← All use cases
Use case 02

Job evaluation

Size a role the same way every time, with a written reason for every call. A skill that follows your method, checks what is missing, and assesses the role, never the person.

Skill Works in Copilot, Claude or ChatGPT Your method, not ours
Try itRead a real result. Watch itFive steps, animated. Build itPrompts you can paste. Check the dataWhat is safe, and where. Take itDownload the skill.

What it does and why it matters

Job sizing goes wrong in two ways. The first is inconsistency: different people, on different days, give different answers for similar roles. The second is sizing the person instead of the role, so a strong performer's job drifts up and a struggling one's drifts down.

This skill fixes the process, not the judgement. It makes the assistant check the role description first, assess only the work, rate each factor against your level descriptions, quote the words behind each call, and flag the calls that could go either way. A person still decides the final size.

For your team, the gain is consistent answers whoever does the sizing, and a written trail you can show when someone asks why a role sits where it does.

A skill is a written set of instructions, saved as a file, that an AI assistant follows every time it does a particular job. It turns "a good first draft, if you ask the right way" into "the same method, every time, whoever is asking".

1Try it

Two saved conversations. In the first, an invented role is attached and sized. In the second, someone pushes back by pointing to the person in the role. Press Play to watch, or Show all to read.

Replay A Captured from Claude following this exact skill, November 2026. Invented role.
Replay B Captured from Claude following this exact skill, November 2026. Invented role.

These are real responses, saved and replayed. Nothing you do here is sent to an AI.

2Watch it

Five steps from a role description to a result you can defend. Press Play, or step through at your own pace.

Step 1 of 5Invented data
sample-role.md
Senior Payroll Analyst
Banksia Example Co (an invented organisation) · Reports to Payroll Manager
Purpose

Make sure every Banksia employee is paid correctly and on time, in line with our enterprise agreement and the relevant Modern Awards.

Main accountabilities
  • Interpret the enterprise agreement and Modern Awards for complex pay cases, and explain the outcome to managers and employees.
Decisions this role makes
  • Approves the pay run for release within the approved payroll process and calendar.
  • Recommends, but does not decide, where the agreement is unclear.

Start with the role description. The skill works from the words on the page. The highlighted phrases are the kind of wording it will quote as evidence.

The gate: is there enough to size this role?
  • Purpose of the roleFound
  • Main accountabilitiesFound
  • Decisions the role makesFound
  • Who the role deals withFound
  • Team sizeStated: none
  • BudgetNot stated. Assumed none, flagged, result marked provisional.Missing

The gate. If something essential is missing, the skill stops and lists what it needs instead of guessing. A silent team or budget does not stop it, but it is flagged and the result is provisional.

Five factors, each with a level and evidence
Expertise4
“Interpret the enterprise agreement and Modern Awards for complex pay cases”
Complexity3
“Recommends, but does not decide, where the agreement is unclear”
Reach of impact4
“every Banksia employee is paid correctly and on time”
Influence3
“Coach Payroll Officers on difficult queries”
Autonomy4
“Approves the pay run for release”

Factor by factor. Each call has a level, a one-sentence reason and the exact words that support it. If there are no supporting words, the call is treated as an assumption.

A call that could go either way
Reach of impact3 / 4
Sits between Level 3 (a team or process) and Level 4 (a significant service).
Stays at 4 because
The pay run covers the whole workforce, and the role completes the tax and super reporting.
Tips down to 3 if
Someone else fully re-checks the pay run before release, so this role's mistakes are caught within the team.

This one matters: at Level 3 the overall level would drop from 4 to 3.

Borderline calls. The skill names the two levels, says what would tip the call, and says whether tipping it would change the overall result.

Result
4of 6
Confidence
Medium
Status
Provisional
Rule
Middle value of the five levels
Expertise 4Complexity 3Reach 4Influence 3Autonomy 4

Provisional because the budget was not stated and was assumed to be none. Medium because two borderline calls could each lower the overall level.

A result you can defend. The level, the rule used, the confidence and the reason, every call tied to the role's own words. A person still decides the final size.

The role and every rating in this walkthrough are invented, for illustration only.

3Build it

You can build your own version with any capable AI assistant your employer allows, for example Microsoft Copilot, Claude or ChatGPT. The skill on this page uses an illustrative five-factor method written for this site. The point is to put your own method in its place.

Use only material you are allowed to share. Do not paste a licensed job evaluation method you do not have the rights to share. If your method belongs to someone else, write down your own plain-English version of how you apply it, or ask your supplier what you may load into an AI tool.

Step 1: Write your method down

Most methods live in people's heads and in slide decks. Get the assistant to interview you, one question at a time, and write it up.

