What A Cofounder Agreement Should Cover That Most First-Timers Forget

Investors expect to see a signed founders' agreement before writing a check, and one of the most common reasons a term sheet stalls is a founding team that skipped the parts of the agreement that felt unnecessary at the time. Missing or vague IP assignment language alone has killed deals at the term sheet stage. Forbes lists the absence of a founders' agreement among the top ten legal mistakes startups make, and founder conflict caused by a poorly constructed one is regularly cited as among the most fatal mistakes that kill early-stage companies.

Most first-time founders don't skip the founders' agreement entirely — they write one, split the equity, define titles, and move on. The gap isn't in whether an agreement exists. It's in what it actually covers. A handful of clauses show up in almost every "what founders forgot" postmortem, and they tend to be the ones that felt unnecessary to negotiate while everyone was still getting along. This piece covers the specific provisions that are most commonly missing, why each one becomes expensive precisely when it's needed most, and how to think about drafting them before the relationship is tested.

Pre-Incorporation IP: The Clause That Kills Deals At The Term Sheet Stage

The most consequential gap in first-time cofounder agreements has nothing to do with equity or titles — it's what happens to the work built before the company legally existed. A standard IP assignment clause transfers intellectual property a founder creates during the company's operation. What it often fails to cover is work created before the effective date — the prototype built over a few weekends, the initial codebase, the designs sketched out before anyone filed paperwork.

Without an explicit retroactive assignment covering that pre-incorporation work, that code, those designs, and those inventions legally stay with the individual founder personally, not the company — even if that work became the actual product. This is exactly the kind of gap investors are trained to look for, and exactly the kind of gap that stalls a raise once diligence starts.

What a complete IP clause needs to cover:

• Assignment of everything created during the company's operation, without exception
• Retroactive assignment of anything created before incorporation that's directly related to the business
• A schedule listing specific pre-existing IP each founder is bringing in, so there's no ambiguity later about what was already assigned versus what wasn't
• A requirement that any third party — contractor, early hire, advisor — signs their own assignment agreement before touching the product

The instinct to skip this clause usually comes from the same place: the founders trust each other, so formalizing something that already feels obvious seems unnecessary. That instinct is exactly backwards. The clause exists for the version of the relationship that isn't going well, not the version that is.

The Deadlock Clause: What Happens When Two Equal Owners Actually Disagree

A 50/50 equity split feels fair at the moment of signing, and it's become the norm for day-one cofounders precisely because it does. What it doesn't come with automatically is a mechanism for what happens when the two people who each own half the company fundamentally disagree on a decision neither will concede.

This is a deadlock, and it's a specific, well-documented failure mode — distinct from ordinary disagreement precisely because no one has the tie-breaking authority to resolve it. A company in deadlock cannot approve a budget, hire a key person, accept an investment, or change direction, and the value built over years can erode while two founders who each still believe they're right wait the other out. Without a contractual resolution mechanism, a prolonged deadlock in a 50/50 company typically ends in dissolution — a court-ordered wind-down that destroys value for both sides, not a negotiated outcome either one would have chosen.

A workable deadlock clause generally includes some combination of the following, layered so each stage has a deadline before escalating to the next:

• A defined negotiation window — commonly 14 to 30 days — before any other mechanism triggers
• A mediation requirement, with a neutral third party facilitating a resolution before anything more formal
• A tie-breaking mechanism — an external advisor with a deciding vote, or a rotating final-say arrangement between the founders
• A binding arbitration clause for matters mediation doesn't resolve
• A buy-sell or "shotgun" provision as a last resort, where one founder names a price for their half and the other either buys at that price or sells at it

The specific mechanism matters less than having one defined before it's needed. A sparsely worded or entirely absent deadlock provision doesn't just fail to prevent a standoff — it actively makes the standoff worse, because neither founder has a defined process to point back to, which tends to escalate a business disagreement into a personal one.

Reserved Matters And Decision Rights: Who Actually Has The Final Call

Closely related to deadlock, and just as commonly skipped, is a clear list of which decisions require joint sign-off and which don't. A common mistake in early agreements is requiring unanimous consent on every decision by default, without articulating what happens if that consent doesn't come — which effectively builds a deadlock trigger into every routine choice the company needs to make.

