A factual appraisal

How can you use AI to structure an evidence-based appraisal and prepare for a fair appraisal discussion?

1250 € per session, excl. VAT

In short

This half-day session of 3 hours helps turn scattered observations into a usable appraisal file. Participants use generative AI to organise facts, formulate open questions and propose objectives to discuss with the employee. They check for bias, traceability and the quality of the file before human validation.

Why this training?

A meeting based solely on recent memories can give undue weight to an incident and overlook less visible contributions. Gathering dated facts, clarifying their context and identifying missing information helps prepare for a more balanced discussion. AI helps classify observations and rephrase questions, but it can invent connections or reinforce an initial judgement. The manager remains responsible for the assessment; no scoring or sensitive decisions are automated. Running case study: At ‘Varnelle Maintenance’, a fictional organisation, a manager is preparing the annual appraisal of a service intervention coordinator. The file contains three initial objectives, documented delays, messages of thanks from customers and a change in the scope of responsibilities. Some observations contradict one another: clarification needs to be prepared without making assumptions about overall performance. Transfer to the workplace: At the next meeting, reuse the evidence matrix and adapt the guide to the internal framework. Have the file reviewed by a qualified HR professional, then check after the discussion whether uncertain points have been clarified and the objectives genuinely discussed.

Objectives of this training

  • Distinguish observable facts, interpretations and missing information.
  • Structure an appraisal linked to objectives and the work context.
  • Draft open questions using generative AI.
  • Formulate proposed objectives with associated indicators and resources.
  • Check for bias and traceability before HR validation.

Who is it for and what are the prerequisites?

Roles and tasks concerned

  • Line managers preparing appraisal meetings with their employees.
  • HR managers supporting managers in preparing appraisals.

Prior skills

  • Have already used a generative AI assistant for a simple writing task.
  • Be familiar with your organisation’s appraisal template and managerial responsibilities.

What you will work on

  1. Build an evidence base

    Objective: Separate observations from judgements. Scope of the appraisal, available sources, confidentiality and selection of only the necessary information.

  2. Build an appraisal without extrapolation

    Objective: Obtain a summary faithful to the information supplied. Instruction specifying context, sources, format and a prohibition on inventing information; recency and confirmation bias.

  3. Prepare the dialogue and objectives

    Objective: Turn the appraisal into a discussion aid. Open questions, factual feedback, space for self-assessment; objectives that take available resources into account.

  4. Test the file and have it validated

    Objective: Measure the quality of the preparation. Traceability checks, balance of the appraisal and validation by a qualified HR professional.

What you take away

  • Matrix of facts, sources, context and points to clarify.
  • Balanced and traceable preparatory appraisal.
  • Meeting guide with questions and proposed objectives.
  • Checklist completed before HR validation.

Method and assessment

  • Short teaching inputs followed by immediate practice on the running case study.
  • Individual work followed by work in pairs, with peer review of the outputs.
  • Guided group discussion: what works, what needs correcting and what remains a human decision.
  • Preparation: Arrange access to an AI tool authorised by the organisation; use only fictional or anonymised cases. Bring a blank meeting template with no personal data; a teaching template can be used instead.

Observable criteria: Each assessment in the appraisal refers to a supplied fact or is explicitly left for verification. ; The appraisal covers the entire period and distinguishes results, resources and changes in the scope of responsibilities. ; Questions avoid personality judgements and allow a contradictory response. ; Each proposed objective specifies an indicator, deadline, resources and review point. ; The checklist flags identified biases, corrections and items submitted for validation..

Frequently asked questions

Can real meeting reports be entered?

The exercises use only fictional or anonymised data, in authorised tools.

Can AI assign a rating?

Not in this approach: the assessment and any sensitive decisions remain human.

What should you do if evidence is missing?

Flag the uncertainty and prepare a clarification question, without artificially filling out the file.

Does the template cover the applicable obligations?

It is a teaching aid; requirements specific to your context should be checked with your legal department.

Last updated: