Read the Term Sheet Backwards: The Clauses That Matter More Than the Valuation
The valuation is the one number everyone asks about, so it gets the negotiation. But a term sheet allocates three things — money, control and downside — and the price speaks only to the first. The clauses near the back decide what founders actually own, who decides, and what happens in a bad year, and they are cheap to change only in the week the document arrives.
Founders negotiate the valuation and accept the rest. It is understandable: the valuation is the one number everyone will ask about, it fits in a sentence, and it feels like the score. But a term sheet is not a price with some boilerplate attached. It is a short document that allocates three things — money, control and downside — and the valuation speaks to only the first. I have sat on both sides of these documents, as a founder and as an angel, and the ones that went badly rarely went badly because of the headline number.
What follows is a way of reading a term sheet that I give founders when one lands in their inbox: start at the back. The clauses near the end — preferences, the pool, the board, protective provisions, exclusivity — are the ones that decide what you actually own, who decides, and what happens when the plan does not work. The price tells you how good the news is on the day. The back pages tell you how the relationship behaves in a bad year. This is a practitioner's view, not legal advice; the document you are signing deserves a lawyer who does this every week.
Liquidation preference: the clause that only matters when it matters
A liquidation preference decides who is paid first, and how much, when the company is sold or wound up. The standard, founder-reasonable version is 1x non-participating: the investor chooses between getting their money back or converting to common and taking their percentage, whichever is larger. It protects the investor in a small outcome and costs the founders nothing in a large one.
The variants are where the money moves. Take an investor who puts in $5 million for 20% of a company at a $25 million post-money valuation, and consider two sale prices. At $30 million, the non-participating investor converts and receives 20%, or $6 million. A 1x participating investor takes the $5 million first and then 20% of the remaining $25 million as well — $10 million in total, a third of the proceeds from a 20% stake. At $15 million, the non-participating investor takes the $5 million preference, while the participating one takes $5 million plus 20% of the remaining $10 million, or $7 million. Same company, same price, same ownership; the founders and employees split $20 million in one world and $15 million in the other.
None of this shows up in the valuation. Participation, multiples above 1x, and seniority over earlier investors are all ways of paying a higher headline price in exchange for a better position in the bad outcomes. Sometimes that is a fair trade, particularly in a hard market. It should be a trade you made on purpose, with the numbers run at three or four exit values, not one you discovered when the first acquirer called.
The option pool: a discount on your valuation
Almost every term sheet asks for an unallocated option pool of a stated size — ten percent of the post-money company is common — and almost every one specifies that the pool is created before the new money, so it sits inside the pre-money valuation. This is the option pool shuffle, and it is not hidden or unfair; it is simply a price adjustment that is easy to miss.
Suppose the sheet says $20 million pre-money, a $5 million investment, and a 10% post-money pool that does not yet exist. The post-money valuation is $25 million, so the pool is worth $2.5 million, and all of it comes out of the existing holders. The pre-money that the founders and earlier investors are really being valued at is $17.5 million — about 12% below the number in the first line. The right response is not to refuse a pool; a company that cannot grant options cannot hire. It is to size the pool from the hiring plan — the roles you will fill over the next eighteen months and the grants they will need — rather than from a round number, and to negotiate the effective pre-money, not the headline.
The headline valuation is what the investor is willing to say out loud. The effective valuation, after the pool and the preference, is what they are willing to pay.
Control: the board and the veto list
At seed, the board is usually the founders and, at most, one investor. At the Series A it often becomes the real power centre of the company, and the composition is written into the term sheet in one line that is easy to skim. A five-person board with two founders, two investors and one independent is a very different company from a three-person board with one founder and two investors, even if the ownership numbers are identical. Ask who chooses the independent seat, and whether it needs the common holders' consent. Ask what happens to the founder seats if a founder leaves; a board seat that disappears with employment changes the balance in exactly the situation where the balance matters most.
