Cap Tables and Liquidation Preferences Explained

Introduction

A capitalization table looks like a spreadsheet of ownership percentages, and that is exactly how most people misread it. The percentages tell you who owns the company. They do not tell you who gets paid, in what order, or how much. Those questions are answered by the rights attached to each block of shares, and in a venture-backed company those rights are asymmetric by design: investors hold preferred stock with a contractual claim on exit proceeds, while founders and employees hold common stock with a residual claim on whatever is left over.

The gap between the two is where careers, fortunes and deal negotiations are won and lost. A company can sell for more money than it ever raised, generate a celebratory press release, and still return a rounding error to the people who built it. That is not a scandal or an accident. It is the arithmetic of the liquidation preference working exactly as it was drafted.

This matters well beyond venture capital. Bankers running sell-side processes for venture-backed businesses have to model the waterfall before they can tell a board what a bid is actually worth to each constituency. Private equity investors see the same structures in preferred equity and rollover arrangements. And in interviews, the ability to walk through a distribution waterfall with real numbers separates candidates who understand capital structure from candidates who have memorized definitions.

Below is the whole machine: what the cap table records, how a priced round is built, how preferences and seniority interact, and two full exit waterfalls with the arithmetic shown line by line.

At its simplest, a cap table is a ledger of every claim on the equity of a private company: who owns what, in which instrument, purchased at what price, and with what contractual rights attached. A serious cap table lists common stock by holder, each series of preferred stock with its original issue price and preference terms, the option pool (both granted and unallocated), warrants, and any convertible instruments such as SAFEs or convertible notes that have not yet converted into equity.

The single most important column is not the percentage. It is the instrument. Two people can each own 10% of the same company and have wildly different economic exposure, because one holds common stock and the other holds a senior participating preferred with a 2x preference.

Why Founders, Employees and Investors Read It Differently

Founders read the cap table as a dilution tracker. Their question is what percentage they retain after each round, and whether they still control enough of the company to matter in a vote.

Employees read it, when they are allowed to see it, as a valuation of their options. What matters to them is the strike price on their grant, the number of shares underlying it, and the size of the preference stack that sits ahead of common stock. An employee holding options struck at $1.00 in a company with $300 million of preferences ahead of them owns a lottery ticket, not an asset.

Investors read the cap table as a claims register. Their question is what they are owed before anyone else gets paid, where they sit relative to other series, and what their stake is worth if they abandon that claim and convert to common. The same underlying analysis of dilution shows up in public markets through the treasury stock method for calculating fully diluted shares, but private cap tables add the layer public companies mostly lack: a preference stack that reorders who gets paid.

Common Stock vs Preferred Stock

Common stock is the base layer. Founders hold it, employees receive options over it, and it carries a residual claim: common shareholders are paid only after every preferred claim has been satisfied. Its advantages are simplicity, a lower valuation for tax purposes (which is why option strike prices sit well below the preferred price per share), and full upside participation when the outcome is large.

Preferred stock is what investors buy in a priced round, and it is not simply common stock with a better name. A typical series of preferred carries a liquidation preference, the right to convert into common at any time, anti-dilution protection, protective provisions (vetoes over specified actions such as a sale below a threshold price or a new senior series), board representation, and often a pro rata right to invest in future rounds. Investors across venture capital, growth equity and buyout funds use versions of the same toolkit, and the preferred instrument itself has close cousins in mezzanine debt and preferred equity structures used in leveraged deals.

One feature usually ends the whole discussion: preferred stock generally converts automatically into common on a qualifying initial public offering. Navan's 2025 registration statement is a standard example, providing that each share of preferred converts on the closing of a public offering above a stated proceeds threshold, in that case $100 million of gross proceeds, as filed in its Form S-1 with the SEC. An IPO collapses the entire preference stack into a single class of common. A sale does not.

A priced round is a negotiation over three numbers that are mathematically linked: the pre-money valuation, the amount raised, and the resulting ownership. Fix any two and the third is determined.

