← All use cases
Use case 11

Position in range analyser

See how everyone is paid against the reference point for their role. Set your target range, see how many people sit inside, below and above it, then dig into any level, department or job to find out who and why.

Vibe-coded tool No AI needed to run it For the remuneration team
Try itUse the real tool. Watch itFive steps, animated. Build itPrompts you can paste. Check the dataWhat is safe, and where. Take itDownload the finished file.

What it does and why it matters

Most pay frameworks give each role a reference point and expect people to be paid within a set distance of it, often 10% either side. Whether that is actually happening is one of the first things an executive, a board or an auditor asks, and it is slow to answer from a spreadsheet.

This tool answers it in one screen. It works out each person's compa-ratio (their pay divided by their reference point, so 1.00 means paid exactly at it) and shows the share of people inside, below and above your target range. Change the target range and every chart and table updates.

It shows the ranges themselves, from most senior to most junior, and then people against them. You can look by level, range, department, job family or job profile, switch between five chart types, and add gender, performance rating or time in role when you need them. A custom analysis page counts the people in any compa-ratio band you type in.

Unlike the other tools on this site, it shows individuals by name and does not hide small groups. That is deliberate: it is a working tool for a remuneration team, who already hold this data and control who sees it.

It was "vibe coded": you describe what you want to an AI assistant in plain English and let it write the code. Once built, it is ordinary code. It runs the same calculation every time and uses no AI when you run it.

1Try it

This is the real tool, running in your browser on 1,000 invented people. Change the target range, switch chart types, hover over a dot, or click any row to list the people behind it.

Every name and figure is invented. If you load your own file, it is read by your own browser and nothing is uploaded.

2Watch it

Five steps from a raw file to knowing where to look. Press Play, or step through at your own pace.

Step 1 of 5Invented data
workforce.csv
employee_idlevelrangejob_profilereference_pointbase_salary
E10384Level 4L4.2Project Manager$119,000$114,900
E10041Level 6L6.3Engineering Manager$201,000$206,300
E10194Level 5L5.3Legal Counsel$159,000$153,500
E10889Level 1L1.1HR Administrator$62,000$59,900

Start with a file. One row per person, with their pay and the reference point for their range. A level can have more than one range (L4.1, L4.2, L4.3) because roles at the same level are paid differently in the market.

One person: a Project Manager in range L4.2
$114,900Pay
÷
$119,000Reference point
=
0.97compa-ratio

Work out the compa-ratio. Pay divided by the reference point. 1.00 means paid exactly at the reference point, and 0.97 means 3% under it. It lets you compare a graduate and an executive on the same scale.

Everyone on one scale

Set the target range. It starts at 10% either side of the reference point. You can change either side, and everything recalculates. A wider total range can be switched on when you need it.

1,000 people against a target range of 0.90 to 1.10
71.5%inside the target range
Below: 17.4% Inside: 71.5% Above: 11.1%

Read the headline. Three numbers that answer the question an executive will ask first: how many of our people are paid where our framework says they should be?

Share inside the target range, by department
Marketing
80.9%
Operations
76.1%
Technology
73.2%
Customer Service
60.5%

Find where to look. In this invented workforce, Customer Service has far more people below target and Technology has more above. Click a bar and the tool lists the people behind it.

All figures in this walkthrough are invented and match the tool's built-in sample at its starting settings.

3Build it

You can build your own version with any capable AI assistant your employer allows, for example Microsoft Copilot, Claude or ChatGPT. You describe what you want, it writes the code, and you check the answers.

Build it with invented data, never your real file. Expect two or three rounds of "that's not quite right, change X".

Step 1: Describe the tool

Say what the tool is, what goes in, and who will use it.

I work in a remuneration team in Australia and I am not a developer. Please build me a "position in range" analyser as a single HTML file that I can save to my computer and open in a browser by double-clicking it. Requirements: - It must run offline. Use no external libraries, fonts, images or links, and send no data anywhere. Draw the charts yourself. - The user loads a CSV file with one row per person. Needed columns: employee_id, level, reference_point, base_salary. Optional columns, each switching on an extra view: name, range, department, job_family, job_profile, gender, performance_rating, time_in_role_years. - A level can have several ranges (for example L5.1, L5.2, L5.3), each with its own reference point, because roles at the same level are paid differently in the market. The reference_point on each row is the one for that person's range. - Pay and reference point are annual full-time-equivalent dollars, so do not adjust for hours. - Build in an invented sample workforce of about 1,000 people across 8 levels, with every column filled in and some believable patterns (new starters lower in range, one job family running high, one department running low, a few people far outside). Add a button to download the sample as a CSV so I can see the file layout. - Wherever levels, ranges or job profiles are listed or charted, put the most senior at the top and the most junior at the bottom. Work out seniority from the reference point, not from the level's name. - The audience is remuneration specialists. Use plain English and Australian spelling. Keep the design clean, with large readable headline numbers. Before you write any code, tell me in a few sentences what you plan to build, and ask me about anything that is unclear.

Step 2: Spell out the calculations

Do not let the AI guess the maths. Tell it exactly what to calculate.

