River
Y CombinatorBacked by Y Combinator
FREE TEMPLATE

Founder Equity Split Agreement Template

Three documents and three sheets, including the model that shows what each weighting pays each founder and what a leaver keeps.

Free download  ·  No account needed

A percentage is not arguable. It is one number with no visible derivation, so the only two responses available are acceptance and rejection, and both feel personal. Every founder equity conversation that goes badly goes badly at exactly that point. A weighting is arguable: whether pre-incorporation work should count for twice as much as capital is a question three reasonable people can disagree about out loud and then write down. So nothing here produces a percentage until a frame exists.

Tessellate Labs is an invented three-founder company with a 7,200,000 share founder pool. Its founders scored five dimensions, then weighted them privately. Run the same scores through all three weightings and one founder lands at 17.77, 19.62 or 21.81 percent, a band four points wide. Equal shares sits at 33.33 percent, outside all three by more than eleven points, which makes it a choice worth 990,000 shares rather than a default nobody priced.

The other half is the one everyone omits, and the split decides it. A founder resigning at month 14 of a 48 month schedule keeps 411,250 shares under the weighting they adopted, which is 6.63 percent of what remains, or 700,000 shares under equal shares, which is 12.73 percent. Same person, same day, same terms. With no vesting papered at all she keeps a third of the company and does no further work, which is how a team ends up beside a permanent absentee shareholder.

One set of scores, three weightings, three splits

The scoring frame, the band each founder lands in, and what the same departure produces under two different splits.

Contribution Analysis

Tessellate Labs, Inc. Scores agreed jointly with evidence on every cell. Weights set by each founder privately, before seeing the others.

DimensionProperty ofAdaRustamLenaWeight AWeight BWeight C
Full-time months before the split is setA month10104301520
Capital contributedA contribution361151540
Work done before incorporationA contribution973202015
Role criticality over 24 monthsA role9107254515
Salary given upA contribution87410510
Weights must total100100100
Proposed and rejectedWhy
CommitmentA property of a person, not a role or a contribution. Reads as a verdict on character
Who had the ideaCollapsed into work done before incorporation, which is measurable
Network and fundraising abilityNobody could name evidence for it that was not circular

Eight dimensions were proposed and five kept. The filter is one test: a dimension has to be phrasable as a property of a role, a contribution or a month, never of a person. Role criticality passes and commitment does not, because the second one lands as a judgement and the conversation stops there. Nobody, as it happens, scored their own role highest.

Split Scenarios

The same scores through every weighting anybody in the room proposed. The output is a band, not a point.

ScenarioAdaRustamLenaTotal
Equal shares, the default33.33%33.33%33.33%100.00%
Weighting A, as agreed39.71%40.67%19.62%100.00%
Weighting B, proposed by Rustam38.05%40.14%21.81%100.00%
Weighting C, proposed by Lena38.40%43.84%17.77%100.00%
Band width in points1.663.704.04
Adopted split against equal sharesAdoptedEqualDifferencePoints
Ada Okonjo2,860,0002,400,000460,0006.39
Rustam Beisenov2,930,0002,400,000530,0007.36
Lena Havelka1,410,0002,400,000-990,000-13.75
Differences sum to7,200,0007,200,00000.00

Weighting C was proposed by the founder it pays least, which is worth noticing and usually shortens the conversation. Lena's band is 4.04 points wide across every frame in the room, so there is very little left to argue about between the weightings. The argument that remains is about equal shares, and it is worth exactly 990,000 shares. That is a choice three people can make knowingly rather than a default nobody examined.

Vesting Schedule

48 months, a 12 month cliff, monthly after that. The same departure, modelled under both splits.

