Exits, Fundraising, and the Capital Landscape

A well-run sale is a process, not an event. From the decision to sell to money in the bank typically takes six to nine months, and the sequence is standard enough that most investment bankers run some version of the same playbook.

Auction or Proprietary

The first decision is how many buyers to invite. In a banker-run auction, an investment bank contacts a curated list, often 20 to 100 potential acquirers, and runs them through parallel rounds on a fixed timetable. Competition is the entire point: a second bidder is the single most reliable way to raise the price. In a proprietary (or negotiated) deal, one buyer approaches directly, often after a partnership or a casual "we should talk" from corporate development. Proprietary deals close faster and leak less, but the seller negotiates without the leverage of an alternative. The practitioner rule is blunt: one buyer is not a buyer, it is a hostage-taker. Even a company flattered by inbound interest should quietly create competition before signing anything.

Teaser, CIM, and Indications of Interest

The bank prepares two documents. The teaser is one to two anonymous pages covering industry, rough revenue, and growth, sent broadly to test appetite without revealing the company's name. Buyers who sign an NDA receive the CIM (confidential information memorandum), a 30-to-80-page book covering the business, financials, and projections. It is the Week 9 pitch deck's heavier sibling: same narrative discipline, aimed at an acquirer instead of a VC. Interested buyers then submit IOIs (indications of interest), non-binding letters with a valuation range and deal structure, which the bank uses to cut the field to a handful of serious parties.

Management Meetings and the LOI

The shortlist gets management meetings: full-day presentations where the executive team walks the buyer through the business and, just as importantly, the buyer evaluates whether it wants to own this team. Final bids follow, and the seller picks one to sign an LOI (letter of intent). The LOI states price and structure but is non-binding on almost everything except exclusivity. The seller agrees to negotiate with no one else, typically for 30 to 60 days. This is the moment leverage flips. Before the LOI, the seller has competing bidders; after it, the buyer knows the alternatives have been sent home, and every issue discovered in diligence becomes an argument for a lower price. Sellers should therefore negotiate hard on price and terms before signing, and keep exclusivity as short as the buyer will accept.

Confirmatory Diligence to Close

During exclusivity the buyer runs confirmatory diligence, the same checklist a VC runs in Week 3 (corporate records, IP assignments, contracts, financials, litigation), but deeper, because the buyer is purchasing all of the liabilities, not a minority stake. Clean data rooms close on schedule; missing IP assignments and surprise contracts become price reductions. The lawyers then negotiate the definitive agreement (the binding purchase contract with representations, warranties, and indemnities), and the deal signs. Closing follows once conditions are met (regulatory clearance where required, third-party consents, and shareholder approval, which is where the preference stack you will meet in the next section gets applied). Then the wires go out.

The headline price is not what sellers receive. Three standard terms sit between the announced number and the wire, and each one moves real money.

Earnouts: The Discounted Dollar

An earnout is contingent consideration: part of the price is paid only if the business hits agreed targets such as revenue, milestones, or retention over one to three years after closing. Buyers love earnouts because they bridge valuation gaps and shift risk to the seller. Sellers should treat earnout dollars with suspicion, for a structural reason: after closing, the buyer controls the resources, the priorities, and often the accounting definitions that determine whether the targets are hit. Earnouts routinely pay out only in part, and disputes over them are among the most common sources of post-closing litigation. The Week 5 discipline applies directly: value an earnout at its probability-weighted expected value, not its face amount, and assume the probability is lower than the buyer's cheerful projection.

Escrows and Holdbacks

A portion of the purchase price, conventionally about 10 to 15 percent, is held in escrow for roughly 12 to 18 months after closing. If the seller's representations turn out to be false (undisclosed liabilities, IP problems, tax exposure), the buyer claims against the escrow instead of suing shareholders one by one. The escrow is the enforcement mechanism for indemnification: the seller's contractual promise to make the buyer whole for breaches, usually capped at the escrow amount for ordinary reps, with higher or uncapped exposure for fundamental matters like ownership of the shares and fraud. For planning purposes, sellers should treat closing proceeds as roughly 85 to 90 percent of the price now and the remainder as a delayed, at-risk payment.

Running the Price Through the Preference Stack

Whatever survives earnouts and escrow runs through the Week 6 waterfall. A worked example, with the Week 6 machinery exactly as taught:

• A company sells for $60M in cash.
• The preferred investors put in $20M with a 1x non-participating liquidation preference and hold 25% of the company as-converted. Common (founders and employees) holds 75%.

Non-participating preferred takes the greater of its preference or its as-converted value:

• Preference: 1x on $20M = $20.0M.
• As-converted: 25% of $60M = $15.0M.

$20M beats $15M, so the preferred takes the preference and does not convert. Common receives the remainder: $60M − $20M = $40.0M, which is 66.7% of the deal despite common holding 75% of the shares. The cap table said 75/25; the waterfall pays 66.7/33.3.

Formula. Conversion point = liquidation preference ÷ as-converted ownership. Here: $20M ÷ 0.25 = $80M. Below an $80M exit, the preferred takes its preference; above $80M, it converts and shares pro rata.

