lab L-08, and it comes after the table is built.
Expected failure modes
Entering a class with the wrong participation setting or rank — the most common error, and the calculator will not catch it. Forgetting the second option-pool line. Rebuilding the spreadsheet and the calculator from different inputs, then "fixing" the spreadsheet to match. Reporting three exits and missing the kinks between them. Writing the employee page in investor vocabulary ("1x participating capped") that the employee cannot read. Saying "options are worth $X" without saying it assumes every option is granted, vested, and exercised.
Done looks like
A packet with five pieces:
- Inputs table — every class as entered in the calculator, with the case-file line it came from.
- Payout table — every holder's payout at $40M, $120M, and $400M, plus the per-share value for common. Each column sums to the exit value to the cent; the packet shows the check.
- Hand rebuild — the $40M case rebuilt in a spreadsheet, step by step: preferences by rank, then who converts, then how the remainder is shared. It matches the calculator.
- Kink map — every exit value where a holder's election changes (takes preference, participates, hits its cap, converts), found by stepping the exit value in the calculator, each stated to the nearest $0.5M, with a sketch of payout against exit value.
- The employee page — one page, no jargon, for an Acme employee holding options: what her shares are worth at each exit, why the first dollars go to investors, the one stretch where investors' payouts stop rising and hers rise faster, and the assumptions behind the numbers. It ends with: "This is a class exercise on a fictional company. It is not legal, tax, or investment advice."
Five C's
CT: separating what the terms say from what the deck implies. CR: finding the kinks by deliberate stepping rather than by luck. CO: a peer enters the cap table independently; any difference in inputs is resolved before the payout table is final. CM: the employee page, read by someone outside the class who has never seen a cap table. CZ: the people who hold common stock and are usually the last to see this math.
Mentor role
A finance professional, startup lawyer, or startup employee reviews the inputs table at Checkpoint 1 — before any payouts are reported, since a wrong participation setting poisons everything after it — and the employee page at Checkpoint 2. Standing instruction: reject any payout table that does not sum to the exit value, and any employee page a non-specialist could not follow. School-supervised, both times.
Rubric calibration
R1: one company, three exits, one audience. R2: every input traced to a case-file line; every column sums to the exit. R3: the calculator is the baseline, and the hand rebuild proves the student understands it. R4: every kink found; the option-exercise assumption stated. R5: the employee page passes the outside-reader test. R6: names who is last in line and refuses to describe options as "worth" a number without its assumptions.
Two ways this goes wrong
(a) A clean three-column table built on Series B entered as non-participating, so every number is wrong and nothing looks wrong. (b) Correct numbers and a kink map, then an employee page full of "liquidation preference" and "as-converted" that the employee cannot read.
Checkpoint suggestions
- Week 1: Labs L-01 to L-04 complete. Inputs table built and peer-checked; mentor review of inputs before any payouts are reported.
- Week 2–3: Payout table with sum checks; $40M hand rebuild matching the calculator; kink map.
- Week 4: Employee page drafted, read by an outside reader, revised; teacher checks the packet against the answer key.
Credit lane fit
Lane A immediately (business or finance capstone, an economics or statistics project, or a business-education competitive event used as a rubric crosswalk). Lane B only where the school's work-based-learning coordinator owns the record. No verified credit claim.
An investment study, not an investment. Fictional case files. Not legal, tax, or investment advice.