Quotations under control

How can you use AI to make supplier quotations comparable without concealing their differences or uncertainties?

1250 € per session, excl. VAT

In short

In a half-day session of 3 hours you use generative AI to extract the terms from several quotations, align their scope and prepare a reasoned purchasing recommendation. You leave with a verified comparison matrix, clarification questions and a method for checking calculations and sources.

Why this training?

A lower total amount is not enough to distinguish between quotations: units, quantities, included services, lead times and payment terms can affect the comparison. Missing information must remain visible rather than become an implicit assumption. AI helps structure information and articulate differences, but may confuse an optional item with an included service. The participant remains responsible for checks, weightings and the recommendation submitted to the decision-maker. Running case study: The fictitious organisation Lentavie Conditionnement needs to replace its shipping boxes. Three suppliers offer different pack sizes, minimum order quantities and delivery arrangements. One quotation omits transport costs; another offers a technical option whose usefulness has yet to be verified. Application at work: For a forthcoming authorised request for quotations, reuse the matrix, have a colleague review the differences and measure the proportion of verified fields, remaining omissions and calculation corrections needed before approval.

Objectives of this training

  • Define a common scope and explicit comparison criteria.
  • Extract commercial terms with a reference to their source.
  • Normalise units, quantities and included items without inventing missing data.
  • Check calculations and test the effect of weightings.
  • Draft a conditional recommendation and requests for clarification.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Buyers and purchasing assistants preparing requests for supplier quotations.
  • Operational managers who need to justify a sourcing choice.

Prior skills

  • Have previously used a generative AI assistant for a simple task.
  • Be able to read a quotation and perform basic calculations in a spreadsheet.

What you will work on

  1. Define the common basis

    Objective: Establish the rules before examining the quotations. Requirement, quantities, technical requirements, exclusion criteria and weighted criteria; distinction between compliance and preference.

  2. Extract without filling gaps arbitrarily

    Objective: Build a matrix that faithfully reflects the documents. Structured extraction instruction: field, value, unit, source and status. Handling options, exclusions and illegible entries.

  3. Harmonise and check

    Objective: Compare genuinely equivalent scopes. Conversion of pack sizes, minimum order quantities, transport, necessary options, distinction between amounts excluding and including VAT, and payment arrangements; independent checking of calculations.

  4. Justify the decision

    Objective: Present a defensible choice and its limitations. Structure of a purchasing note: compliance, decisive differences, risks, open questions and conditions for approval.

What you take away

  • Scope definition sheet and weighted criteria grid.
  • Comparison matrix with source references and verified calculations.
  • List of clarification questions for each supplier.
  • Recommendation note and record of checks.

Method and assessment

  • Short instructional sessions followed by immediate practice using the running case study.
  • Individual work followed by work in pairs, with participants reviewing each other’s outputs.
  • Guided group discussion: what works, what needs correcting and what remains a human decision.
  • Preparation: Ensure access to an AI assistant and a spreadsheet application authorised by the organisation. Prepare a fictitious or anonymised file containing three quotations and a purchasing requirement; remove confidential information before entering any data.

Observable criteria: Every decisive data item refers to an identifiable section of the quotation. ; The quotations are aligned with the same requirement, with units and scopes made explicit. ; Missing information is flagged, never replaced by assumptions. ; Calculations are reproduced in the spreadsheet and discrepancies corrected. ; The recommendation distinguishes between facts, assumptions and caveats that prevent approval..

Frequently asked questions

Should you automatically select the cheapest quotation?

No. Compliance with the requirement comes before ranking; the selected criteria must explain the trade-offs.

What should you do if transport has not been priced?

Flag the total as incomplete and request clarification, without treating the absence of an amount as meaning that it is free of charge.

Can you upload real quotations to the assistant?

The exercises use fictitious or anonymised cases, exclusively within tools authorised by the organisation.

Does AI validate the calculations and approve the supplier?

No. Calculations are checked separately, and the decision remains subject to internal purchasing approval rules.

Last updated: