Home · Business · HR & People · Performance & training

PARTLY

As of 13 August 2026, AI can only partly build an employee skills matrix.

This still needs a person who signs their name to it.

Can you do it?

15 minutesto a draft.

30 minutesto something you’d act on.

Cost, all in£0

Skill neededchat-fluent

Who has to check ita colleague

What the alternative costsManatal is AI recruitment software that provides candidate scoring, enrichment and pipeline automation, although it is not a dedicated employee skills matrix tool.

If this goes wrong: inaccurate or inconsistent ratings influence training, promotion or performance decisions and the employer has to correct the process.

What to actually do

  1. Hand it to a person

    The route this page recommends

    A person who owns the outcome does this end to end, worth it when the failure is dear.

  2. Use a tool built for this

    Second choice
  3. Do it yourself

    The distant third

    A chat interface, chat-fluent skill, and roughly 30 minutes until you can act on the result.

    How to actually do it

    1. Open your current competency framework and copy the skill names, definitions and proficiency scale into the prompt.
    2. Gather the latest anonymised employee records, including role, self-assessment, manager assessment, qualifications, work samples and dated observations, and remove unnecessary personal information.
    3. Paste the framework and employee evidence into the prompt, then ask the chatbot to return the matrix in a spreadsheet-ready table.
    4. Copy the generated table into Excel or Google Sheets and compare every employee name or ID, skill level and evidence note with the source records.
    5. Ask each relevant manager to check whether the evidence supports the assigned level and to mark missing evidence rather than correcting ratings without a reason.
    6. Ask HR or the process owner to check that the same proficiency definitions and evidence standard were applied across employees before using the matrix for training planning.
    7. Send each employee their relevant section or summary through the organisation's approved system and record agreed development actions separately from the assessment matrix.

    Prompt

    Build an employee skills matrix from the information below. Use only the evidence provided and do not invent skills, scores, qualifications or performance facts. Treat each employee consistently. If evidence is missing, write "not evidenced" rather than guessing.
    
    Competency framework:
    [Paste the skill or competency names, definitions and any agreed proficiency levels here]
    
    Proficiency scale:
    [Paste the organisation's scale and definitions here, for example beginner to advanced, if one exists]
    
    Employee evidence:
    [Paste anonymised employee names or IDs, current roles, self-assessments, manager assessments, qualifications, work samples and relevant observations here]
    
    Create:
    1. A table with one row per employee and one column per skill.
    2. A level for each skill only where the evidence supports it.
    3. A short evidence note for every level assigned.
    4. A separate list of missing evidence and assumptions, with no assumptions treated as facts.
    5. A team summary showing skills that are well covered, skills with possible gaps and skills that need more evidence.
    6. Individual development priorities linked to the stated framework, without making promotion, pay, disciplinary or redundancy recommendations.
    
    Keep the language neutral and suitable for an internal UK workplace document. Flag any rating that may be inconsistent with the framework or lacks enough evidence. Return the matrix in a format that can be copied into a spreadsheet, followed by the notes and checks.

    Open it prefilled in ChatGPT or Claude, or copy it into Gemini, which takes no prefill link.

What it gets wrong

  • AI cannot know whether an employee's work meets your organisation's standard unless you provide reliable evidence and definitions.
  • AI cannot resolve conflicting self-assessments, manager views or informal feedback without a human deciding which evidence matters.
  • AI cannot guarantee that ratings are consistent or free from bias across teams and protected characteristics.
  • AI cannot take responsibility for using the matrix in promotion, pay, performance or redundancy decisions.
  • AI cannot replace the conversations needed to agree development priorities with employees and managers.

What caps this at PARTLY: judgement under ambiguity, context depth and legal accountability.

How we scored this

Five axes, each scored nought to two by hand: ten means AI carries the task cleanly, and the thresholds that turn a total into YES, PARTLY or NO are published in the methodology. Each axis name links to its definition.

AxisScore (0–2)
Output2
Inputs1
Verification1
Liability1
Effort delta2
Total7 / 10

FAQ

Can ChatGPT create an employee skills matrix?
Yes, it can create the table and organise skills, levels, evidence and development gaps. It cannot decide whether each rating is fair, so a manager or HR colleague must check the evidence and approve the matrix.
What information do I need for an employee skills matrix?
You need the competency framework, proficiency definitions, employee roles and current evidence such as assessments, qualifications, work samples and observations. Missing evidence should be recorded as missing rather than filled in by the model.
Can AI assess employee skills fairly?
Not by itself. AI can apply the same stated structure to the information you provide, but people must check the evidence, identify inconsistent standards and consider whether the process creates unfair effects.
Can I use an AI skills matrix for promotion decisions?
It can support evidence gathering and development planning, but it should not make the promotion decision. The employer and responsible managers must check the evidence, apply the organisation's process and own the consequences.

Nearby answers

Assessed by gpt-5.6-luna (gpt-5.6-luna) on 2026-08-13, second-checked by an independent model. Wrong somewhere? Email [email protected] and it gets re-checked.

The newsletter

AI news, new answers and product picks, straight to your inbox.