Pre-Money, Post-Money and Price Per Share

Post-money valuation equals pre-money valuation plus the new money invested. The investor's ownership is the new money divided by the post-money valuation. Price per share is the pre-money valuation divided by the pre-round fully diluted share count.

Take a company with 7.5 million fully diluted shares raising $5 million at a $15 million pre-money valuation. The post-money valuation is $20 million. The price per share is $15 million divided by 7.5 million shares, or $2.00. The investor buys $5 million divided by $2.00, or 2.5 million new shares. Post-round the company has 10 million fully diluted shares, and the investor owns 2.5 million of them: 25%.

Notice what the denominator does. Price per share is set off the pre-round fully diluted count, which includes unissued option pool shares. Every share added to that denominator before the round lowers the price per share and therefore dilutes existing holders rather than the incoming investor. That mechanical detail is the entire basis of the next negotiation.

The Option Pool Shuffle

Investors almost always require an option pool sized as a percentage of the post-money company, and they almost always require it to be created in the pre-money. In our example, suppose the investor demands a 10% post-money pool. The pre-round fully diluted count rises from 7.5 million to 8.5 million, so the price per share drops from $2.00 to roughly $1.73, and the investor's $5 million now buys about 2.9 million shares rather than 2.5 million. Existing holders keep their 7.5 million shares, but the company now has roughly 11.5 million fully diluted shares, so those 7.5 million represent 65% rather than 75%.

Option Pool Shuffle

The practice of requiring a new employee option pool to be created out of the pre-money valuation rather than the post-money company, so that existing shareholders alone absorb the dilution while the new investor's percentage is unaffected. Because the reserved but unissued pool shares are counted in the pre-round fully diluted share count, they lower the price per share and reduce the effective pre-money valuation. A 10% post-money pool carved out of a stated $15 million pre-money produces an effective pre-money of roughly $13 million.

Compare the two outcomes. If the pool had been created after the round, diluting everyone equally, existing holders would retain 75% of 90%, or 67.5%, and the investor would hold 22.5% rather than 25%. The pre-money carve-out therefore transfers 2.5 percentage points of the company from the founders to the investor. In cash terms, existing holders end up with 65% of a $20 million post-money company, or $13 million, which means the true pre-money valuation was never $15 million at all.

The liquidation preference is the contractual right of preferred holders to receive a specified amount of exit proceeds before common shareholders receive anything. It applies on any liquidation event, which the charter typically defines to include a sale of the company or of substantially all its assets, not merely a bankruptcy.

Liquidation Preference

The amount a preferred shareholder is contractually entitled to receive from exit proceeds before common shareholders receive anything. It is expressed as a multiple of the original investment, most commonly 1x, meaning the investor gets their money back first. A liquidation preference is not a guarantee: if the sale price is below the total preference stack, preferred holders are paid in order of seniority until the money runs out, and holders further down the stack take a loss alongside common.

To see the structures side by side, take one investor who put in $20 million for 25% of a company that later sells for $120 million.

1x Non-Participating: The Market Standard

Under a 1x non-participating preference, the investor chooses: take the preference of $20 million, or convert to common and take 25% of the proceeds. Here conversion is worth 25% of $120 million, or $30 million, so the investor converts and takes $30 million. Common shareholders receive the remaining $90 million.

This is the standard for a reason. It gives the investor downside protection without taking value from common in good outcomes, and law firm surveys of US priced rounds consistently show the overwhelming majority of deals using a 1x non-participating structure. The model documents published by the National Venture Capital Association, which most US venture financings are drafted from, treat it as the default.

Participating Preferred and the Double Dip

Under 1x participating preferred, the investor takes the $20 million preference and then also shares in the remainder as though they had converted. The residual is $120 million minus $20 million, or $100 million, of which the investor takes 25%, or $25 million. The investor's total is $45 million and common receives $75 million. Compared with the non-participating case, the investor gains $15 million and common loses exactly that amount.

Participating Preferred

A class of preferred stock whose holder receives the liquidation preference and then also participates pro rata with common stock in the remaining proceeds, rather than choosing between the two. Often called a double dip, it transfers value from common shareholders to investors at every exit price. Participation is sometimes limited by a cap, typically at 2x or 3x the original investment, after which the holder is better off converting to common.

