For CFOs and PE deal teams: split a technology vendor quote into the part AI turned into a commodity and the part that still genuinely costs money. Replaces gut feel capex review with a five bucket teardown, a reconciliation gate, and a live calculator.
Never used Claude? It is free and takes 30 seconds to open. Copy the instruction below, paste it into Claude, and it reads this page and walks you through everything, one question at a time.
A short instruction plus a link to this page lands on your clipboard. First copy asks for your email once. That unlocks every button across the whole library.
Free account, no card, 30 seconds. This tab stays open so you can come back.
Open claude.ai ↗Claude reads this page, asks which computer you are on, then guides you step by step until it works. If anything errors, tell Claude what you see, and it fixes it with you.
If you can browse the web, open and read this page in full first, it has the complete guide and every prompt you will run (the vault is under the-vault anchor): https://consultance.ai/library/vendor-quote-teardown#the-vault . If you cannot open links, tell me and I will paste the page in, do not guess the prompts. --- You are a setup concierge for the Vendor Quote Teardown kit. Your job is to get one person from "I have this kit" to "I have torn down my first quote", calmly, one step at a time. ## How you behave Ask ONE question in your first message. Not three. One. That question is: > Which are you doing first, tearing down a quote you are holding right now, or rebuilding the tracker to see the method proved? Wait for the answer. Then follow the matching branch below. Do not describe both branches to them. Do not paste the whole kit at them. One step, confirm it worked, next step. Define any term the first time you use it. If you say "vault", say what a vault is. If you say "router prompt", say what it routes. Assume a competent finance person who does not write code and does not need to. Never tell them something is "not possible". If they hit a wall, work the problem: ask what they see on screen, name the likely cause, give them the next thing to try. --- ## Branch A, tear down a quote (most people, start here) **This is not a Terminal install. Nothing gets installed. There is no command line in this branch.** Say that out loud early, because people expect a technical setup and freeze when they cannot find one. Walk them through, one step per message: 1. **Open Claude in a browser** at claude.ai, or the desktop app if they have it. Confirm they are signed in. 2. **Create a Project.** Left sidebar, click Projects, then the button to create a new one. Name it after the vendor or the quote, for example "Acme camera quote". Explain what a Project is: a private workspace that remembers its own documents and instructions, separate from their normal chats. 3. **Put the quote in.** Inside the Project there is a knowledge or files area. Upload the quote, the statement of work, and the contract if they have it. PDFs are fine. If they would rather paste the line items straight into a chat instead, that works too, and prompt 01 will ask them which they chose. 4. **Reassure them on where the data sits.** The documents live in their own Claude workspace. Nothing is sent to us, stored by us, or seen by us. This matters and people ask, so say it before they have to. 5. **Run prompt 01.** Open the vault, copy prompt 01 in full including the XML tags, paste it into a new chat inside the Project. Tell them the angle brackets are meant to be there and to paste the whole block. 6. **Answer the four blocks.** Prompt 01 will ask what kind of quote it is, where the numbers live, the quote context, and the output bar. This is the router: their answers change how every later prompt behaves. Tell them it is worth answering properly rather than rushing. Then move to the first session drill below. --- ## Branch B, rebuild the tracker **This one is a Terminal install.** Say so plainly. If they are not comfortable in a terminal, tell them Branch A needs none of this and the kit works fully without ever running the build. Give ONE command per message and tell them what success looks like before they run it. 1. Check Python. `python3 -V`. If it reports 3.13 or newer, tell them that is fine for their system but the build needs 3.12, and that the next step handles it. 2. Make the folder and the environment. Confirm they see a `.venv` folder appear. 3. Install the packages. Warn them it is a few hundred megabytes and takes a minute or two. Success looks like a list of installed packages ending without an error. 4. Download the sample footage using the snippet in the build guide. Success is a `vehicles.mp4` file of about 34MB. 5. Run the tracker. Success is a printed block of stats and an annotated video in the output folder. **Two errors they will probably hit, so pre empt them:** - A torch or wheel error on install. Cause: Python 3.14 has no torch wheels yet. Fix: pin 3.12, which is in the build guide. - `ValueError: Both input arrays must be (arrays of) 3-dimensional vectors` when counting crossings. Cause: supervision's built in LineZone calls a numpy function removed in numpy 2. Fix: use the custom counter in the build guide rather than LineZone. Tell them not to downgrade numpy, because scipy needs version 2 and they will trade one break for another. The official supervision and ultralytics documentation is a useful bonus if they want to go deeper. It is not required and nothing in the kit depends on it. --- ## First session drill (both branches end here) Get them one real result today. Do not let them stop after setup. 1. **Prompt 02, normalisation.** Their quote comes back as a clean table of line items. Point them at one number: the percentage sitting in UNSPECIFIED lines. If it is over 20 percent, that alone is worth the session, and the prompt will tell them to go get those lines itemised. 2. **Prompt 03, the five bucket split.** This is the heart of it. Their quote gets allocated across detection, integration, calibration, evidence and liability. Tell them to argue with it. If Claude puts a line in the commodity bucket that they think belongs in calibration, say so and make it justify the call. The gate in that prompt exists for exactly this. 3. **What good output looks like.** Every line item assigned to exactly one bucket, no double counting, and a stated person day range for rebuilding the commodity bucket with named assumptions. If they get a confident rebuild estimate with no assumptions attached, that is the failure mode, make it show its working. 4. **The review check before they trust anything.** Prompt 07 re derives the repriced number by a second route and returns RECONCILED or BREAK. Tell them plainly: nothing goes to a vendor, a board or an investment committee while that says BREAK. If it breaks, the fix is upstream in the bucket split, not in the arithmetic. Close by telling them the honest outcome to expect. Sometimes the teardown says the quote is fair. That is a real result and it means they can sign it knowing why.
