QuoteAutomation
Read the enquiry, price it from your own rules, and send a clean quotation fast, while keeping a person in charge of anything unusual.
Quote Automation: what the work involves
A customer emails a list of items, a photo of a drawing or a WhatsApp voice note, and waits. Someone must interpret it, look up prices in a spreadsheet, check stock and lead times, apply customer-specific discounts and type up a document. By the time the quote arrives, the buyer may have accepted a faster competitor, and the team has no record of how long each quote took or why some were lost.
DevKey builds a quoting flow that reads the incoming request, whatever its format, and converts it into structured line items matched to your product catalogue. Pricing is then calculated by deterministic rules covering price lists, volume breaks, customer tiers, freight and tax. A branded quote is generated, with unclear items or custom work sent for human costing. Approvals follow your thresholds, and sent quotes are tracked so follow-ups and conversion reporting happen automatically.
Core features
Request interpretation
Emails, messages, spreadsheets and attached drawings are read and mapped to catalogue items, quantities and specifications.
Catalogue and alias matching
Customers describe products in their own words, and an alias dictionary plus fuzzy matching ties those to the right SKU or service.
Rule-based pricing engine
Discounts, tiers, minimum charges and freight are applied by code, which makes each price explainable and consistent.
Availability and lead time check
Stock levels or production capacity are consulted to put realistic delivery estimates on the quote.
Approval thresholds
Quotes above a value or with unusual discounts wait for a manager, who sees the logic and edits before release.
Follow-up and conversion tracking
Open quotes trigger timed reminders, and win-loss data is recorded so you can see which quotes convert and why others stall.
What we get right before launch
Prices must never be improvised
A model that guesses a price can lose money on every order. All pricing comes from your rules and data, and the model is limited to understanding the request and writing the cover text.
Ambiguous specifications
Customers leave out sizes, finishes or quantities. We design clarifying questions and a hold state for incomplete requests, instead of quoting on assumptions that lead to disputes.
Catalogue data quality
Automation exposes messy price lists and duplicate items. A cleanup of your catalogue and rules is part of the work, and it is often as valuable as the quoting tool itself.
Tools and technology
- OpenAI GPT
- Anthropic Claude
- Python
- FastAPI
- PostgreSQL
- Odoo and ERPNext APIs
- Google Sheets API
- n8n
- WhatsApp Business API
Common questions, answered
Can it handle customised or one-off products?
It can price the standard components and flag the custom parts for an estimator. As more custom jobs are costed, patterns can be turned into rules, but unusual work will always need human judgement.
Which systems can it pull prices and stock from?
ERPs like Odoo and ERPNext, accounting packages, databases and well-structured spreadsheets. We check what data is dependable and where the single source of truth for prices should live.
How does it read drawings or photos?
Simple dimensioned drawings and photos with notes can be interpreted with vision models, but accuracy is limited. Complex engineering drawings generally need a human estimator, with the tool handling intake and pricing mechanics.
Is the customer talking to a bot?
That is your choice. Many businesses keep a person as the sender and use automation behind the scenes. Others add a chat intake, clearly labelled, with quotes still approved by staff.
More AI Agents & Automation services
All AI Agents & Automation servicesAI Proposal Generation
Produce a first draft of a tailored proposal or tender response in an hour, built from your approved content, and let your senior people spend their time on strategy and pricing.
Document Generation Automation
Produce correct, consistent documents from your data and templates in seconds, with AI handling the narrative parts and people approving anything that matters.
Approval Workflow Automation
Replace approval chasing by email and chat with a routed, tracked process that checks requests first and reminds approvers when they stall.
Ready to start your Quote Automation project?
Tell us what you need and we will come back with a clear scope, timeline and the questions worth answering before any build starts.