Capped Participation and Multiple Preferences

A cap limits the double dip. With participation capped at 2x, our investor cannot receive more than $40 million in total. Uncapped participation would have produced $45 million, so the cap binds and the investor takes $40 million, leaving $80 million to common. Once the exit price is high enough that straight conversion beats the cap, the investor simply converts, which is why caps matter only in the middle of the outcome range.

Multiple preferences raise the base entitlement. A 2x non-participating preference entitles the investor to $40 million or conversion value of $30 million, so they take $40 million and common receives $80 million. A 2x participating structure, the most aggressive of the set, pays $40 million plus 25% of the remaining $80 million, giving the investor $60 million and leaving common with $60 million. The same sale price produces a common outcome ranging from $90 million down to $60 million depending entirely on terms nobody mentions in the funding announcement.

Multiples largely disappeared from healthy markets, but they return whenever capital gets expensive. Down rounds, recapitalizations and insider bridges are where 2x preferences, senior stacking and pay-to-play provisions reappear, because investors pricing a risky round would rather protect the downside than argue about the headline valuation.

Waterfall math shows up in more interviews than most candidates expect: Work through liquidation preference, capital structure and valuation questions with full written answers, start practicing interview questions for free and find the gaps before an interviewer does.

Preference terms tell you how much each investor is owed. Seniority tells you in what order those claims are paid, and it only matters when there is not enough money to satisfy everyone.

Pari Passu vs Stacked

Under a pari passu structure, all series of preferred rank equally. If proceeds are insufficient, every series is paid the same fraction of its preference. Under a stacked structure, the later series is paid in full before the earlier series receives anything, so a Series D investor with a senior preference can be made whole while the Series A investor takes zero.

Standard practice for most of the last decade has been pari passu, partly because it is simpler and partly because later investors did not need to fight for priority in a market where valuations kept rising. Seniority becomes a live negotiation when the company needs money and the new investor has leverage.

Convert or Take the Preference

A non-participating preferred holder faces a binary choice at exit: take the stated preference, or convert to common and take a pro rata share. A rational holder takes whichever is larger, and the crossover between the two is the indifference point.

For a single-series company where an investor holds a $30 million preference and 20% of the fully diluted shares, conversion is worth 20% of the exit proceeds. Setting 20% of proceeds equal to $30 million gives an indifference point of $150 million, which is simply the post-money valuation of that round. Below $150 million the investor takes the preference; above it, they convert.

With multiple series the calculation becomes iterative rather than a single formula, because one series converting changes the residual available to everyone else and can flip another series' decision. Practitioners solve it by testing the outcomes rather than by algebra, running each combination of conversion decisions and checking that no holder would prefer to switch. That is exactly what a waterfall model automates.

Anti-dilution protection adjusts the price at which preferred stock converts into common when the company later issues shares at a lower price. It does not stop dilution from new capital. It changes how much of that dilution the preferred investor bears versus how much lands on common.

Full Ratchet

Full ratchet is the punitive version: the conversion price of the earlier preferred resets all the way down to the new issue price, regardless of how few shares are sold at that price. Selling a single share at $1.00 would reset a $2.00 conversion price to $1.00 and double the earlier investor's share count.

Take an investor who put in $12 million at $2.00 per share, receiving 6 million shares. The company later raises $10 million at $1.00 per share, issuing 10 million new shares. Under full ratchet, the earlier investor's conversion price becomes $1.00 and their $12 million now converts into 12 million shares instead of 6 million.

Broad-Based Weighted Average, With the Arithmetic

The market standard is broad-based weighted average, which adjusts the conversion price in proportion to how much cheap stock was actually issued relative to the size of the company:

Reading the terms in order: CP1 is the old conversion price, A is the fully diluted shares outstanding before the new issuance, B is the number of shares the new money would have bought at the old conversion price, and C is the number of shares actually issued. Broad-based means A includes options and as-converted preferred, which makes the denominator larger and the adjustment gentler; narrow-based definitions exclude the option pool and bite harder.

