Not the pitch — the mechanics. This is what a Monday looks like once the workflow is running: who does what, what you'd stop doing by hand, and what the workflow won't do for you.
Four parts, each with one job. None of them does another's job — that separation is the whole point.
This page describes the same workflow in operator terms. For the technical architecture behind each role — what "governed execution" actually enforces — see Platform.
A request comes in — on WhatsApp, email, wherever your customers already reach you. You paste it in as it arrived; nobody has to re-type it into a quoting form first. Automatic intake straight from a WhatsApp number is not connected yet — today the request is handed to the workflow rather than collected by it.
A draft is ready in minutes — priced from your real list, not a guess. If something's missing — an item not in your catalogue, a price that isn't current — it stops and asks instead of filling the gap.
You, or whoever you've named, approve it. Every quote, whatever its value — a named person signs off on the exact version shown to them. There is no threshold below which it goes out by itself, and nothing reaches your books without that signature.
It's registered — and checked. Automatically, against your own books, by an account that can't write to them. If it matches, you hear nothing. If it doesn't, you hear about it first, with the exact line that's different.
Pruvato is a concierge partnership right now, not a self-serve product. That is not a disclaimer at the bottom of the page — it changes what your first month actually looks like, so here it is in the middle where you'll read it.
The engine drafts from the request, stops for your named approval, writes the draft to Zoho, and reads it back independently. Your decision is recorded durably against the run, with your name on it. Accepting a mismatch with a written reason is designed and not built yet.
The engine checks a customer's credit standing against your declared threshold before approval goes through — over the limit, it stops and tells you why, rather than quoting first and finding out later.
Automatic intake from a WhatsApp number isn't connected. Requests are handed to the workflow — pasted in — rather than collected from a live inbox.
Two policy rules run automatically today: your credit threshold and your discount floor. Changing either is a conversation with us — see your rules on the platform page.
Every run is recorded. The signed, independently checkable receipt is published for your runs by us rather than minted automatically per run — the next thing being built.
Checking a quote against your current price list before it goes out — the Guard does this on every line, every time, not just the ones you remembered to check.
Reading a message, then keying the same lines into a quoting form, then again into Zoho. The draft starts from the request text itself — you paste it once.
Finding out a quote in your books doesn't match what was approved because someone happened to notice. The Witness catches it on the next read, not on the next audit.
Scrolling back through chats and spreadsheets to work out what's still waiting on a customer. Every quote's state is a lookup, not a memory exercise.
A workflow this specific should be honest about where its help ends. Three places, each mapped to one of the three parties that make the workflow work.
The Drafter reads the request, looks up the real items, prices them from your real catalogue. What it produces is a quote whose every number is grounded in something real. That is not the same as a quote your business should send. A discount you'd give this particular customer, a special arrangement not written in your rules, a strategic decision to price this deal aggressively — those are judgments the Drafter cannot make for you, and the workflow doesn't try to. The approval step exists so you see the accurate quote and decide whether it's the right quote.
If you set your discount floor at 20% and one of your customers has a standing agreement for 25%, the Guard will happily let a 22% quote through — because 22% is above the rule you wrote. Rules faithfully enforced are not the same as rules wisely written, and one of the reasons the design-partner phase involves us setting up your rules with you is to catch this class of mismatch before it costs you a customer. The Guard is a machine; it does exactly what you tell it, no less and no more.
If someone on your team edits a quote directly in Zoho after it was approved — to fix a typo, adjust a price, change a customer — the Witness will notice on the next read and flag it: "this quote no longer matches the version that was approved." What the Witness cannot do is stop the edit from happening. Your Zoho is your system; your team has the access they have; a vendor who claimed to control that access would be claiming too much. What we can do is make the change visible, with the specific line that changed, so nothing drifts silently.
And one thing no workflow can do: even together, the three parties give you a quoting process you can trust. They cannot give you a quoting process that sells more. Better quoting hygiene wins deals that would otherwise be lost to bad quotes, and it saves margin that would otherwise leak — but the customers you win still have to want what you're selling. Pruvato removes a category of failure; it does not replace the work of running a business well.
The fastest way to see whether this fits your business isn't a demo of the workflow — it's a free report on your own quoting history. Same rows, same honesty, no commitment.