Protective provisions are the second half of control. These are decisions — selling the company, issuing senior stock, changing the charter, taking on debt above a threshold, changing the size of the board — on which the preferred holders have a veto regardless of the board vote. A short list tied to the investor's actual economic exposure is normal. A long list that reaches into the annual budget, hiring above a salary, or ordinary-course contracts turns an investor into a co-manager. The test I use is whether each item protects the investor's money or merely gives them a say. The first is a reasonable ask. The second is something to trade, and early is the time to trade it.
Anti-dilution, pro rata and the quieter rights
Anti-dilution protection adjusts an investor's conversion price if the company later sells shares at a lower price. The broad-based weighted-average version is the market standard and is mild: it softens a down round without wrecking it. A full ratchet resets the investor's price to the new low one regardless of how many shares are sold, and the cost lands almost entirely on the founders. If you see the second in a term sheet, treat it as a statement about how the investor expects the company to go, and ask why.
Pro rata rights — the right to buy a share of future rounds in proportion to current ownership — are generally a fair thing to grant, and the difficulty is in the accumulation. Each seed investor who holds one expects to be offered an allocation at the Series A, and a lead who wants to put in a large cheque may find a quarter of the round already spoken for. Count the rights outstanding, SAFE side letters included, before you promise a clean allocation to anyone new.
Information rights are the opposite case: cheap to grant and valuable to both sides. Monthly or quarterly financials and an annual budget are reasonable to promise, and a company that produces them anyway loses nothing. It is worth being precise about what is promised, in what form and by when, because a vague information right written into a legal document becomes a standing obligation the first time a month is late.
Exclusivity: the clause with a clock
Most of a term sheet is non-binding. The exclusivity clause usually is not. A no-shop of thirty to sixty days means you may not talk to other investors while the lead completes diligence, and in that window your leverage drops to roughly zero: the investor can reprice, add conditions or simply slow down, and your alternative is to restart a process you stopped. Founders sign it because the term sheet felt like the finish line, and it is not. Diligence is where a deal changes.
Keep the window short, tie it to specific milestones where you can, and make sure your own side of the work is ready before you sign — the cap table reconciled, the financials tied to the bank, the data room organised, which is covered in what investors actually check in a data room. The faster the investor can finish diligence, the less the clock costs you. If you have a second investor warm, tell them before you sign, not after.
How to compare two term sheets
When a founder has two offers, the instinct is to compare valuations and ownership. The better comparison is a small model: for each term sheet, compute founder ownership after the round including the pool, and then proceeds to common at four or five sale prices — a bad outcome, a modest one, a good one, a very good one. Add the investor's rights as a column rather than a footnote. It is an afternoon of work, it is done in a spreadsheet, and it frequently shows that the offer with the lower price is the better one for the founders in the outcomes they are most likely to see. It is also the work the lead investor has already done on their side.
It also pays to look at the investor as well as the paper. Ask for two founders they backed who had a bad year, and call both. How an investor behaves when a plan fails is not something the term sheet will tell you, and it matters more than the difference between a $20 million and a $22 million pre-money.
What is the effective pre-money after the option pool, and was the pool sized from a hiring plan? What does each of the four outcomes — a modest sale, a good one, a very good one and a wind-down — pay the common holders under the proposed preference? Who controls the board after the round, and what happens to a founder's seat if they leave? Which decisions can the investor veto, and does each one protect their money or only give them a say? And how long is the exclusivity period, and are you ready for diligence before it starts?
The point of reading it backwards
Nothing here is a reason to distrust investors; most of them want the company to succeed and write standard terms. The point is that a term sheet is a rare moment when the relationship is still cheap to adjust. Every clause above is negotiable in the week you receive it and close to immovable a year later. A founder who has spent an hour on the preference table, the pool and the board will negotiate from a position the investor recognises. A founder who has only discussed the valuation will have agreed to most of the deal without knowing it.
If you have a term sheet in hand, or expect one this quarter, I am glad to run the comparison with you and to model your cap table through the round before anything is signed.