Run the same numbers. Before the down round the company has 10 million fully diluted shares (4 million common, including the option pool, and 6 million preferred), so A is 10 million. The new money is $10 million, which at the old $2.00 price would have bought 5 million shares, so B is 5 million. The round actually issues 10 million shares, so C is 10 million. The adjusted conversion price is $2.00 multiplied by 15 million over 20 million, or $1.50. The earlier investor's $12 million now converts into 8 million shares rather than 6 million.

Anti-Dilution Protection

A provision in preferred stock that lowers the price at which those shares convert into common when the company issues new stock below the original purchase price, increasing the investor's share count to compensate. Broad-based weighted average, the market standard and the default in the NVCA model charter, scales the adjustment to the size of the discounted issuance. Full ratchet, the aggressive alternative, resets the conversion price to the new price regardless of how few shares are issued.

The difference lands on common. With no anti-dilution protection at all, the post-round company would have 20 million shares and common would hold 4 million of them, or 20%. Under broad-based weighted average, total shares become 22 million and common falls to 18.2%. Under full ratchet, total shares become 26 million and common falls to 15.4%. The protected investor is made whole, and everyone without protection pays for it: founders and employees absorb the largest share, and the incoming investor's own stake shrinks as well, which is why sophisticated new money prices the ratchet into what it is willing to pay in the first place.

Pay-to-play provisions cut the other way. They condition anti-dilution protection (and sometimes the preference itself) on the existing investor writing a check in the new round, converting non-participants into common. In practice they are how a syndicate forces its own members to support a struggling company, and they are one of the few structural terms that can improve the common shareholders' position.

A waterfall applies the cap table's rights in order until the proceeds are exhausted. The sequence is always the same.

Take a company with 25 million fully diluted shares: 10 million common (8 million founders, 2 million option pool), and three series of 1x non-participating preferred that rank pari passu. Series A invested $5 million at $1.00 for 5 million shares. Series B invested $15 million at $3.00 for 5 million shares. Series C invested $30 million at $6.00 for 5 million shares. Total raised is $50 million, the Series C post-money valuation was $150 million, and common holds 40% of the company.

Waterfall One: A $400 Million Exit

The company sells for $400 million with clean terms and no debt.

Test each series. Series A can take $5 million or convert for 20% of $400 million, which is $80 million, so it converts. Series B can take $15 million or convert for $80 million, so it converts. Series C can take $30 million or convert for $80 million, so it converts. With every series converted, the preference stack disappears entirely and proceeds are split purely by share count.

Now change one term. Suppose Series C had negotiated 1x participating preferred instead. Series C takes its $30 million off the top, leaving $370 million, then participates for 20% of that, or $74 million, for a total of $104 million. Series A and Series B each receive 20% of $370 million, or $74 million. Common receives 40% of $370 million, or $148 million. One clause moved $12 million away from founders and employees and $6 million each away from the two earlier investors, in a deal everyone would describe as a success. Had that participation been capped at 2x, Series C would have been limited to $60 million, which is worse than the $80 million it gets by converting, so the cap would have returned the outcome to the clean case.

Sharpen the technical foundations behind the waterfall: Our 160-page PDF covers capital structure, valuation and the accounting that sits underneath deal math, access the IB Interview Guide and work through the frameworks interviewers actually test.

Waterfall Two: A $120 Million Exit That Reads Like a Win

Same company, different history. Growth stalls after the Series C, and the company needs money. Rather than accept a down round, the board agrees to keep the $6.00 headline price and give the new investor structure instead. Series D invests $30 million at $6.00 per share for 5 million shares, with a 2x participating preference that is senior to every other series.

The company now has 30 million fully diluted shares and a headline post-money valuation of $180 million. Common holds 10 million shares, or 33.3%. The preference stack is $60 million for Series D (2x on $30 million) plus $50 million across Series A, B and C, for a total of $110 million against $80 million of capital raised.

Two years later the company sells for $120 million. The press release notes that a company that raised $80 million sold for $120 million. Here is what actually happens.