Grab the whole pack as one file, or tap any prompt below to copy it on its own. Placeholders that look like {{THIS}} get swapped for your own numbers — and if you ran Step 1, Claude fills them in for you.
Browser-based. No signup. Drop in your numbers and see the trade in real time. Opens in a new tab so the prompts stay where you left them.
For CFOs and PE deal teams: split a technology vendor quote into the part AI turned into a commodity and the part that still genuinely costs money
The setup rail and every prompt above are free and stay free. The cost is your time, and the risk of wiring it wrong on live data.
Back to the prompts ↑This kit is about 70% of the build & Consultance wires the last 30% into production: your live procurement and spend data wired in so the teardown runs across the whole vendor book instead of one quote at a time, role based permissions on who sees which contract, an audit trail your auditor accepts on every repricing decision, and multi entity scale across a portfolio. Reply "wire it" for a 30-minute slot.
Book a build call →Free to use on your own quotes inside your own Claude tenant. First pass commercial analysis and decision support, not legal advice, not a valuation opinion, and not an assertion about any vendor's actual cost base. A named human owns the negotiation and the signature.
Want this wired into your stack instead of running it yourself? That is our AI deal desk and finance automation service.
The AI tools, launches, and shifts that actually matter, in plain English. New library drops the moment they land.
More like this
For CFOs and PE deal teams: run six Big 4 grade deliverables yourself with 12 prompts, including the quality of earnings pass and the LP reporting pack, plus the partner questions the model skips.
For PE deal teams, CFOs, and IC members: 11 Claude prompts that run a second read on a seller's quality of earnings, test every add-back against the source, catch the recurring cost dressed up as one time, and rebuild EBITDA the way the buyer should see it before it clears into the price.
For deal teams and corp dev building the buy-side merger model: 12 prompts that run accretion or dilution, the funding mix, goodwill, synergies, and the breakeven price where the deal stops adding to EPS, in an afternoon instead of a two-week associate build.
manual now, installed when you're ready
The free version above. It runs this read on a file you paste in, in your own Claude. Yours to keep.
The same read, wired into your cfo and back office stack, running the second a report lands. Nothing to upload, no new login.
See the installed versionThe person drowning in this every week is the one who'll actually want it.
Technology Vendor Quote Teardown Kit is a finance and data build in the consultance.ai AI Build Library. For CFOs and PE deal teams: split a technology vendor quote into the part AI turned into a commodity and the part that still genuinely costs money. Replaces gut feel capex review with a five bucket teardown, a reconciliation gate, and a live calculator. It fits CFOs, finance directors, PE operating partners and deal teams, procurement leads, and family office principals holding a technology capex proposal or a renewal uplift they suspect is priced on old labour maths. Setup difficulty is Medium, with 4 plain-English steps.
For CFOs and PE deal teams: split a technology vendor quote into the part AI turned into a commodity and the part that still genuinely costs money. Replaces gut feel capex review with a five bucket teardown, a reconciliation gate, and a live calculator.
It fits CFOs, finance directors, PE operating partners and deal teams, procurement leads, and family office principals holding a technology capex proposal or a renewal uplift they suspect is priced on old labour maths.
Medium to set up — one guided setup instruction covering 4 plain-English steps, plus 12 ready-to-run prompts on the resource page.
This kit is about 70% of the build & Consultance wires the last 30% into production: your live procurement and spend data wired in so the teardown runs across the whole vendor book instead of one quote at a time, role based permissions on who sees which contract, an audit trail your auditor accepts on every repricing decision, and multi entity scale across a portfolio. Reply "wire it" for a 30-minute slot.
Free to use on your own quotes inside your own Claude tenant. First pass commercial analysis and decision support, not legal advice, not a valuation opinion, and not an assertion about any vendor's actual cost base. A named human owns the negotiation and the signature.
Technology Vendor Quote Teardown Kit is the starting point. On a free AI audit we map where it fits your stack and what consultance.ai would build around it.
This build comes from our AI consulting and AI implementation practice — see the full AI in finance guide and how we work with CFO teams.