The 30 days before you sign: five structural checks
The month before signing is where deals are won or lost — not at the term sheet, and not at closing. These five checks cover what the documents usually leave open.
1. Which claims are evidenced, and which are promises?
Market access, introductions, pipeline and revenue forecasts arrive as relationships rather than evidence. The check is simple: for each material claim, is there something in the agreement that requires it to be demonstrated, and a consequence if it is not? If not, you are carrying that risk alone.
2. Who can block a decision at month three?
Equal equity is not equal power. Identify who controls the bank mandate, the client list, hiring, and the ability to suspend operations. Then find the mechanism for disagreement. Most disputes are not personal — they are the predictable result of a structure with no resolution path.
3. In what order does the money move?
Single-tranche commitments reduce your options to persuade. Staged commitments tied to verifiable milestones keep a decision reversible for longer. The question is not how much you commit, but what you can still do after the first payment clears.
4. Where is the information asymmetry?
Ask what the other side knows that you do not, and how you would find out before signing rather than after. This is usually a diligence question, occasionally a governance question, and rarely a legal one.
5. Can you exit, and at what cost?
An exit that requires a negotiation you have not planned for is not an exit. Confirm notice periods, valuation mechanics, restrictions and what happens to your position if the relationship deteriorates.
Doing these checks in practice
Five checks, one page, before you sign. If three of them cannot be answered from the documents you already have, that is the finding — and it is cheaper to discover now than at month three.
The 12-point Decision Risk Checklist covers these five checks in more detail. If several remain unanswered and the commitment is material, that is what a decision review addresses.
Why the last thirty days matter more than the term sheet
By the time a term sheet exists, the commercial logic has usually been accepted by both sides. What remains open is structural: who verifies what, who decides what, and what happens if the relationship deteriorates. Those questions are rarely settled by the document that gets signed first, and they are the ones that reappear at month three.
What this costs to get wrong
A structure with an undefined control question tends to express itself as delay rather than catastrophe: decisions stall, capital sits, and the cost accumulates quietly in opportunity terms. The point of the five checks is to find those points while they are still cheap to fix — before signature, when both sides still want the deal.
The one-page version
If you only do one thing in the next thirty days, write down what you would need to see before you felt comfortable signing. If any item on that list is a promise rather than a verification, you have found your first condition.
What the last thirty days actually decide
A structured decision review is a written, independent analysis of one pending decision: four assessment dimensions, two or three scenario paths, and a list of what has to be evidenced before you commit. The final thirty days are the only window in which all of it can still be changed cheaply. After signature the control question moves out of the negotiation and into the agreement, where changing it costs a renegotiation.
Three numbers frame that window. A scoping request is answered within 24 hours. A written memo is delivered in 48 hours. One follow-up exchange is answered within 72 hours of delivery — which is why a review started in the last ten days is still workable, and one started in the final three usually is not.
What to verify, in order: four steps
- Step 1 — List the claims that carry the decision. Write down the three or four statements that, if false, would change your answer. Everything else is context, not evidence.
- Step 2 — Mark each claim evidenced, assumed or unverifiable. "Assumed" is the dangerous middle, and it is where most agreements are lost.
- Step 3 — Name who can verify each assumption, and by when. Your lawyer, your accountant, a registry search, the counterparty — each with a date attached to it.
- Step 4 — Convert what is left into conditions. Anything still unevidenced becomes a condition precedent rather than a hope that survives into the agreement.
This is deliberately close to the twelve-question protocol Kahneman, Lovallo and Sibony proposed for CEO-level decisions in Before You Make That Big Decision (Harvard Business Review, 2011) — the difference is that a review is produced as a written deliverable with the assumptions logged, rather than as an internal checklist.
A review versus a legal memo, a consultancy engagement or a reading
These four are alternatives, not versions of the same product, and choosing the wrong one is why advice can feel expensive and inconclusive. A legal memo interprets the documents you already have: it answers whether something is enforceable, not whether it is the decision you should make. A consultancy engagement models a market, which helps with sizing and not at all with the control question. A reading interprets a situation in prose. A structured review maps what the decision depends on — who can block it at month three, what sequencing the documents imply, where the irreversible points sit — and then hands the legal questions back to your lawyer with the list attached.
Where the last thirty days usually break
- Control is assumed rather than written. Equal equity is not equal power: the operating agreement, not the shareholding, decides who can block a distribution.
- Sequencing is reversed. A consent or reporting obligation that has to be satisfied first is treated as an administrative detail, then blocks the capital transfer.
- The exit is untested. Nobody prices the exit at the same time as the entry, so the cost of leaving is discovered later rather than negotiated now.
- Verification turns into assurance. A counterparty promise is filed as a fact, and the assumption outlives the person who made it.
Why there is no forecast in this article
Nothing above depends on predicting an outcome. The public record on forecasting is instructive: the Good Judgment Project (gjopen.com) spent years testing whether forecasting skill survives measurement, and the fact that such tournaments are necessary is itself an argument against selling one client a probability. What can be verified is the analysis — whether the assumptions were identified, whether the verification steps were listed, and whether the irreversible points were named.
Cross-border agreements: the two layers that matter
Where the counterparty sits in another jurisdiction, two public references are worth knowing before you negotiate: the UNIDROIT Principles of International Commercial Contracts and the ICC's model contract and arbitration material (iccwbo.org). Neither substitutes for local counsel; both tell you which questions are standard enough that your lawyer should have a settled answer, and which are specific to your structure.