Common holds a third of the company and receives 5.6% of the proceeds. Employees holding options struck anywhere near the preferred price receive nothing at all, because their strike price exceeds the per-share value of common.

The two waterfalls above make a point that repeats across the private markets: a valuation headline is a function of the price per share paid for the most senior, most protected instrument in the capital structure. It is not a statement about what the whole company is worth, and it is certainly not a statement about what common stock is worth.

What the Research Says About Post-Money Valuations

Academic work has put numbers on the gap. Research by Will Gornall and Ilya Strebulaev, published in the Journal of Financial Economics and summarized by Stanford Graduate School of Business, valued 135 US unicorns using a model that prices each share class separately and found reported post-money valuations averaged 48% above fair value, with common shares overvalued by 56%. Roughly half the companies studied lost unicorn status once the structure was priced properly. The mechanism is precisely what we have been calculating: post-money valuation multiplies the newest, most protected share price by every share outstanding, including common shares that carry none of those protections.

Clean $80 Million Beats Structured $100 Million

Founders face this trade directly. Suppose a founder owning 100% of a business receives two term sheets, each investing $20 million. Offer one is a $100 million post-money valuation for 20%, with a 2x participating preference. Offer two is an $80 million post-money valuation for 25%, clean 1x non-participating.

If the company later sells for $150 million, offer one pays the investor $40 million of preference plus 20% of the remaining $110 million, or $22 million, for $62 million total, leaving the founder $88 million. Offer two pays the investor the greater of $20 million or 25% of $150 million, so they convert and take $37.5 million, leaving the founder $112.5 million. The lower headline valuation is worth $24.5 million more to the founder.

Now test a weak outcome of $60 million. Offer one pays $40 million plus 20% of $20 million, or $44 million, leaving the founder $16 million. Offer two pays the $20 million preference, since 25% of $60 million is only $15 million, leaving the founder $40 million. The clean, lower-headline round wins in both directions. Valuing a company where the equity story is entirely about future outcomes is hard enough on its own, as any attempt at valuing a company with no profits shows, and structure makes the headline number even less informative.

For a banker running a sell-side process for a venture-backed company, the cap table is not background reading. It determines who has to approve the deal, what price clears, and what the negotiation will actually be about.

Pricing a Venture-Backed Sell-Side Process

The first thing to establish is the total preference stack, because that is the effective floor for any transaction that management and the board will support. A bid below the stack means common receives nothing, which means the founders and management team running the process have no economic interest in completing it. Boards in that position often negotiate a management carve-out, a slice of proceeds set aside for key employees ahead of the preferred, precisely to keep the people who must deliver the company through diligence engaged.

The second thing is the approval map. Protective provisions typically give one or more series a veto over a sale below a specified price, and a senior series can block a deal that pays it in full simply because the outcome is unattractive relative to waiting. Knowing which holders can say no, and what each of them receives at each price point, is what allows a banker to tell a board whether a $200 million bid is genuinely executable.

Building the Waterfall Into the Model

Mechanically, the waterfall is a separate schedule that takes an enterprise value input, bridges to net proceeds to shareholders, and applies the rights in order. It must handle conversion decisions dynamically rather than with hardcoded assumptions, since the optimal decision for each series changes with price. It must handle option and warrant strike proceeds, unvested equity treatment and any escrow or earnout that delays part of the consideration.

The same discipline transfers directly to buyout work, where sponsors model preferred returns, rollover equity and management incentive plans through equivalent waterfalls. It is also one of the more transferable skills for anyone considering a move from investment banking into venture capital, where reading structured term sheets quickly is a daily requirement rather than an occasional deal task.

Structure is the part of a deal that survives the announcement. Valuations get revised, growth rates disappoint, and markets reprice, but the charter keeps paying out in the order it was drafted. That is why the ability to build a waterfall from a cap table is a genuinely useful skill rather than an academic exercise: it converts a set of legal provisions into the only number anyone actually cares about, which is what each party takes home. Learn to read the instrument column first, model the conversion decisions honestly, and you will understand a private company's capital structure better than most people sitting around the table.