Now add the calculations. Follow these definitions exactly. Do not substitute your own. 1. Compa-ratio = base_salary ÷ reference_point. Show it to two decimals (for example 0.97). 2. Target range: two settings, "% below the reference point" and "% above the reference point", both starting at 10. A person is "in target" if their pay is from (100 − below)% to (100 + above)% of their reference point, with both edges counted as inside. Otherwise they are "below target" or "above target". Every chart, table and figure must recalculate when either setting changes. 3. Total range: an optional wider range, starting at 80% to 120% of the reference point, that I can switch on or off and edit. When it is on, also count the people below its minimum and above its maximum. 4. Headline: the % of people in target, below target and above target, with the headcount for each, plus the average and median compa-ratio. 5. Range structure view: a chart of every range, most senior at the top, showing the reference point, the target range and (when switched on) the total range. Use a scale where equal percentage widths look equal at every level. Add a table of the ranges with the % step up from the matching range one level below. 6. Position view: let me look by level, range, department, job family or job profile, and switch between five charts: box plots, one dot per person, a distribution of compa-ratios, 100% stacked bars of below / in / above, and a heat map of level against department. Shade the target range on every chart that has a compa-ratio scale. Under the chart, show a table for the chosen grouping with headcount, average, median, lowest, highest, and % below / in / above. 7. Lenses: gender, performance rating and time in role (under 1 year, 1 to 3, 3 to 5, over 5). Each is off to start with. Switching one on adds a filter, a summary table, and the option to colour and split the charts by it. Only one lens colours the charts at a time. 8. Custom analysis: let me add any number of compa-ratio bands (for example 0.97 to 1.02). For each, show the number of people, the % of the people currently filtered, and a button to list them. Include both ends, using compa-ratios rounded to two decimals. 9. Filters for level, department, job family and job profile that apply to everything. 10. Add buttons to download any table, and any list of people, as a CSV. Explain each definition in one plain-English sentence on the page.

Step 3: Be deliberate about who sees what

This tool shows individuals. Say so on purpose, and add the protections that still matter.

Now set the rules for showing people. - This tool is for the remuneration team only, who already hold this data. Do not hide small groups and do not apply a minimum group size. Show a clear notice near the top saying the tool shows individual pay and should only be shared with people allowed to see it. - When I hover over a dot, show that person's name, ID, job profile, level, range, pay, reference point and compa-ratio. When I click any group, bar, cell or band, list the people in it. - Add a "Hide names" switch that replaces names with the employee ID everywhere, for when I am sharing my screen. - Add a print-ready summary that shows group figures only and never lists individuals, because that is the version most likely to be forwarded. - In every CSV download, put an apostrophe in front of any text that starts with =, +, - or @, so a spreadsheet cannot run it as a formula. - Do not save anything. Do not use cookies, local storage or any network request. Nothing may leave the computer. - Show a clear line near the top: "Your file is read by your own browser. Nothing is uploaded or saved." When you finish, list every place in the code that could store or send data, and confirm there are none.

Step 4: Make it prove itself

Load a tiny file where you can work out every answer by hand.

Please help me check that the calculations are right. 1. Create a test CSV of exactly 10 people, all with a reference point of $100,000, and these salaries: 85,000; 89,900; 90,000; 95,000; 100,000; 100,000; 105,000; 110,000; 110,100; 125,000. 2. Before I load it, tell me what the tool should show for: the number below, in and above target at 10% either side; the same at 5% either side; the average and median compa-ratio; the number above a total range maximum of 120%; and the number in a custom band of 0.97 to 1.02. 3. If any result does not match a hand calculation, say so plainly and fix the calculation. Do not adjust the test data to make it fit.

The answers you should get

Invented example
CheckBelowIn targetAbove
Target range 10% either side262
Target range 5% either side343
Average compa-ratio: 1,010,000 ÷ 1,000,0001.01
Median compa-ratio: the middle two are both 1.001.00
People in the band 0.97 to 1.022

Two salaries are there to test the edges. $90,000 is exactly on the lower edge at 10%, so it counts as inside. $89,900 is just under, so it counts as below, even though its compa-ratio shows as 0.90 when rounded.

If the tool gives a different answer, it is wrong. Tell the AI what you expected and ask it to fix the calculation.

Good habits

  • Test the edges. Most mistakes in range analysis are about whether "exactly 90%" is in or out.
  • Write down your definitions (which pay figure, which reference point, how part-timers are treated) before you compare two reports.
  • Keep the prompt you used so you can rebuild or audit the tool later.
  • Get someone else to check the numbers before anything goes to an executive or board.

4Check the data

What this touches

Named individual pay, plus gender and performance ratings if you switch those lenses on. This is among the most sensitive data HR holds. The tool does not hide small groups, so anyone who sees the screen can see individuals.

What is safe

The finished tool runs in your browser and nothing is uploaded. Keep it and its downloads inside the remuneration team. Use "Hide names" when sharing your screen, and the print-ready summary when something needs to go further. When you build your own, give the AI assistant invented data only.

Ask before you start

  • Who is allowed to see individual pay here?
  • Where will the file and any downloads be stored?
  • Is this AI tool approved for our data?
  • Does our privacy officer need to know?

If people outside the remuneration team need this view, build a version that hides small groups, as the pay gap analyser does. 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 tool Open it in a new tab

It is a single file. Save it anywhere and double-click to open it. The sample file can be downloaded from inside the tool.

What next: for a wider check of pay across the organisation, try the remuneration health check, or go back to all use cases.