The Scope Nobody Wrote Down: When a Job Was Agreed in a Corridor

Here is a constructed example. Nothing in it is drawn from a real customer, and the figures are invented so you can check the arithmetic. A six-page brochure site is quoted on 2 June as Q-118: build ₹36,000, contact form with server-side validation ₹6,000, and one round of revisions ₹4,000. The call on 2 June ends with the client saying they will check with their partner. The quote is accepted on 4 June, and the work runs for five weeks. On 13 July the invoice arrives as IN-204. Both parties remember the same call and describe it differently.

Nobody in this story did anything wrong. The studio did not inflate anything: both extra lines are priced at the rate the client had already accepted. The client did not forget anything either — a person remembers a phone call as a conversation and remembers a written quote as a list, and the conversation is the part that felt like a commitment. The failure is structural. A conversation cannot be quoted, invoiced, or produced six weeks later, so the only version of the agreement with a date on it is the one that was accepted, and it is missing everything discussed after 4 June.

Once you accept that NoxOrigin has no contract editor, the rest is straightforward. The quote is the document. It has a number, a date, a customer, lines with amounts, a status that moves from draft to accepted, and a project it belongs to. When it is accepted, the business has an agreed scope: a list of things with prices, in a form that can be read by both sides, that can be printed, exported, and looked at again in a year. That is a real thing, and most small service businesses were operating with far less than it before they had it.

What the accepted quote cannot do is absorb a conversation that happened after it was accepted. It is a snapshot with a date. Treating a snapshot as if it could absorb later changes is how the invoice ends up describing work that no document ever agreed to. The control is not a better quote template. It is a rule about what happens when a conversation changes the scope: something new and dated has to exist before the extra work is billed.

This is the mechanism, and it is deliberately plain. In NoxOrigin a scope change is a new quote raised against the same project. Not an edit to the old quote — the old quote stays accepted, with its own date and its own total, and the new one sits on top of it. The project holds both. The invoice then draws on whichever quotes belong to the work being billed. The result is that the six-week argument above becomes answerable: the client can be shown 2 June as the scope, 13 July as the scope, and the difference between them as a numbered, dated document rather than a memory contest.

It also means the ordering matters more than people expect. The second quote should be raised when the change is agreed, not at invoicing time. Raised at invoicing time it looks like a correction, and a correction on a quote that was already accepted reads as an admission that the original was wrong. Raised the day the change is discussed, it reads as what it is: a piece of work the client agreed to buy, with a price attached, sitting next to the first one.

In the constructed example the money was the easy part, because the extras were small and priced at a known rate. The expensive part of an unwritten agreement is the set of decisions nobody made. Every project has a handful of these, and they are the ones that surface at handover, at month three, or when the client asks for something that was always assumed. Naming them in advance is most of the work of preventing the argument.

A project in NoxOrigin carries the accepted quotes, the tasks, the client, and eventually the invoice and the payment. Where something was discussed but never priced — a hesitation, a preference, a "we might also want" — that belongs in a note attached to the project rather than in a quote. It is not agreement, and treating it as agreement is exactly the mistake this article is about. But it is a dated, attributable record that something was raised, which is a great deal more than a memory on a Friday.

There is one more identity question worth naming here, because it is where unwritten scope goes to hide. The same client can end up with two customer records — one from a phone number at enquiry, one from a walk-in — and then the quote, the project, and the invoice attach to two different identities that are really one person. NoxOrigin flags duplicate candidates; it does not merge them. Match rules and merge behaviour are a setup decision, a person reviews each candidate, and nothing merges automatically. The reason this matters for scope drift is that a second customer record can also carry a second set of notes, so the notes about what was agreed for half the project may be sitting in an inbox nobody will ever read. Our piece on one customer with two records goes into the seam in detail.

The client calls. They have seen the invoice for ₹68,440 and they expected ₹54,280, and they are not angry, they are confused, which is worse. The person taking the call has never seen the June call. The project has two quotes if the studio raised the second one, and one if they did not. In the first case the call takes four minutes and ends with a calendar link; in the second case the call takes an hour and ends with a discount, because there is no document that can win the argument for the studio. The business has lost the same ₹14,160 either way — as unbilled time in one branch, as a discount in the other.

That is the whole argument for writing scope down, and it does not depend on the software. The tool matters because a dated record is easier to produce than a recollection, not because the tool makes anyone honest. A studio that raises a second quote for every agreed change will have the same argument with a different client; what it will not have is the argument above, where two sincere people disagree about a corridor conversation with nothing to point at.