AI use case · Operations
AI quotation drafting from your own price lists and rules, with sales approving every quote.
For B2B sales and sales-support teams that build each quote by hand from spreadsheets, old quotes and email threads. AI reads the request, applies your price list and rules, and prepares a draft. A person approves it before it reaches the customer.
What is AI quotation drafting?
AI quotation drafting reads a customer's request, matches each item to your catalogue and applies your own price lists, discount rules, validity periods and terms to prepare a draft quote. Each line shows the rule that set it. Anything outside the rules, such as a special discount or an unknown item, goes to a salesperson before the quote is sent.
The problem
Every quote is rebuilt from spreadsheets and old emails.
In many UAE trading, distribution and service companies, a quote request arrives as a short email or WhatsApp message, often a photo of a list. Someone in sales support finds each item in the catalogue, opens the current price list, checks the customer's tier and copies everything into a template.
The rules live in people's heads as much as on paper. Which customers get which discount band, minimum order quantities, delivery charges by emirate and how long a price holds all depend on who prepares the quote. When that person is on leave, quotes slow down or go out with the wrong terms.
Customers often want the quote in Arabic and English, then a revision after changing quantities. Each revision repeats the process, and nobody can easily see why two customers received different prices.
Manual assembly
Items, prices, discounts and terms are copied by hand from several files into one template.
Rules in people's heads
Discount bands, minimum quantities and delivery charges depend on who prepares the quote.
No trail
When a price is questioned later, it is hard to see which rule or approval produced it.
How it works
One quote request, from message to approved draft.
An illustrative workflow. Intake channels, pricing rules, approval limits and the target system are agreed for each project.
Intake
Requests arrive in the agreed channel: a sales mailbox, a web form or messages your team forwards from WhatsApp.
Item matching
AI reads the request and matches each line to your catalogue by code, description or specification. Lines it cannot match with confidence are marked, not guessed.
Pricing rules
Prices come from your current price list through a fixed calculation step, not from the AI model. Customer tier, quantity breaks, delivery and validity are applied, and each line shows the rule used.
Questions and gaps
Specification questions are answered from your approved product data. Missing details, such as a delivery location or a quantity, become a short question for the salesperson to ask.
Routing
Requests outside the rules go to the right person with the context attached: special discounts, new customers, large orders, unusual items or non-standard terms.
Approval and send
A salesperson reviews the draft beside the original request, edits it if needed and approves. The approved quote is created in your ERP or CRM, in Arabic, English or both.
Who does what
What the AI does · What a person checks.
What the AI does
Reads quote requests from emails, forms and forwarded messages
Matches requested items to your catalogue and flags unclear lines
Applies your price list, customer tiers and quantity rules, with the rule shown on each line
Answers specification questions from your approved product data only
Drafts the quote and cover note in Arabic, English or both
Sends out-of-rule requests to sales with a summary of why
What a person checks
Every quote before it reaches the customer
Any discount, term or item outside the agreed rules
New customers and customers with special agreements
Large or strategic deals above the threshold you set
Changes to the price list and to the rules themselves
First pilot
A bounded first pilot on your own quote history.
We test on your past requests and quotes before anyone talks about results. Scope, timeline and price are confirmed in a written proposal.
One product range
Start with a range that is quoted often and has clear price lists and rules.
One channel and one system
One intake channel, such as the sales mailbox, and one target system for the approved quote.
Past requests as the test set
Real requests and the quotes your team actually sent, shared safely after the first call. Drafts are compared line by line with what was sent.
Agreed acceptance
Agree what counts as a correct line, which differences matter and when a request must go to a person, before the evaluation runs.
Data, hosting and Arabic
Where the data goes, and how Arabic is handled.
Hosting
The application and stored data can be hosted on Azure UAE North, on private servers managed by Vloud, or in your own cloud account. The AI model is used through a provider agreed per project, and its processing region may be outside the UAE.
Price lists
Price lists and customer terms are commercially sensitive. Who can see them is set by sales role, and what the model provider may store is agreed before the pilot.
Arabic
Arabic requests and bilingual quotes are tested as their own sample, including product names that customers write in Arabic.
Retention
What is stored, for how long and who can see it is agreed with your sales and finance owners.
Honest limits
Where it does not work yet.
- Drafts are only as good as the price list. If your rules are not written down yet, the first step is writing them down with your team, and we will say so.
- Vague requests such as 'same as last time' need a person, unless the customer's order history is connected.
- Engineered or project-based quotes that need a site survey or design work stay with your estimators. The workflow can prepare only the standard lines.
- We do not quote accuracy or time savings before testing on your own requests. Results are measured on a held-out sample first.
- If your ERP or CRM already has a quoting tool that covers the need, configuring it may be the better first step.
What drives cost and timeline. The size of your catalogue, the number of price lists and customer tiers, and how complex the discount and approval rules are. Intake channels, languages, the ERP or CRM integration route and hosting also count. We don't publish a price list; scope, timeline and price come in a written proposal after the discovery call.
Evidence
What we can show you.
Every example is labelled: delivered work, our own product, a demonstrator, or an illustration with synthetic data.
- Delivered work
Odoo for Ashbilia Contracting
We implemented Odoo accounting, purchase, employees and inventory for Ashbilia Contracting LLC, replacing a fully manual process. It is ERP work, not an AI quotation project.
- Synthetic data
The illustrative AI workflow
A step-by-step walkthrough with a synthetic invoice, not a quote. It shows the pattern this workflow follows: AI prepares, a person reviews, and only the approved record reaches the system. It is not a live client connection.
- Illustrative
What drives the cost of an AI implementation
Our guide to the factors that set the cost of an AI workflow in the UAE, from data and integrations to review and hosting.
Before you start
Questions buyers ask.
Does the AI decide the price?
No. Prices come from your price list and rules through a fixed calculation step. The AI reads the request, matches items and drafts the wording. It never invents a price or a discount.
What happens to requests outside our rules?
They go to the right salesperson with a short summary: what the customer asked for, which rule it falls outside and what the draft would be. Nothing outside the rules is sent without approval.
Can it answer technical questions about our products?
Only from product data you approve, such as datasheets and specifications. If the answer is not there, the question is marked for your team instead of being guessed.
Can quotes be in Arabic and English?
Yes. Drafts can be prepared in Arabic, English or both, using your approved product names and terms. Arabic requests are tested as their own sample in the pilot.
Which systems can it create quotes in?
An ERP or CRM with a supported API or an agreed import route. We review your system and version before confirming the integration scope, because an API alone does not prove that every update you need is supported.
What drives the cost?
The size of the catalogue, the number of price lists and rules, intake channels and the integration route. We confirm scope and price in a written proposal after a free 30-minute discovery call.
Who owns the code?
You own the source code and IP of the custom workflow we build for you. Our own products stay ours, and any licence terms are agreed separately.
Related
Related AI use cases.
Supplier quotation comparison
Read supplier quotes into one comparison, align items, units and terms, and flag missing lines and exclusions before your procurement team makes the award.AI and intelligent automation
The service page for AI workflows with review steps and system integration.AI automation in Dubai
How we scope a first AI workflow with your team.Enterprise operations
Systems we build for operations-heavy companies in the UAE.
Next step
Test it on your own documents.
Tell us what the work looks like today. We reply with the questions we need answered before a pilot can be scoped.
