AI System for Comparing Supplier Quotation Items

The problem

In day-to-day operations, similar items may have different scopes. When quotations, specifications, quantity lists, currencies, delivery terms and exclusions are spread across different screens, documents or team records, the issue may not be noticed in time.

This fragmented data delays preparation of the quotation comparison table and forces staff to search for supporting records again by hand. Before deciding, the procurement specialist and technical lead must complete missing fields and verify each inconsistency individually.

A common exception is that an equivalent product may not deliver the same performance. If this context is not checked, an incomplete or incorrect automation could treat a non-compliant quotation as equivalent, so the output cannot be acted on directly.

What we set out to improve

• Number of incorrect matches separated by the specialist
• Share of outputs corrected or rejected by the procurement specialist and technical lead
• Time between the source record and the review output

How the system works

1. 01 Define the review scope and the authorised roles of procurement specialist and technical lead
2. 02 Collect the inputs: quotation, specification, quantity list, currency, delivery terms and exclusions
3. 03 Check the source identifier, date, version and data freshness
4. 04 Separate items by unit and scope
5. 05 Create candidate matches against the specification
6. 06 Check the exception: an equivalent product may not deliver the same performance
7. 07 Place low-confidence or conflicting records in a separate human review queue
8. 08 Write the source identifier, rule, output and timestamp to the audit trail
9. 09 Have the procurement specialist and technical lead approve, correct, defer or reject the output

Methods we used

• Unit, product and scope dictionary matching
• Source, format, version and required-field checks for quotations, specifications, quantity lists, currencies, delivery terms and exclusions
• A missing-data alert instead of an estimate when inputs are absent or conflicting
• Separate handling of exceptions and the main rule
• Joint retention of the source identifier, rule result, quotation comparison table and human decision

Where people stay involved

Data and security

Access to quotations, specifications, quantity lists, currencies, delivery terms and exclusions is limited to the minimum permissions needed for the task. Identifiers and sensitive fields are masked where possible, and only the necessary section is sent to the model rather than the full raw file. Access, output and the decisions of the procurement specialist and technical lead are logged; production use does not begin until the organisation approves the retention period and data location.

Who this suits

Frequently asked questions