Now layer the earlier terms back on. With a 12.5% escrow, $7.5M of the $60M is held back and only $52.5M wires at closing, allocated in the same waterfall proportions, with the escrowed remainder following 12 to 18 months later if no claims arise. If $10M of the headline price had been an earnout instead, the contractually assured portion of the deal is really $50M, and the preferred's $20M preference consumes 40 percent of it. Deal structure and the preference stack interact; neither should be evaluated in isolation.

Two alternatives to the bookbuilt IPO got heavy use in recent cycles, and each is best understood by what it gives up.

A direct listing puts existing shares on an exchange with no underwritten offering: no bookbuilding, no allocation list, and traditionally no new capital raised (the SEC has since permitted primary raises alongside a direct listing, though they remain rare). The opening price is set by an exchange auction matching actual supply and demand, which is precisely why Spotify (2018) and Slack (2019) chose it: no 18% gift to an allocation list, and typically no lockup, so existing holders can sell immediately. What it trades away: the guaranteed capital raise, the underwriter's marketing and aftermarket support, and the curated shareholder base a bookbuild produces. It suits companies with strong brands and no need for cash; it does little for a company that actually needs to raise.

A SPAC (special purpose acquisition company) inverts the sequence: a shell company raises cash from public investors first (units priced at $10, proceeds held in trust) and then hunts for a private company to merge with, the so-called de-SPAC. The private company gets a negotiated price and a faster path to being public. What it trades away is substantial: the sponsor typically receives a promote of roughly 20% of the SPAC's shares for a nominal price, public SPAC holders can redeem their $10 rather than fund the deal (so the cash actually delivered is uncertain), and warrants add further dilution. The 2020-2021 SPAC boom ended with most de-SPACed companies trading far below $10, and the structure has since retreated to a niche.

The three paths side by side

The lesson is that there is no free path to the public markets. The traditional IPO pays in underpricing and fees, the direct listing pays in foregone capital and support, and the SPAC pays in promote and dilution. The right question is Week 8's question, transplanted: which cost are you best positioned to bear?

Everything so far returns capital at the end of the story. A secondary sale returns it in the middle: an existing shareholder (founder, employee, or early investor) sells shares to a buyer, and no new money reaches the company. As companies stay private longer (a decade or more is now routine), secondaries have become the main pressure valve for people whose net worth is locked in a certificate.

The common structures:

• One-off sales: a holder finds a buyer directly or through a secondary broker. Slow, and subject to company transfer restrictions.
• Company-sponsored tender offers: the company organizes a window in which employees and early holders can sell to a designated buyer (often the new lead investor), usually alongside a primary round. This is the clean, scaled version: one price, one buyer, one legal process.
• Founder secondaries inside a round: a portion of a large primary round is used to buy founder shares directly, typically justified to investors as taking sell-the-house pressure off the founders so they can keep swinging for the large outcome.

Nearly every startup's bylaws include a ROFR (a right of first refusal), giving the company, and often major investors, the right to buy any shares a holder proposes to sell, at the negotiated price, before the outside buyer can. The ROFR protects the cap table from unknown shareholders, but it also chills the market: buyers hesitate to spend weeks negotiating a purchase the company can simply take over at the last step.

Why secondaries price at a discount

A common mistake is to treat the last round's price per share as the price of the company's stock. It is not: it is the price of preferred stock, and Week 6 explains why that matters. The Series D investor who paid $10 per share bought a liquidation preference, antidilution protection, and other rights; a secondary buyer purchasing an employee's common stock buys none of that and sits at the bottom of the waterfall. Add illiquidity, thin information (a secondary buyer gets nothing like the Week 3 diligence access a lead investor gets), and ROFR friction, and common in secondary transactions routinely trades at a discount to the last preferred round: practitioner ranges of roughly 20-30% are common, wider for shakier companies, and the discount can compress to zero or even invert into a premium for the handful of names every fund wants exposure to.

For the seller the discount is usually worth paying: a certain dollar today against a Week 5-style weighted set of outcomes in which the modest-exit scenarios may leave common with little. For the company, one caution: a large common secondary at a high price is a data point that can pull the 409A valuation (and therefore new employees' option strike prices) upward.

An acquihire is an acquisition whose real object is the team, not the product. The product is usually shut down within months. Pricing follows a convention that has nothing to do with revenue: buyers think in per-engineer terms, with practitioner ranges on the order of $1-3 million per retained engineer depending on seniority, specialty, and how hot the talent market is. A twelve-person team might fetch a headline price in the low tens of millions, which sounds like an exit until you run it through the waterfall.

Run the Week 6 arithmetic. Suppose a company raised $8 million of preferred with a standard 1x liquidation preference and is acquihired for $9 million. The preference stack takes the first $8 million; common (the founders and every employee holding options) splits the remaining $1 million. If the price had come in at $7.5 million, common receives exactly nothing. Acquihire prices are so often in the neighborhood of the preference stack precisely because buyers know the stack is the investors' walk-away floor: pay enough to get the investors to consent, and not much more.

