Every B2B sales team knows the quoting bottleneck. A customer sends a request. Someone opens the email, reads the document, identifies the products, looks up prices, builds the response in the customer's format, routes it for internal review, and sends. For a single request, this takes anywhere from twenty minutes to several hours depending on complexity.
At low volume, this is manageable. At scale, it becomes a structural problem that no amount of effort or headcount can sustainably solve.
The scale problem
Why Manual Quoting
Doesn't Scale
The math is unforgiving. A team processing forty quote requests per day at an average of forty-five minutes each is spending thirty hours of working time per day on quoting alone. That is the equivalent of nearly four full-time roles doing nothing but reading documents and entering data. As volume grows, the options are stark: hire more people, let response times slip, or triage, which means some customers wait longer and some quote requests never get answered at all.
The downstream effects are real and largely invisible. Missed tender deadlines mean lost opportunities that never appear in any report. Slow response times hand first-mover advantage to competitors on every multi-supplier RFQ. Errors in quotes create corrections, complaints, and margin erosion that shows up in the numbers but rarely gets traced back to the quoting process itself.
30h
of working time per day spent on quoting alone, for a team processing 40 requests at 45 minutes each
4×
equivalent full-time roles doing nothing but reading documents and entering data
0
missed tender deadlines appear in any report, because the lost opportunity is invisible
What automation covers
What Aspects of Quoting
Can Really Be Automated
Not every step in the quoting process is equally automatable. Setting accurate expectations before implementation prevents the most common rollout mistake: automating the easy steps while leaving the actual bottleneck in place.
Steps that can be fully automated
- Inbox monitoring and request classificationReading incoming emails, identifying which ones are quote requests versus orders, inquiries, or complaints, and routing each to the correct workflow.
- Document reading and data extractionExtracting all relevant fields from the incoming request regardless of format: the customer's identity, product references or descriptions, quantities, required delivery dates, and any specifications or special requirements.
- Product matchingResolving the customer's description or article number to the correct internal SKU, including customer-specific codes, description-based matching where no code is provided, and contextual references to previous orders.
- Pricing applicationApplying the correct price for each matched product based on the customer's applicable price list, volume tier, or contract agreement, and flagging deviations above a configured threshold.
- Draft quote generationBuilding a complete draft response in the required format: the customer's Excel template, a GAEB X84 file, a PDF, ready for human review before it goes out.
- ERP record creationWriting confirmed quote data directly to the ERP system without a manual import step.
Steps where AI assists but humans confirm
- Exception resolutionWhen a product cannot be matched, a specification is ambiguous, or a price deviates significantly, the AI assembles the context and surfaces a suggested resolution. The human confirms or overrides.
- Margin optimisation decisionsWhen multiple products meet the specification, the AI can rank candidates by margin and surface the best option for the human to approve. The final pricing decision on complex or sensitive quotes stays with the team.
- New customer handlingQuote requests from contacts not yet in the ERP require a human to confirm or create the account.
Steps that always require human judgment
- Commercial negotiationAny situation where the customer is contesting a price, requesting a special arrangement, or escalating a concern.
- Genuinely ambiguous specificationsSome requests cannot be resolved from the product catalog or order history without clarification from the customer.
- Strategic account decisionsPricing and product decisions on key accounts where commercial judgment, relationship history, and competitive context all matter.
