Every quote goes out with the customer’s price list and the margin protected.
Claude builds every quote under your commercial policy: applies the customer’s price list, calculates defensible discounts, protects the margin, adds taxes and multi-currency, and publishes a versioned proposal with a public link, no login required. A discount above the cap waits for a person’s approval.
Commercial proposal
Distribuidora Andina
No. P-9F3A · v2 · € EUR
Total · margin 16.40% · ITBIS applied
rate frozen in the snapshot
1
price list
per client and per country
0
margin errors
hard floor per line item
L1
the discount
prepared, waiting for approval
6
industries
same pricing engine
UNDER THE HOOD
It shows how it built the total, line by line.
A quoting tool that spits out a number without explaining the price list, discount, and margin is a tool your team will override by hand. Claude exposes every pricing decision: which list it applied, why that discount, and where the margin landed.
Sample request
The request comes in
Recurring distributor requests 480 units of the base SKU + 60 premium. Has "Wholesale" price list assigned and net-30 payment terms. The rep suggests an extra 12% discount to close this week.
Correct price list
customer = Wholesale → uses that list, not retail (−18% base already applied)
Volume discount
480 units crosses the 400+ tier → −7% automatic per policy, not negotiated
Extra discount requested
12% requested + 7% volume = 19% → margin would drop to 11%, below the 15% floor
Defensible counteroffer
approves 9% extra (margin 16.4%) + conditions the 12% on prepayment
Score
88/100
Verdict from Claude
Quote ready. Margin protected at 16.4%. Claude left a note explaining why not the 12%.
The request comes in
Energy opportunity: 7 line items (imported equipment USD, local installation DOP, engineering). Customer requests the proposal in EUR. ITBIS applies to only 4 items; imports are tax-exempt.
Mixed-currency line items
equipment in USD, labor in DOP → consolidated to EUR at today's rate, frozen in snapshot
Tax by line
ITBIS 18% only on the 4 taxable items; 3 imports marked exempt with legal basis
Margin per item, not global
engineering 42%, equipment 9% → weighted margin 19.6%, above BU floor
Version and approval
amount > BU threshold → requires management approval before publishing the link
Score
81/100
Verdict from Claude
Draft v1 generated. Pending management review by amount. Public link released upon approval.
The request comes in
Proposal sent 9 days ago, public link opened 6 times but not accepted. Customer asked by email to "adjust the scope." The rep wants to resend the same PDF.
Price friction signal
6 opens without acceptance + "adjust scope" → price objection, not product
Version confusion risk
resending the same link without versioning creates ambiguity about which they agreed to
Re-scope suggestion
create v2 removing 2 optional line items → reduces 14% without touching unit margin
Change traceability
v2 keeps v1 in history; link shows "updated" and notifies the contact
Score
34/100
Verdict from Claude
Alert to manager. Generate v2 with reduced scope and record the reason for the change.
WHAT AI DOES IN YOUR QUOTING ENGINE
Five pricing jobs you shouldn't be doing in Excel anymore
Each one runs automatically on every incoming order. Your rep receives a proposal ready to send, not a spreadsheet to fight with.
claude.configure(order)
Turns an order into structured line items
Reads the requirement in natural language and builds it as line items: product, quantity, unit, applicable price list, and dependencies. The order comes in raw and comes out as a configurable quote.
› input
Order: "I need 480 base units, 60 premium, installation included, invoice net 30"
› claude →
line 1: SKU-BASE × 480 · list=Wholesale
line 2: SKU-PREM × 60 · list=Wholesale
line 3: Installation · 1 service
condition: net-30 credit
→ quote configured · 3 line items
claude.price(quote)
Applies price list, discounts, and taxes with judgment
Takes the customer's price list, applies volume discount tiers, respects the margin floor per line item, and adds the tax that applies to each line. Multi-currency with frozen rate.
› input
Quote: 3 line items · customer=Wholesale · requested currency=EUR
› claude →
Wholesale list applied · base −18%
volume discount 400+ · −7% (line 1)
ITBIS 18% on taxable lines only
weighted margin: 16.40% · above floor
→ total: € 41,280.00 · frozen rate
claude.version(proposal)
Every change is a version, not a lost PDF
When scope or price changes, creates a new version keeping the previous one as an immutable snapshot. Nobody argues about "which of the three PDFs we agreed on" again.
› input
Customer asks to remove 2 optional line items from v1
› claude →
v1 → archived as immutable snapshot
v2: −2 line items · total −14%
unit margin unchanged
visible diff: what changed and why
→ public link points to v2 (current)
claude.approve(flow)
Routes to approval only when needed
If the amount, discount, or margin crosses a threshold, triggers the flow: draft → review → approved. If within policy, it passes through directly. No manual rules.
› input
Proposal: € 41,280 · extra discount 9% · margin 16.4%
› claude →
amount > BU threshold → review required
approver: commercial management
status: draft → under review
discount within policy ✓
→ on approval: enables publishing
claude.publish(link)
Generates the branded PDF and public link, no login required
Once approved, builds the branded PDF and publishes a web proposal at /proposal/:id that the customer opens without an account — and accepts with one click, leaving a timestamp.
› input
Proposal v2 approved · customer=Andean Distributor
› claude →
branded PDF generated · with your logo
public link: /proposal/9f3a (no login)
status: sent · view registered
accept = one click + timestamp
→ on acceptance: notifies + creates project
LIVE SYSTEM · Grupo Latitud
This is what it looks like inside an AI-operated system — and how it is governed.
A quote is a commitment of money. Here you see how it is built with the current list and the country tax, and what happens when the discount exceeds the cap.
Modules in this tour
- 01The current price list rules
- 02Move the discount and watch where it stops
- 03The commercial director decides with the full case
- 04What the customer did with the proposal
- 05The rule that holds all this up
Boosty Standard for Operating with AISimulated AI · demo data
THE SAME ENGINE
One quoting engine. The rules change, the judgment doesn't.
We don't build a different quoting engine per industry. The same pricing engine reads the specific rules of each business: lines of business, items, variants, taxes. This is already running in production.
Signals specific to the industry
Signals specific to the industry
Signals specific to the industry
Signals specific to the industry
Signals specific to the industry
Signals specific to the industry
LIVE QUOTE BUILDER
Play with the quote builder. The price rebuilds itself.
Change list, quantity, discount or currency: the engine recalculates the quote instantly and warns you if the margin falls below the floor. This is what your client builds step by step.
1 · Customer price list
2 · Number of units
120 uno volume tier yet (200+ / 400+)
3 · Negotiated extra discount
6%4 · Installation
5 · Currency
PUBLIC PROPOSAL
One link, no login. The client opens and accepts.
This is how your client sees the proposal at /propuesta/:id: no account to create, with your brand, and they accept in one click leaving a timestamp. Every opening is logged.
Commercial proposal · v2
Distribuidora Andina
| Line item | Qty | Total |
|---|---|---|
| Core platform · annual license | 1 | $ 18,400.00 |
| Implementation and API integration | 1 | $ 9,200.00 |
| Additional users · pack of 25 | 2 | $ 3,000.00 |
| Premium support · 12 months | 1 | $ 4,800.00 |
| Total · USD (ITBIS included) | $ 41,772.00 | |
Public link status
IN THE SYSTEM · PROPOSALS
Who opened the proposal, on which device and for how long.
The follow-up the agent proposes comes from those opens, not from a fixed date. Publishing the public link is a person’s action.
Real screenshot of the system · demonstration data
APPROVAL FLOW
Draft to sent, with no PDFs lost.
Every proposal moves through its flow and every change is a version, not another file in the inbox. If the amount crosses the threshold it goes to approval; if not, it passes straight through.
Claude builds v1: 7 line items, Wholesale list, selective ITBIS
Amount > BU threshold → routed to sales management
Management approves · margin 19.6% above floor · v1 sealed
Branded PDF + public link /propuesta/9f3a published
Version history · no ambiguity about “which one did we agree on”
initial draft · 7 line items · archived immutable
−2 optional line items · total −14% · unit margin intact
the public link always points to the current versiondiff visible to the client
IN THE SYSTEM · APPROVALS
The discount above the cap reaches the phone of whoever decides.