Now the part that creates real conflict: the buyer's total spend is split between purchase price, which flows through the waterfall and mostly reaches the preferred, and retention packages, signing bonuses and multi-year equity grants paid directly to the employees who join. Retention comp bypasses the waterfall entirely. The buyer is often indifferent between paying $9 million of price plus $4 million of retention or $6 million of price plus $7 million of retention; the founders and engineers are decidedly not indifferent, and neither are the investors, whose recovery falls dollar for dollar as value migrates from price to retention. Boards scrutinize this split because directors owe fiduciary duties to shareholders, not to the employees being retained. In practice, the negotiation over the split is the acquihire.

Closing the loop: exits, DPI, and the power law

Week 7 defined DPI (distributions to paid-in capital) as the metric that moves only when cash actually comes back. Each exit type moves it differently. Acquihires typically return a fraction of invested capital to the preferred and round to zero for the fund's returns. Modest acquisitions return capital plus something. Secondaries generate early partial DPI, which is exactly why later-stage funds prize them. IPOs can drive enormous DPI, but slowly: the lockup delays any sale by 180 days, and funds typically sell down or distribute shares over quarters afterward, so IPO-driven DPI arrives years after the headline.

Seen this way, Week 7's power law is not just a distribution over outcome sizes; it is a distribution over exit types. A typical early-stage portfolio resolves into many shutdowns and acquihires that return almost nothing, a band of modest acquisitions that return one to a few times the money, and, rarely, the large acquisition or IPO that returns the entire fund. The fund's DPI is, to a first approximation, determined by whether that last category shows up at all. That is why investors push portfolio companies toward the big outcome even when a comfortable acquihire is on the table: the fund model does not need more small exits; it needs one large one.

How you enter the funnel determines its conversion rates. Cold outreach, an unsolicited email to an investor with no shared connection, converts poorly, with response rates in the low single digits at most funds. The reason is not rudeness; it is filtering economics. A partner sees thousands of companies a year and can diligence a few dozen, so the first filter is the network itself. A warm introduction, a referral from someone the investor already trusts, carries a signal a cold email rarely can: a person with reputation at stake chose to spend it on you. This is the Week 8 signaling lens applied to outreach. The strongest introduction path of all is a founder the investor has already backed, because that referral is informed by having worked with the investor, not just met them.

The practical method is unglamorous: map who you know in common with every name on the target list, ask each connector for an introduction, and make the ask easy with a short forwardable note of three or four sentences the connector can send without writing anything themselves. Names with no warm path either wait or get a carefully personalized cold note, with expectations set accordingly.

Run the Conversations in Parallel

The single most common process mistake is raising sequentially: pitching the favorite fund first, waiting for its answer, then moving to the next. Sequential raising fails twice. Each pass through one fund's process costs weeks, so ten sequential conversations can consume a year. Worse, it destroys negotiating position: an investor who knows they are the only conversation has little reason to move quickly or price generously, and a term sheet typically comes with a short fuse, often a week or two, so a founder with no other live conversations is left to take the only offer on the table or start from zero.

Running conversations in parallel batches fixes both problems. Launch outreach to the list in waves over two to three weeks so that first meetings cluster, partner meetings cluster, and, if the funnel converts, term sheets arrive within days of one another. Simultaneous term sheets are option value: the founder compares terms instead of accepting them, and each investor prices knowing others are pricing too. Batching has a second use: some founders schedule a first wave of lower-conviction targets to pressure-test the pitch, saving the highest-conviction funds for the second wave when the answers are sharp.

Parallelism is also how you protect momentum. Investors read time-on-market as a signal: a round that has been open for five months prompts the question every founder dreads (why hasn't anyone taken this?) regardless of the company's merits. A compressed process rarely lets that question form.

Inside a multi-partner fund, a deal does not go from one meeting to a wire. It climbs a meeting cascade, and each level screens for something different. Knowing what each stage is actually testing keeps founders from over-preparing for the wrong questions and from mistaking enthusiasm at one level for a decision by the fund.

The associate screen is a filter, not a negotiation. The associate's product is a recommendation upward, so the founder's job in that meeting is to make the fit legible fast, covering stage, sector, traction, and round size, and to give the associate the material to champion the deal internally. The partner meeting is where conviction is built or lost: the partner who takes the deal forward becomes its sponsor, staking internal credibility on it, and sponsors generally spend that credibility only on companies they believe can return the fund. The final stage is the full partnership, or investment committee (IC): the body with formal authority to approve an investment. This is where the sponsor presents and the other partners stress-test. A term sheet normally follows IC approval, not partner enthusiasm; "I love this, let me socialize it with my partners" is a real stage of the process, not a soft yes.