MonthAda, of 2,860,000Rustam, of 2,930,000Lena, of 1,410,000Vested
110000.00%
12715,000732,500352,50025.00%
14834,166854,583411,25029.17%
241,430,0001,465,000705,00050.00%
482,860,0002,930,0001,410,000100.00%
Lena resigns at month 14Under the adopted splitUnder equal sharesWith no vesting at all
Shares she held1,410,0002,400,0002,400,000
Vested and retained411,250700,0002,400,000
Repurchased at cost998,7501,700,0000
Founder shares outstanding after6,201,2505,500,0007,200,000
Her permanent stake6.63%12.73%33.33%

Same person, same day, same vesting terms, and the permanent stake nearly doubles between the first two columns: 288,750 extra shares, decided months earlier in a conversation about percentages. The third column is the real default, where nobody papers a repurchase right and a founder leaves in year two holding a third of the company with no further obligation and no mechanism available to change it. Vested is vested in four of the five departure cases, which is why the definition of cause is the one clause that has to be enumerated rather than described.

What's in the pack

01

Contribution Analysis

The scoring frame, with one filter on what may become a dimension: it has to describe a role, a contribution or a month, never a person. Evidence required in every cell, and rejected dimensions kept with their reason.

02

Split Scenarios

The same scores through every weighting anybody proposed, reported as a band per founder rather than a single answer, with equal shares priced in shares and points instead of argued against.

03

Vesting Schedule

Vested shares by month, five departure cases, and the same resignation modelled under two splits and under no vesting at all. Acceleration worked at a sample month, with the parts summing back to the grant.

04

Founder Agreement

A working draft for counsel with visible brackets wherever nothing has been decided, including the clause that records how the split was reached and why equal shares was not adopted.

05

Vesting Note

What founder vesting actually is, which is the company losing a repurchase right rather than shares arriving, plus the two dated facts the paperwork turns on.

06

Role and Decision Rights

Fifteen decisions across three tiers on one page, the consult-then-decide tier nobody writes, and a deadlock mechanism with a name in it rather than a promise of good faith.

How to use it

  1. 1

    Open in River, or download it

    Take the blank Word and CSV files with no account, or install the pack in River and have the agent hold each founder's weighting until all of them are in.

  2. 2

    Agree the dimensions before any number

    Run each proposal through one test: does it describe a role, a contribution or a month, or does it describe a person. Keep the rejected ones with their reason.

  3. 3

    Weight separately, then reveal at once

    Each founder distributes 100 points alone, on paper. The first number spoken out loud anchors everything after it, so this step does not happen in the room.

  4. 4

    Settle the leaver terms in the same sitting

    Never split the two apart. The same resignation produces very different permanent stakes depending on a split agreed weeks earlier.

Frequently asked questions

Is this template free?

Yes. Three documents and three sheets download as Word and CSV files with no signup and no credit card. Edit with AI is the optional path where the agent runs your own scores through your own weightings. The rest of the library is at the template index.

What format are the downloaded files?

Word documents (.docx) for the Founder Agreement, the Vesting Note and the Role and Decision Rights table, and CSV (.csv) for Contribution Analysis, Split Scenarios and the Vesting Schedule. They open natively in Word, Pages, Google Docs, Excel, Numbers and Sheets.

Does this say equal shares are wrong?

No, and it deliberately takes no position. Equal shares are often right. What the pack does is price the choice: in the worked example the difference between equal shares and the weighting the founders adopted is 990,000 shares for one person, which is a decision they can then make knowingly.

Why weight the dimensions privately?

Because the first number spoken out loud anchors every number after it, and an anchored conversation cannot be un-anchored. Weighting alone on paper also surfaces the useful case where a founder's own weighting pays them less than a co-founder's does.

What happens if a founder leaves?

They keep what has vested, permanently, in four of the five departure cases. At month 14 of a 48 month schedule that is 411,250 shares under one split and 700,000 under another. With no repurchase right papered at all, they keep their entire stake and owe nothing further.

Can the agreement give a founder a veto, and do we need a lawyer?

Among the founders, yes. Over the company, no: Delaware vests the business and affairs of a corporation in a board of directors. Counsel is still needed, and the pack shortens that meeting. One example: a re-papered grant is a new transfer, so a section 83(b) election runs 30 days from it.

What comes after the split is settled?

The founder rows land in a reconciled cap table, the vesting terms and consents become numbered items in the startup data room checklist, and both get asked about again by name on the diligence request list.

Move the argument off the percentage

Download the blank pack as Word and CSV files, or open it in River and have it hold each founder's weighting until every one of them is in.

Edit with AI