The quote builder prepares the proposal and leaves it waiting with the amount, the margin, the customer history and the risk. Approving or rejecting is one tap, and who and when is recorded.
- Within the cap, it is issued
- Above the cap, it waits for an approval
- Below the minimum margin, the policy blocks it
Real screenshot of the system · demonstration data
CONNECTED STACK
Lives on top of your core. Doesn't replace it.
The pricing engine connects to what you already use. If your system has a REST API, we talk to it; if it's legacy, we bridge it.
Claude · Anthropic
Claude PartnerThe pricing engine: configures line items, calculates discounts, protects margins, and reasons through it
Kommo CRM
PartnerThe quote originates from the deal and moves the pipeline stage upon acceptance
Monday.com
PartnerAccepted proposal → the project starts automatically with its line items
WhatsApp Business
The public proposal link goes out through the channel where your customer responds
Make / n8n
Webhooks to your legacy ERP or invoicing system without a modern API upon proposal acceptance
Supabase
Deno Edge Functions for pricing calculation, version snapshots, and public links
Frequently asked questions about Quotes & Proposals
No. It's a pricing layer that lives ON TOP of your core. It reads your price lists and discount rules where they already are; if you have a REST API we integrate natively, and if it's a legacy system we bridge it via Make/n8n. Your team keeps working where they work, except now the quote builds itself with judgment.
You define the floors. By default it applies price lists and policy discounts automatically (low risk), but any discount that crosses the margin floor or an amount threshold enters a human approval flow. It never publishes a proposal that breaks your margin without someone approving it.
Every approved proposal generates a URL like /proposal/:id (or /quote/:id, /offer/:id) that the customer opens without an account. They see the current version, download the branded PDF, and accept with one click leaving a timestamp. Every view and acceptance is recorded so you know the real level of interest.
Each change creates a new version keeping the previous one as an immutable snapshot. The public link always points to the current version and shows a diff of what changed. No more arguing about "which of the three PDFs we agreed on."
Yes. It consolidates line items in different currencies to the requested currency with the day's rate frozen in the snapshot, and applies the tax that corresponds to each line (taxable, exempt, differentiated rate). Tested with ITBIS, VAT, and multi-currency USD/DOP/EUR schemes.
Yes. We use it in insurance for 8-line-of-business quoting wizards where the customer builds their policy step by step, and in fintech for proposals with variants. The same engine works for both the rep's internal quoting tool and the public wizard.
No. A PDF generator fills in a template; this reasons through the price: which price list, which discount is defensible, where the margin lands, and whether it needs approval. The difference is pricing judgment and explainable versioning, not the document format.
It is defined in the assessment. The scope — what gets built first and what waits — comes from what we see in your operation, not from a catalog. Book 30 minutes and we give you the range in writing.
A WORD FROM THE FOUNDER
“A sale is won with the right quote on time: the customer’s price list, the defensible discount and the protected margin.”
A quote works like a recipe: the ingredients are the customer’s price list, the discount policy, the cost and the country’s tax. When each rep builds it from memory in a spreadsheet, the result depends on who prepares it and at what hour: the wrong list, one discount point too many, a proposal that lands after the competitor’s.
The quoting engine applies those rules line item by line item: the list that applies, the discount within the cap, the margin against its floor and the exact tax. It publishes the proposal versioned, with a link the customer opens and accepts without a login. The AI prepares the quote and the reason behind every number; a discount above the cap is decided by a person, with the evidence in front of them.
For the business, that means consistent prices across reps and a margin you see before signing. Book 30 minutes with me: we build one of your quotes with the engine, line by line. Who decides a discount above the cap today, and where is it written down?
Gabriel Montiel
CEO · Boosty Digital
LET'S TALK
Ready to quote with judgment, not with Excel?
Schedule a 30-minute assessment. Bring a real order of yours and we'll show the engine building the quote live: price list, discount, margin, taxes, and the public link. No corporate deck.