Diligence Runs Both Directions

While the fund diligences the company, the founder should be running the Week 3 discipline in reverse: diligence on the investor. This investor will hold a board seat and information rights for the better part of a decade, a relationship far longer than most jobs and nearly impossible to unwind. The single best source is reference calls with portfolio founders, and the calls that matter most are with founders whose companies struggled or failed. Most investors behave well when the chart goes up and to the right. What you are buying is the investor's behavior when it doesn't: Did they bridge the company or abandon it? How did they act in the down round? Did they help recruit, or just attend board meetings? Ask the investor directly for references, including a failed company, and treat reluctance to provide one as data. Confirm the practical facts too: whether the fund has capital left to deploy, its typical follow-on reserves, and who would actually sit on your board, because the partner who charmed you in the pitch is not always the partner you get.

The term sheet Week 6 taught you to negotiate is, with narrow exceptions, not a binding contract. The economics and control terms are statements of intent; typically only two provisions bind: exclusivity (a no-shop clause, commonly 30 to 45 days, during which the founder may not solicit other offers) and confidentiality. Everything else must be converted into definitive documents before any money moves.

The document stack in a priced round

For a standard priced equity round, the industry works from the NVCA model documents, and the stack has five pieces:

• Stock Purchase Agreement (SPA): the actual sale of shares. It contains the company's representations and warranties (factual statements about the business that, if false, give investors a claim) and the conditions that must be satisfied before closing.
• Amended and Restated Certificate of Incorporation: where the preferred stock's rights actually live. The liquidation preference, antidilution formula, and protective provisions from Week 6 are not in the SPA; they are written into the charter itself.
• Investors' Rights Agreement (IRA): information rights (regular financial reporting), registration rights for a future IPO, and the pro rata right to invest in future rounds.
• Right of First Refusal and Co-Sale Agreement: if a founder tries to sell shares, the company and then the investors get first claim (ROFR), and investors can sell alongside the founder (co-sale). This is the machinery that constrains the secondary sales discussed in Part One.
• Voting Agreement: board composition and the drag-along provision that can compel minority holders to approve a sale.

In parallel with drafting, the investor runs confirmatory diligence: verifying that the company they described in the term sheet is the company in the data room. This is checking, not exploring. The investment decision was made before the term sheet. But confirmatory is not ceremonial. A material gap between what was represented and what the documents show can reprice the round, add escrow-like protections, or end it, and the no-shop clause means the founder has already sent every other interested investor away.

Closing mechanics

In venture practice, signing and closing are usually simultaneous: the day the documents are executed, the conditions are confirmed and the wires go out. Larger or multi-investor rounds sometimes run an initial closing with the lead and one or more subsequent closings, within a window the SPA defines, to collect smaller checks that were not ready on day one.

SAFEs work differently, and the difference matters for cash planning. A SAFE round has no single closing. Each SAFE is a standalone two-party contract: one signature block, one wire, often within days of a verbal yes. Founders run rolling closes: money lands investor by investor, usable immediately, while the round stays open. That speed is the operational reason Week 6's convertible instruments dominate early rounds: a priced round's five-document stack takes weeks of legal work and coordination; a SAFE takes an afternoon. The trade is that the priced round's governance and investor protections are exactly what the five documents create, and a company that raises on SAFEs is deferring that structure, not escaping it.

Everything above takes calendar time, and calendar time is paid for in burn. Week 4 gave the arithmetic: runway equals cash divided by monthly net burn. Part Three's funnel runs roughly 3 to 6 months from first outreach to money in the bank, and the confirmatory diligence and documentation this part describes are the last 4 to 8 weeks of that window, from term sheet to wire. A raise, in other words, consumes something like 3 to 6 months of runway end to end, with the closing stretch the part a founder controls least.

Formula. Runway at Close = Runway at Launch - (Months to Term Sheet + Months from Term Sheet to Wire), assuming constant net burn

Work the numbers. A company with $1.5 million in cash and $125,000 of monthly net burn has 12 months of runway. It launches the raise, reaches a signed term sheet in month 4, and wires close at the end of month 6. Cash spent during the process: 6 months × $125,000 = $750,000. Cash at close: $750,000, which is 6 months of runway. The company signed its documents with real options still on the table, including the option to walk.

Now run the same process starting at 6 months of runway. The term sheet arrives with 2 months of cash left; the close, if nothing slips, lands at roughly zero. And the counterparty knows. The bank statements and burn rate are in the data room; the investor can compute the founder's runway to the week. Every negotiation during confirmatory diligence, every finding that might justify a price adjustment, happens against a visible countdown. This is how a raise becomes the distressed financing of Part Five, where Week 6's antidilution machinery fires and the down round reprices everyone. The discount for negotiating desperate is far larger than the dilution from raising a few months early.

Hence the practitioner rule: launch the raise with 9 to 12 months of runway. That budget covers the 3-to-6-month funnel (whose final 4 to 8 weeks are the close itself) plus a cushion for the process slipping, and it means the alternative to a bad term sheet is still "keep operating" rather than "shut down." It also composes cleanly with Week 4's sizing rule: raise 18 to 24 months of runway each round, and starting the next raise at the 9-to-12-month mark gives every round roughly a year of heads-down execution before fundraising resumes.

Suppose the bridge buys time but not a recovery, and the company must price a new round below the last one: a down round. Week 6 introduced the three antidilution forms and showed that broad-based weighted average is the market standard. Here we execute the full computation on a small cap table, end to end.

