It does the work, then it stops and asks
Nova reads your workspace, shows every call it made, and proposes the change. Nothing is written until a person confirms it.
↳ Watch the sequenceProduct

Findings / Field research
Every firm we sat with described the same sale in the same four movements, and produced the same four documents doing it. None of them had compared notes with each other.
Software for salespeople assumes a funnel: stages a deal passes through and leaves behind. A project sale does not behave that way. The drawing made at stage three is the reason the price at stage two changed, and both are live until the day somebody installs something.
What we did was narrow. We sat with firms whose deals live in quotes, drawings and site visits, and wrote down what the sale actually looks like rather than what a CRM assumes it looks like. We were not testing a model. We were trying to find out whether the shape we kept seeing in one business was a habit of that business or a property of the work.
It was a property of the work. Four stages, and at each one a document that outlives the conversation that produced it.
Someone describes a room, a budget and a date. Nothing is specified, and the estimate is a guess held loosely. The document is a brief, and it is almost always informal: an email, a note taken during a call, a photograph of a sketch.
It is also the only artefact that records what the client originally wanted, which starts to matter once the specification has moved three times. Where it breaks: the brief is rarely attached to anything. It stays in the inbox of whoever picked up.
The first number that can be argued with. It moves when the drawing moves, and the reason it moved is worth more than the number itself.
Where it breaks: the quotation is versioned in a filename. Two people have different versions open, and the one that gets sent is whichever was most recently saved rather than most recently agreed.
Where a project sale is actually won or lost. Three revisions is unremarkable, and the third is the one nobody can find afterwards.
Where it breaks: the revision survives and the reason for it does not. Six weeks later the question is never “what does the drawing say”, it is “why did we change it”, and that answer was in a conversation nobody wrote down.
The work lands on a site, on a date, and it has to be built to the specification that was signed rather than the one that was quoted.
Where it breaks: the handover is a conversation. The people who did the selling know things the people doing the fitting do not, and the transfer happens by memory and goodwill.
Every break is the same break. The document survives. The reason does not.
A funnel stage is a place a deal has been. These four are not places; they are layers that stay live. The brief still governs at installation, because it is the only record of what was wanted before anyone drew anything. The drawing set still governs after handover, because it is what somebody has to build against.
Which is why the systems these firms had were not wrong so much as mis-shaped. A CRM stored the deal. A folder tree stored the files. A spreadsheet stored the numbers. Nothing stored the relationship between them, and that relationship is the sale.
The four stages are the pipeline model in Venuvo OS, unchanged from what we wrote down. Each stage carries its document, each document keeps the reason it changed, and delivery opens the same record sales was working in rather than receiving a summary of it.
That is a smaller claim than most software makes. It is also the whole finding: the firms did not need a better funnel, they needed the four things they already produce to stop living in four different places.
Tell us roughly how you work and we will come prepared.