Who can block a decision after you sign?
Two owners with 50% each are not equals if only one of them can approve a payment. Control is written into clauses, not into percentages.
What equal equity does not give you
A structured decision review is a written, independent analysis of one pending decision: four assessment dimensions, two or three scenario paths, and the list of what has to be evidenced before you commit. In a partnership, the first thing that analysis separates is the claim on value from the ability to decide. Equity determines what you own when the business is sold. Control determines whether it is sold at all, who is hired, what is borrowed, and whether cash can leave the company. Those two questions are answered by different documents, and signing the cap table does not settle the second one.
The relevant numbers arrive before signature. A scoping request is answered within 24 hours, a written memo is delivered in 48 hours, and one follow-up exchange is answered within 72 hours of delivery — which is why a control review started three weeks out is workable, and one started in the final three days is not.
Where blocking rights hide
Blocking rights are rarely described as blocking rights. They appear as consent thresholds in seven places:
- Board composition — who appoints directors, and whether the chair holds a casting vote.
- Reserved matters — the list of decisions that require a specific majority or unanimity.
- Shareholder consent thresholds — often 50%, 66.7% or 75%, which decides whether a minority holder has a veto.
- Bank and payment mandates — who must co-sign, and above what amount.
- Information rights — who receives what, and how quickly. The party who controls the numbers usually controls the account of how the business is doing.
- Transfer restrictions — pre-emption, tag-along and drag-along rights that decide whether you can leave at all.
- Deadlock clauses — what happens when two equal holders disagree, which is the most commonly untested clause in the whole agreement.
Deadlock: five mechanisms, ranked
When two parties with equal power disagree, the agreement has to say what happens next. There are five answers, and they are not equally good:
- Escalation to a third party. Cheap and fast, and it only works if the third party has real authority rather than a courtesy title.
- A casting vote or tie-break appointment. Decisive, and it hands control to one side permanently.
- Buy-sell (shoot-out). Effective where both sides can fund a purchase, brutal where one cannot.
- Mediation or expert determination. Slower, and it prices the dispute in time rather than money.
- Nothing at all. The most common clause in practice, and the reason ventures sit frozen for months while the business loses value.
How to check this in one afternoon: four steps
- Step 1 — List the five decisions that matter. Hiring the second employee, borrowing, related-party payments, selling the business, and changing what the business does. Everything else is routine.
- Step 2 — Trace each one through the documents. Find the clause that decides it and write down who has to agree.
- Step 3 — Test the deadlock. Ask what happens at month three if the two of you disagree and neither moves. If the answer has to be improvised, that is the finding.
- Step 4 — Convert the gaps into conditions. Anything unresolved becomes a condition of signing rather than a hope that the relationship stays good.
A control review versus a legal opinion
These are alternatives, not versions of the same product. A legal opinion tells you whether a clause is enforceable and what a court would do with it. A consultancy engagement sizes the market. A control review answers a third question: which decisions you can no longer make alone, who can stop what, and where the irreversible points sit. Most reviews end with a short list of questions for your lawyer, which is where the legal layer should begin — with the question already framed, and the metered hour spent on the answer rather than on the search.
Numbers that decide control
- 50/50 — deadlock by construction, unless a tie-break exists somewhere else in the pack.
- 51/49 — control on paper, instability in practice: the minority knows it can be diluted at the next round.
- 75% — the threshold that turns a minority stake into a veto over structural changes in many jurisdictions.
- Two-thirds — common for amending constitutional documents, so a holder above 33.4% can block a restructure.
- Any threshold with a related-party carve-out — the carve-out, not the threshold, usually decides who can pay themselves.
Where people get this wrong
- Assuming the board decides everything. Many agreements reserve specific matters to shareholders, which bypasses the board entirely.
- Reading the cap table and stopping there. The shareholding schedule is the least informative page in the pack.
- Confusing information with control. The party who controls the numbers can delay, reframe or withhold a decision without holding any formal veto.
- Testing the exit last. Exit mechanics are decided at entry; discovering them later means negotiating from weakness.
What to send your lawyer
Five questions, in this order, convert a structural review into a legal work plan: which clause creates each veto; whether the deadlock mechanism has been tested in this jurisdiction; whether the transfer restrictions permit a sale you would actually accept; what the bank mandate requires for a payment above your working threshold; and what changes if one party dies, divorces, or becomes insolvent.
What this costs, and when it is not worth paying for
A review of this kind is a fixed fee — a Written Brief at $197, a Single Decision Review at $380 (up to $580 where several parties are involved) — against a renegotiation that costs a multiple of any advisory fee. It is not worth buying where the decision is small, reversible and entirely within your control: a two-person consultancy with no external capital and a written exit is a checklist, not an engagement. We say so at scoping, and that conversation is free.
If three or more of the questions above are unresolved, the structure is not ready to sign. The shorter version is the twelve questions.