The cap table before the down round

The Series A invested $6,000,000 at $3.00 per share with broad-based weighted-average protection, and its preferred initially converts to common one-for-one at a conversion price of $3.00. The company now raises $3,000,000 at $1.50 per share, half the Series A price, issuing 2,000,000 new Series B shares.

Formula. NCP = OCP × (A + B) / (A + C), where NCP is the new conversion price, OCP is the old conversion price, A is the fully diluted shares outstanding before the new issue, B is the number of shares the new money would have bought at the old conversion price, and C is the number of new shares actually issued.

Compute each input:

1. A = 8,000,000 fully diluted shares before the round.
2. B = $3,000,000 / $3.00 = 1,000,000 shares the new money would have bought at the old price.
3. C = 2,000,000 new shares actually issued at $1.50.
4. NCP = $3.00 × (8,000,000 + 1,000,000) / (8,000,000 + 2,000,000) = $3.00 × 0.90 = $2.70.

The Series A's $6,000,000 now converts at $2.70 instead of $3.00: $6,000,000 / $2.70 = 2,222,222 common-equivalent shares, an extra 222,222 shares. Those shares are not newly issued for cash; they appear on the as-converted cap table, and each additional Series A share dilutes the founders. The post-round cap table under three cases:

Read the decomposition carefully, because it is the founders' negotiating map. The down round itself costs the founders 15.0 points of ownership (75.0% to 60.0%): that is the price of the new money, and issuing shares for cash dilutes existing holders by definition. Broad-based weighted-average protection costs a further 1.3 points (60.0% to 58.7%), because the formula scales the adjustment by how much cheap stock was actually sold relative to the whole capitalization. Full ratchet, in its place, would cost 10.0 points instead of the 1.3 (60.0% to 50.0%), because it resets the conversion price all the way to $1.50 regardless of how small the offering. This is the arithmetic behind Week 6's advice: accept broad-based weighted average as the market standard, and treat full ratchet as a term worth trading almost anything else to remove.

Down rounds also reprice the relationship among investors, and three structures govern that repricing.

Pay-to-Play and Pull-Ups

A pay-to-play provision, introduced in Week 6's term-sheet discussion, conditions preferred privileges on continued support: an existing preferred investor who does not invest its pro rata share of the down round has its preferred stock converted to common, losing its liquidation preference, antidilution rights, and protective provisions. The pull-up is the carrot paired with that stick: investors who do participate get some or all of their old preferred exchanged into the new, senior series (pulled up the preference stack) while non-participants sit below them or in common. The logic is symmetry. Antidilution and preferences are downside protection, and pay-to-play says the protection is only available to investors who fund the downside when it arrives.

Full Recapitalizations and Washouts

When the company's condition is worse than a down round can express, the new money may demand a full recapitalization: the new round is priced so low, or existing preferred is converted or repriced so severely, that prior investors and founders are washed out to a small residual stake. Because a washed-out team has little reason to stay, recaps almost always pair with a fresh management incentive pool: new option grants, commonly restoring meaningful ownership to the executives the new investors want to keep. The people most damaged by a washout are the ones who hold old common or old preferred and are no longer with the company; they hold pre-recap paper and get no refresh.

Clean Price Versus Dirty Terms

Adversity often presents a specific temptation: an investor offers to protect the headline valuation (no down round announcement) in exchange for structure. A flat valuation with a 2x senior liquidation preference, full participation, or a ratchet is routinely worse for common than an honestly lower price with a standard 1x nonparticipating preference. Week 6's waterfall shows why: structure is invisible in the press release and fully visible at exit, where the stacked preferences absorb the proceeds before common is paid. The practitioner shorthand is that a clean down round beats a dirty flat round. Price your shame once, at financing, rather than compounding it silently until the exit.

The Board's Duty to Common: the Trados Lesson

Adversity is also where board duties get tested, and the reference case is In re Trados (Delaware Chancery, 2013). Trados sold for $60 million; the preferred stockholders' preferences and a management incentive plan absorbed essentially all of it, and common received nothing. The court held that directors, including the venture investors' designees, owe their fiduciary duties to the common stockholders, not to the preferred whose funds appointed them, and it reviewed the conflicted sale under the demanding entire-fairness standard. The board ultimately escaped liability only because the court found the common's shares had no economic value even standing alone. The lesson for any board approving a bridge, a recap, or a preference-clearing sale is procedural: recognize the conflict, consider what the transaction does to common, document alternatives considered, and where possible use independent directors or a common-stockholder vote to cleanse the decision.

The instrument on the offer letter determines when the tax bill arrives and what rate applies. Three instruments cover nearly every startup offer.

The ISO trap is subtle: exercising and holding creates no regular-tax income, so it feels free, but a large spread can trigger a five- or six-figure AMT bill in a year when the shares cannot be sold to pay it. The NSO trap is blunter: the spread is ordinary income the moment you exercise, cash tax due on paper gains. RSUs remove the exercise decision entirely, which is why late-stage companies favor them, but they also remove the ability to start the capital-gains clock early: the full value is compensation income at vest, at whatever the stock is then worth.

