Field Notes: I Read 40 Seed Term Sheets and Found Who Actually Pays for the Legal Fees

Archi4 min read

startupsfundinglegalfounders

Contents
  1. What I Actually Read
  2. The Founder Who Paid to Be Diligenced
  3. What the Diligence Bucket Holds
  4. The Option Pool Math Nobody Runs Twice
  5. Pro Rata Rights and the Small Check
  6. Board Seats and the Quiet Majority
  7. The Clauses I Would Push Back On First
  8. What Surprised Me

What I Actually Read#

Over the past five weeks I went through 40 seed-stage term sheets shared by founders, associates, and one very tired lawyer in exchange for anonymity. The sample skews toward software and hardware startups raising between $500,000 and $4 million, mostly in the US and UK. I read every clause, then called 11 founders to ask what each line cost them in practice. This is not a survey. It is a set of notes from the paperwork, and from the people who signed it.

The pattern that kept surfacing was simple: the terms that hurt most are rarely the ones founders negotiate hardest. Valuation gets the dinner conversation. The clauses below get the invoice.

The Founder Who Paid to Be Diligenced#

Nine of the 40 sheets put legal fees on the company, usually capped between $15,000 and $35,000, payable whether or not the round closes. Two had no cap at all, just "reasonable fees." One founder I spoke to, building a three-person climate analytics tool, spent $22,400 on legal work for a $750,000 round that died at the final partner meeting. The investor's own diligence cost, roughly $9,000 of that total, came out of a bank account with 14 months of runway in it.

Who is affected most here is not the founder raising a $10 million Series A. It is the pre-seed company with one employee and no finance hire, where a $20,000 legal bill is a month of engineering. In my sample, every sheet with an uncapped fee clause went to a company raising under $1.5 million.

What the Diligence Bucket Holds#

The fee line is only the visible part. I counted the documents each investor requested and grouped them. The heavy hitters: , Cap table history, including every SAFE and convertible note with its own side letter., IP assignments for every contractor, going back to the first commit., Employment agreements with non-compete and invention assignment language that two founders discovered was never signed., Customer contracts with change-of-control clauses, which mattered most to the hardware companies.

One founder had 23 separate SAFEs from a friends-and-family round. Her cap table model took a contractor nine days to reconstruct. The investor asked for it in three.

The Option Pool Math Nobody Runs Twice#

Twenty-six sheets included a pre-money option pool, ranging from 8% to 18% of fully diluted equity. The pool is carved out before the investment, which means the founder absorbs the dilution, not the new investor. A 15% pool on a $3 million pre-money valuation moves roughly $450,000 of value from founders to future hires, and every founder I asked had modeled it only after signing.

The people hit hardest are second-time founders with existing employees. First-timers often have no one to grant options to yet, so the pool sits unused while the dilution is permanent. One founder with six employees told me he discovered that his pool was sized for a team of 30 and that three of his current engineers had been promised grants that no longer fit inside it.

Pro Rata Rights and the Small Check#

I flagged 17 sheets with pro rata rights attached to investors writing checks under $100,000. On paper this looks generous. In practice it means a small angel can block or complicate a later round by choosing to exercise, or by negotiating the right away. Two founders described weeks of email with micro-investors before their Series A. The dollar amount was small. The calendar cost was not.

Board Seats and the Quiet Majority#

Eleven sheets gave a single institutional investor board control with a two-person board or a tie-breaking clause. Another six created a board observer seat with information rights that survive the round. Observers sound harmless until a founder has to explain why four people get monthly financials and one of them is a competitor's advisor.

Who is affected most: companies with a solo founder and no independent director. In my sample every solo founder who accepted a two-person board later described the same problem, which is that every disagreement is a stalemate and every stalemate resolves in the investor's favor by default.

The Clauses I Would Push Back On First#

If I were advising a first-time founder tomorrow, the order of negotiation would not start with valuation. It would start here:

  1. Fee caps, with a walk-away. Cap legal fees at a number you can pay from the operating account and split them if the deal dies.
  2. Pool sizing tied to a hiring plan. Ask for the pool to match the next 12 months of hires, not the next four years.
  3. Information rights with a floor. Monthly reporting to a board observer should not include anything you would not send a customer.
  4. Board composition for solo founders. Three seats, or a named independent, before you sign.

None of this is exotic. It is the boring middle of the document, and it is where the money actually moves.

What Surprised Me#

Three sheets had a clause requiring the company to reimburse the investor's travel to board meetings, uncapped. One founder flew an investor to four cities in a year. The total was under $12,000, but it came out of a budget with no line item for it.

The other surprise was consistency. The terms that hurt small companies showed up again and again in the same shapes, from different investors, in different countries. That is not a negotiation problem. It is a default, and defaults are what you have to read for.

More articles