A better structure separates decisions into two categories from the start:

• Reserved matters — decisions significant enough to require both founders' agreement (raising capital, issuing new equity, taking on debt above a threshold, selling the company), each paired with a deadlock-break procedure if consent isn't reached within a defined window
• Operational decisions — the day-to-day calls that fall within each founder's defined role and don't require the other's sign-off at all

Without this split, founders often default to consulting each other on everything, which slows the company down, or defaulting to consulting on nothing, which erodes trust the first time a significant decision gets made unilaterally. Neither failure mode is really about the decision itself — it's about the absence of a rule for which decisions needed to be joint in the first place.

What Happens If A Founder Leaves, Becomes Unable To Work, Or Simply Stops Contributing

Vesting is the piece of this most founders have absorbed by now — a standard four-year vesting schedule with a one-year cliff protects the company if someone leaves early. What's less consistently covered is the range of ways a founder can stop contributing without formally "leaving" in a clean, voluntary sense.

A complete agreement addresses more than the voluntary-departure case:

• Involuntary departure — what happens to unvested and vested equity if a founder is removed by the others, not just if they resign
• Death or disability — whether unvested equity accelerates, reverts to the company, or passes to an estate, and who makes that call
• Reduced contribution — what happens if a founder stays on paper but stops meaningfully contributing, without a clean resignation triggering the standard leaver provisions
• Time commitment obligations — an explicit definition of what "full-time" means for each founder, and what obligation each has around outside opportunities that could conflict with the company

These scenarios feel remote and slightly uncomfortable to negotiate at the founding stage, which is exactly why they're the ones most often left out. They're also, by definition, the scenarios where having a clear answer already on paper matters the most — because they tend to arrive without warning and under conditions where a calm, from-scratch negotiation isn't realistic.

Capital Contributions: Loan, Investment, Or Gift

When one founder puts personal money into the company early — covering incorporation costs, early tooling, a few months of runway — first-time agreements frequently treat this as an informal understanding rather than a defined term. Left undefined, this becomes one of the more common sources of later dispute, particularly if the company starts raising outside capital and the terms of that early contribution suddenly matter for cap table purposes.

A complete agreement should specify, in writing, whether that money was a loan to be repaid, an investment convertible into equity, or a gift with no expectation of return — along with any future financing plans that depend on the answer. Leaving this ambiguous doesn't just risk a future disagreement over reimbursement; it risks a future disagreement over dilution, which becomes far harder to resolve once outside investors are involved and cap table cleanliness suddenly carries real weight in a raise.

Confidentiality And Post-Departure Restrictions

The final gap shows up less at signing and more after a founder has already left. Standard agreements often stop at defining what happens to equity on departure, without addressing what a departing founder can and can't do afterward — specifically, whether they can immediately go work for or start a competing company, and whether they can solicit the remaining team or the company's clients on the way out.

A confidentiality and non-solicitation clause covering the post-departure period protects the company from exactly the scenario a founder split is most likely to create: someone who knows the product, the roadmap, and the client relationships intimately, now with every incentive to use that knowledge somewhere else. This is also one of the more legally sensitive clauses to draft well, since enforceability varies significantly by jurisdiction — which makes it a section worth getting real legal input on, rather than adapting from a generic template.

None of these clauses are exotic. Every one of them shows up, repeatedly, in postmortems of cofounder relationships that fell apart — not because the founders didn't trust each other enough to write them down, but because writing them down felt unnecessary at a moment when the relationship didn't need protecting yet. The agreement's entire purpose is to hold up in the moment it's actually needed, which is by definition the moment trust alone isn't enough.

Getting the cofounder relationship itself right in the first place makes every one of these conversations easier to have honestly and early, rather than retrofitting an agreement onto a partnership that's already strained. That's the layer CoffeeSpace focuses on — helping founders and early hires find matches worth building this kind of foundation with, so the agreement reflects real alignment rather than papering over a mismatch neither side wants to name yet. For anyone about to sit down and draft this agreement with a cofounder, or still searching for the right one to draft it with, CoffeeSpace is built to help that search start on solid ground.