The 90-day window forces a decision at the worst time

By statute an ISO must be exercised within roughly 90 days of leaving the company to keep ISO treatment, and most option plans simply cancel any unexercised option (ISO or NSO) 90 days after termination. So an employee who leaves faces a forced choice: write a personal check for the strike price on every vested share, plus any AMT or ordinary income tax on the spread, to buy illiquid stock that may never be sellable, or walk away and forfeit equity that may have been most of the reason for taking the job. Some companies now extend post-termination windows to 5 or 10 years; under the statute an ISO exercised more than 90 days after termination converts to NSO treatment, so the extension trades ISO tax status for optionality. Ask which policy applies before you sign.

Early exercise plus 83(b): the Week 6 tool, employee edition

Some plans allow exercising options before they vest. Exercising early (ideally at grant, when the spread is at or near zero) and filing the Section 83(b) election within its 30-day window works exactly as it did for the founders in Week 6: tax is assessed on a spread of roughly zero, all later appreciation becomes capital gain, and both the long-term holding clock and the Qualified Small Business Stock clock start immediately. The 30-day deadline is as unforgiving for employees as it is for founders. The upside case is QSBS under Section 1202: if the company qualifies, gains can be substantially or entirely excluded. Under the rules for stock acquired after July 4, 2025, the exclusion phases in at 50% after three years, 75% after four, and 100% after five, with a per-issuer cap of $15M. For an early employee at a company that works, early exercise plus 83(b) plus QSBS can be the difference between ordinary-income and near-zero tax on the same gain.

An option grant is a claim on uncertain future outcomes, so it should be valued the way Week 5 valued the company itself: define scenarios, weight them, and (this is the step candidates skip) run each scenario through the Week 6 preference stack before your shares see a dollar. Common gets paid only after the preferences, so a headline exit price is not your exit price.

Worked example

You are offered 30,000 options in a company with 12,000,000 fully diluted shares, so your stake is 30,000 / 12,000,000 = 0.25%. Investors have put in $30M of preferred with a standard 1x non-participating preference, a $30M preference stack. Weight three scenarios:

• Downside (60%): the company shuts down or sells for $25M. That is below the $30M stack, so common receives $0. Your equity: $0.
• Base (30%): a $60M acquisition. The preferred takes its $30M, leaving $30M of residual. Applying your 0.25% to the residual gives 0.25% × $30M = $75,000. (This is the conservative convention: if the preferred takes its preference rather than converting, the residual actually splits among common shares only, so your true share is somewhat higher. Precision requires the full waterfall from Week 6.)
• Upside (10%): a $300M exit. That is far above the stack, so the preferred converts to common and everyone shares pro rata: 0.25% × $300M = $750,000.

Formula. Expected equity value = Σ probability × common proceeds in that scenario × your ownership of the relevant pool.

Expected value = 0.60 × $0 + 0.30 × $75,000 + 0.10 × $750,000 = $0 + $22,500 + $75,000 = $97,500. Spread over a four-year vest, that is roughly $24,400 per year, before subtracting the exercise cost, before tax, and before the future dilution described in the first note above compounds the 0.25% down. That is the honest number to weigh against the salary you gave up to take the offer, and it is dominated by a 10% chance of the upside case: the power-law shape Week 7 taught, experienced from the employee's seat.

The four questions every candidate should ask

1. What is the fully diluted share count? Without it, the grant is a numerator with no denominator.
2. How large is the preference stack, and does any of it participate? This sets the floor an exit must clear before common is worth anything.
3. What was the last 409A common-stock price, and what did the last round's preferred sell for? The 409A valuation sets your strike; the gap between it and the preferred price (commonly the common is appraised at a substantial discount to preferred at early stages) is a rough measure of the built-in spread and of how much the preferences weigh on common.
4. What is the post-termination exercise window, and is early exercise allowed? This determines whether leaving the company means writing a check under the 90-day gun, and whether the 83(b) route is even available to you.

A company that answers all four quickly is telling you something good about itself. A company that treats them as impertinent is also telling you something.

A syndicate sits between the angel and the fund. A lead investor, often an experienced angel, sources a deal and invites backers to invest alongside them through a special purpose vehicle, an SPV: a single-purpose LLC formed to hold shares in exactly one company. Platforms such as AngelList standardized the plumbing, so an SPV can be formed, funded, and closed in weeks for a formation cost in the high four figures, typically passed to the investors.

The lead's economics

The lead usually charges no management fee but takes deal-by-deal carry, commonly in the 10 to 20% range, on that single SPV's profits. Compare this with the Week 7 fund: a GP earns 20% carry only on fund-level profits, after the whole portfolio nets out winners against losers, and lives off a 2% fee in the meantime. A syndicate lead's carry is per deal: one winner pays carry even if the lead's other syndicates went to zero. That asymmetry shapes behavior. The lead has a strong incentive to run many SPVs, since each is a free option on carry, and comparatively less incentive to be selective, since there is no fund-level netting and usually little of the lead's own capital at risk. Good leads invest meaningfully in their own deals; the amount of the lead's personal check is the single most informative number a backer can ask for.

What founders should check before taking syndicate money

• Cap table mechanics. A properly formed SPV appears as one line on the cap table, one signature block, and one vote: dozens of small checks without dozens of stockholders. Confirm this in the documents; a syndicate that puts individual backers directly on the cap table recreates the mess the SPV exists to avoid.
• Who exercises the rights. Voting, information rights, and any pro rata rights sit with the SPV, exercised by its manager. Know who that manager is and how they behave when a hard consent is needed: a down round, a Week 6-style pay-to-play, an acquisition vote.
• Can the lead actually close? Syndicate allocations are filled backer by backer, so a committed $750K can quietly become $400K if backers drop. Ask for the fill status before building the round math (Week 4) around the commitment.
• Who is behind the SPV. Founders rarely see the backer list. Ask. A competitor's executive, a sanctioned person, or a future strategic acquirer can sit invisibly inside an SPV, and some acquirers' counsel will ask during Week 3-style diligence in reverse, when the company is the one being diligenced at exit.

Corporate venture capital is a corporation investing off its own balance sheet or through a dedicated arm (Google Ventures, Intel Capital, and Salesforce Ventures are long-running examples). CVC participates in a meaningful share of venture deal value, roughly a quarter of dollars invested in recent years as a practitioner approximation, so most companies raising past the seed stage will encounter it.

Strategic versus financial mandates

The first question about any CVC is what it is paid to do. A financially mandated CVC is judged on returns, behaves much like the Week 7 fund, and is compensated on investment performance. A strategically mandated CVC is judged on what the parent learns or gains, whether market intelligence, early access to technology, or pipeline for partnerships or acquisition, and its staff are often salaried corporate employees without carry. This matters because incentives predict behavior at the board table. A financial investor wants the outcome that maximizes exit value. A strategic investor may rationally prefer the outcome that maximizes the parent's benefit, and those can diverge: a strategic may resist a sale to the parent's competitor, favor a partnership that locks the startup into the parent's ecosystem, or lose interest entirely when the parent's strategy shifts. CVC arms also turn over with corporate priorities, so the person who championed the deal may be gone in two years.

The commercial-agreement trade

CVC deals frequently arrive as a package: investment plus a commercial agreement, such as a distribution deal, a supply contract, cloud credits, or a co-development arrangement. The package can be genuinely valuable; a distribution agreement with the corporate parent can be worth more than the check. But price the two components separately. A common failure is accepting a lower valuation or heavier terms because the commercial agreement feels valuable, then discovering the agreement has weak minimums, an easy termination right, or an exclusivity clause that blocks the startup from the parent's competitors, often the majority of the market. Run the commercial agreement through the Week 8 lens on its own: what does it cost in dilution, control, and optionality, independent of the money?

Signaling and the M&A option problem

Two structural concerns follow CVC money. First, signaling: taking money from one strategic can mark the company as that corporation's territory, chilling partnerships and future acquisition interest from its rivals: the signaling logic Week 8 applied to a firm's financing choices, but now across an entire industry. Second, the M&A option: some CVCs seek a right of first refusal or right of first negotiation on a sale of the company. Sophisticated counsel resist these almost categorically, because a known ROFR deters other acquirers from bidding (why spend months on diligence and a bid if the incumbent can simply match it?), and a weak auction is the main determinant of a weak exit price. Information rights deserve similar care: a CVC whose parent competes with the startup should sit behind an information firewall, with board observer rights rather than a seat if necessary.

When CVC money is the right choice

CVC is the right capital when the strategic asset is real and priced: capital-intensive or deep-tech companies that need a partner with infrastructure; companies whose growth genuinely runs through the parent's distribution; and rounds where the corporate validates the technology to a market that trusts the parent's judgment. The guardrails are consistent: financial terms at parity with the round's other investors, no ROFR on sale (a right of first offer or simple notice is the common fallback), termination-tested commercial agreements, and an information firewall. Taken on those terms, CVC is a legitimate, sometimes the best, source of capital. Taken as a package deal negotiated once and read never, it is how startups wake up owned in every way except on the cap table.

Equity Crowdfunding: Reg CF and Reg A+

Regulation Crowdfunding (Reg CF) lets a private company sell securities to the general public, accredited or not, through a registered funding portal, up to a cap of roughly $5 million in any rolling 12-month period. The company files a Form C disclosure, and its financial statements must be reviewed or audited depending on the amount raised. Regulation A+ is the larger sibling: a Tier 2 offering can raise up to $75 million in a 12-month period, but only after the SEC qualifies an offering circular, and it carries ongoing reporting obligations afterward. Practitioners call Reg A+ a mini-IPO for a reason; it takes months and meaningful legal spend.

The mechanical problem with crowd money is the cap table. A raise that succeeds by attracting nine hundred small checks can fail the Week 3 diligence checklist two years later, when a Series A lawyer asks who, exactly, holds the company's securities. The market's answer is aggregation: a crowd SAFE or a special purpose vehicle (SPV) rolls the entire crowd into a single line on the cap table, with one signature block and one entry in the ledger.

