What happens instead is that a two-week diligence becomes a six-week diligence, the questions get more basic rather than less, and somewhere in week four the conversation stops being about what the business could be worth and starts being about what might be wrong with it.

Nothing has failed. Something has simply taken long enough for doubt to form.

What delay actually costs

Momentum is the asset you have least of and value least accurately. An investor who was enthusiastic in the first meeting is not enthusiastic six weeks later, not because anything changed, but because enthusiasm has a half-life and yours has been spent answering questions about how the revenue figure was arrived at.

Delay also moves who is negotiating from strength. A term sheet negotiated by a founder with four months of runway is a different document from the same term sheet negotiated with nine. Every week spent reconstructing information you should already have had is a week of runway converted into somebody else's negotiating position.

The state most businesses are in

Not bad. Ordinary.

The accounts are filed and broadly correct. The model lives in a spreadsheet the founder built and has been extended eleven times. Revenue in the model does not tie to revenue in the accounts, for reasons that are explicable but have never been written down. Customer numbers come from the CRM and disagree with the invoicing. The cap table is a spreadsheet with three tabs, two of which are versions.

None of that is negligence. It is what accumulates when a business grows faster than the way it keeps its records, and it is invisible until somebody outside asks.

What has to be true before you go out

One revenue number, and one explanation of it. Not a number in the model, another in the board pack, and a third in the accounts. If they differ for good reasons, the reconciliation exists in writing before anybody asks for it.

A model somebody else can interrogate. Assumptions separated from calculations, drivers you can point to, and the ability to change one input in front of an investor without the file breaking. A model only its author can operate is treated as an opinion.

Unit economics you can defend to a second decimal. What it costs to win a customer, what they are worth, how long they stay, and how each of those has moved over eight quarters. Not because the investor needs the precision, but because being unable to produce it says you have been running on impression.

A clean cap table. Every instrument, every option, every SAFE or convertible, fully diluted, agreed with what is filed. This is the single most common source of a bad week late in a process.

The paper findable. Customer contracts, the shareholders' agreement, employment terms, any facility, any lease. Findable in a day, not assembled over a fortnight.

A story that reconciles. The narrative in the deck has to be visible in the accounts. If the deck says the enterprise segment is the future, the accounts should show enterprise growing. Where they do not, say so first.

The thing nobody prepares for

The questions do not stop at the file. They test whether the person answering understands their own business.

An investor asking why gross margin moved two points is rarely interested in the two points. They are finding out whether you knew, whether you know why, and whether the answer arrives with confidence or with a promise to come back. That is a judgement about management, formed in the first twenty minutes, and it is not recoverable by sending a better spreadsheet on Friday.

When to start

Earlier than feels necessary, and long before the deck.

Most of the work is not analysis. It is deciding what the numbers mean, writing it down, and making the various places they appear agree with each other. That is unglamorous, it takes a few months rather than a few weeks, and it cannot be done in parallel with running a process, because running a process is a full-time job for the person who would otherwise be doing it.

A business that goes into a raise with all of the above is not more likely to be funded. It is likely to be funded faster, on better terms, by an investor whose impression of management was formed by how quickly the awkward questions got good answers.