I want to write down how my organisation sizes jobs, so that an AI assistant can follow the same method every time. Please interview me. Ask one question at a time and wait for my answer before you ask the next. Cover these, in this order: 1. The factors we use to size a job. Use ours, not factors you already know from other methods. 2. For each factor, what separates one level from the next. Ask me for plain-English differences in the work, not points or scores. 3. How we turn the factor levels into one overall size. 4. What a role description must contain before we will size it. 5. Anything we always ignore, such as the jobholder's performance, length of service, pay or potential. Rules: - Use only what I tell you. Do not add factors, levels or wording from any other method. - If an answer is vague, ask me to describe a real role at that level, with the name and employer removed. - When we finish, write the method up as one clear document with the factors, the level descriptions, the overall rule and the list of things a role description must contain. Then ask me to correct anything that is wrong. I will only share material I am allowed to share.

Step 2: Turn it into a skill

Now ask for the skill file itself. This prompt lists the behaviours that make the difference: the gate, the role-not-person rule, quoted evidence, borderline flags, confidence and a consistency check.

Using the method document we just wrote, create a skill for an AI assistant. A skill is a single file called SKILL.md. It starts with a short header (a name, and one sentence saying when to use it) and then gives the instructions. The instructions must make the assistant do these things, in this order, every time: 1. Gate. Check that the role description covers purpose, main accountabilities, decisions the role makes, who it deals with, and team and budget if any. If something essential is missing, stop and list what is missing. Do not guess. 2. Assess the role, never the person. Ignore anything about the current jobholder's performance, length of service, pay or potential. If the user offers it, say it has been set aside. If the user asks for a different size because of the person, do not change any rating. 3. Rate each factor with the level, a one-sentence reason, and the exact words from the role description that support it, in quotation marks. 4. Flag borderline calls. Say which two levels each one sits between, what would tip it, and whether tipping it would change the overall level. 5. Give an overall level using our rule, plus a confidence rating (high, medium or low) with the reason. Mark the result "provisional" if anything essential was assumed. 6. Consistency check. If I supply roles that are already sized, compare the result with them. If I do not, say the check was not done. 7. Output in this order: a short summary, a factor table, borderline calls, then what was missing or assumed. Use only the factors, levels and rules from our method document. Write in plain English with Australian spelling. When you finish, list anything you had to assume so that I can check it.

Step 3: Test it against roles you already know

A skill is only as good as its results on roles where you already trust the answer. Test before anyone else uses it.

  1. Pick three roles that people have already sized and that you are confident about. Choose a low, a middle and a high one. Remove names and anything about the jobholder.
  2. Run the skill on each, without telling it the existing size.
  3. Compare. Where the skill differs from the people, do not change the answer. Work out which words in the instructions or level descriptions led it there.
  4. Fix the instructions, then run all three again. Repeat until the results sit where you would expect.
Here is my job-sizing skill: [paste the SKILL.md text]. Below are three role descriptions. Size each one by following the skill exactly. Do not ask me for their existing sizes. [paste role description 1] [paste role description 2] [paste role description 3] After you have finished, I will tell you the sizes that people in my organisation gave these roles. For each difference, do not change your answer to match. Instead, tell me which words in the skill's instructions or level descriptions may have caused the difference, and suggest a precise change to that wording.

Step 4: Put it where your team works

Once the skill gives answers you trust, make it easy for the team to use it.

  • In Claude: add the SKILL.md file as a skill. Check Anthropic's help pages for the current steps, because menus change.
  • In Microsoft 365 Copilot: create an agent in Agent Builder and paste the skill's instructions into it.
  • In any other approved assistant: paste the instructions at the start of the conversation, then attach the role description.

Microsoft's documentation says agent instructions can be up to 8,000 characters, and a skill file up to 20,000 characters (Microsoft Learn, checked November 2026). We have not yet tested this skill inside Copilot.

The sample SKILL.md on this page is about 7,550 characters long, counted by us, so it is close to the first of those limits. If you add your own level descriptions it will probably need trimming to fit an agent's instructions.

Good habits

  • Read the evidence quotes. A rating whose quote cannot be found in the role description is a guess.
  • Keep your skill file in one shared place with a version date, and re-run your three test roles every time you edit it.
  • Treat the output as a recommendation. A named person decides and signs off the final size.
  • Give it only the role description. Leave out names, and leave out anything about how the jobholder performs.

4Check the data

What this touches

Role descriptions are about jobs, not people, but they often carry names. Conversations about sizing also drift into the jobholder. Remove names before you start, and keep the conversation on the work.

What is safe

Role descriptions without names are usually low risk. Your organisation's sizing method may be licensed or confidential, so check before you paste it anywhere. Use tools your employer has approved.

Ask before you start

  • Is this AI tool approved?
  • Are we allowed to load our sizing method into it?
  • Who signs off the final size?

Privacy rules for employee records differ between private employers and government agencies, so check with your privacy officer. Start with the OAIC's guidance on the Australian Privacy Principles and on employment and privacy. General information, not legal advice.

5Take it

Download the skill Download the sample role View the captured output

To use it: add SKILL.md to your approved assistant, attach a role description with the names removed, and ask it to size the role. Then replace the factors and levels with your own method.

What next: Back to all use cases