Grants: SBIR and STTR

The Small Business Innovation Research (SBIR) and Small Business Technology Transfer (STTR) programs are federal R&D grants, awarded competitively by about a dozen agencies. Phase I awards are feasibility studies, typically in the tens to low hundreds of thousands of dollars over six to twelve months. Phase II awards fund development, typically in the high hundreds of thousands to low millions over about two years. STTR differs from SBIR in requiring a formal partnership with a nonprofit research institution.

The money is non-dilutive, which is the whole attraction: it is the only entry in this part's table with no cost in equity and no repayment claim. The price is paid elsewhere. Proposals take weeks to write and months to decide, the funded work is scoped to what the proposal promised, and the reporting and government-accounting burden is real. A grant is cheap capital for a company whose research agenda already overlaps a federal one; it is a slow detour for a company bending its roadmap to chase the award.

Accelerators

The reference point is Y Combinator's standard deal: approximately $500,000 for roughly 7% of the company, structured as a small check that buys the 7% plus a larger check on an uncapped SAFE that converts at the next round's terms. Measured purely as a price per point of equity, it is expensive money at a low implied valuation. What the founder buys is a three-month program, a network, and a demo day in front of concentrated investor attention.

Interpreting accelerator outcomes requires separating selection from treatment. Top programs admit a small fraction of applicants, so their alumni would have outperformed the average startup anyway; the raw performance gap overstates the program's value-add. Studies that compare accepted companies with similar near-miss applicants find that good accelerators do accelerate outcomes in both directions: portfolio companies raise follow-on capital sooner, and the ones that were going to fail shut down sooner. Faster resolution of uncertainty is itself worth something, per Week 5's scenario logic, but it is a different claim than the brand's marketing makes. The signaling value, meanwhile, is steeply rank-dependent: a top-tier badge opens investor doors, while an unknown program's 7% buys little that the founder could not have gotten free.

Venture Studios

A venture studio inverts the accelerator model. Instead of admitting existing companies, the studio generates the idea, assembles the founding team, and provides the first capital, and in exchange it commonly takes 30 to 50% of the equity at formation. For an operator who wants to found a company but has no idea and no capital, the trade can be rational. The cost shows up later, twice: the founder's own stake is small before any outside investor has bought a share, and downstream VCs, who use founder ownership as a proxy for founder motivation, discount teams that arrive already heavily diluted. The studio is the heaviest-equity source in the landscape, and it should be evaluated as such.

Revenue-Based Financing

Week 8 introduced revenue-based financing (RBF) among the debt instruments: capital repaid as a percentage of revenue, with no fixed repayment schedule, available once a firm has recurring revenue to share. The mechanics deserve one worked example. A company takes a $500,000 advance with a 1.5x repayment cap, paying 6% of monthly revenue until it has repaid $750,000 total (1.5 × $500,000). If revenue is modest and repayment stretches over three years, the effective annual cost lands in the high 20s in percentage terms. If revenue grows quickly and the cap is repaid in eighteen months, the effective annual rate is well above 50%. The instrument's defining quirk is that success makes it expensive: the faster the company grows, the sooner it repays, and the higher the annualized cost of the same $250,000 fee.

Formula. Total RBF repayment = advance × cap multiple. The dollar cost is fixed at signing; the effective annual rate depends on how fast revenue repays it, rising as the repayment period shortens.

Against Week 8's venture debt, the contrast is clean. Venture debt is sized off the last equity round (commonly 25 to 35% of it) and priced at 8 to 15% interest plus warrants, so it generally requires an institutional equity sponsor. RBF requires no sponsor and takes no warrants; it requires revenue instead. One is a complement to venture capital; the other is a substitute for it.

Here is the fact that reframes the whole course: the overwhelming majority of new firms do not raise institutional venture capital. Robb and Robinson (2014), the same Kauffman Firm Survey evidence Week 8 used, found that new firms are built primarily on owner equity, formal bank debt (often personally guaranteed), and business credit; venture capital finances a vanishingly small fraction of company formations. VC is a specialized instrument for a specialized business: one whose potential outcome distribution can justify a fund's power-law arithmetic from Week 7. Most good businesses are not that, and are not worse for it.

Which means the survey in this part is not an appendix to the real material; for most founders, it is the real material. The decision it frames has three inputs. The business: its cash-flow shape, its asset base, and its plausible exit paths determine which instruments will even engage, because a lender needs revenue or collateral, a grant agency needs a research agenda, and a VC needs a shot at a power-law outcome. The founder: Week 1's question about what the founder actually wants, control or wealth, a durable company or a large exit, points different people at different rows of the table. The goal: capital sized and shaped for a specific milestone, per Week 4's runway math, rather than the largest amount available.

Matching the source of capital to the business, the founder, and the goal is the entire course compressed into one decision. Every tool the course has built, the diligence checklist, the runway arithmetic, the scenario weights, the waterfall, the cost-dilution-control-signaling lens, exists to make that one match well.