Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for compensation analysts

Job Architecture Title Cleanup Agent

Reduce titles to a clean family and level structure with pay spreads checked

Data cleanupFinanceHuman resources
See how it worksGet access
Job Architecture Title Cleanup Agent: what goes in, what the agent does and what you get

What it does

Over ten years the company has collected 1,200 titles, many of them near-duplicates like Sr. Analyst, Senior Analyst and Analyst III. This agent reads the job list and groups near-duplicate titles. It maps each group to a job family and level using the architecture rules and the job descriptions. It then checks the pay spread within each mapped title, flagging groups where pay differs too much, and titles that do not fit any family or level. For those it asks the HR business partner for clarity and remaps with the answer. It builds a mapping table with before and after counts. HR approves each mapping before any system changes. Edge case: a title used for different jobs in different teams is split rather than merged.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Cleanup cycle begins 2 USES A TOOL Load titles, descriptions and pay data 3 DOES Group near-duplicate titles 4 DOES Map each group to a family and level 5 CHECKS THE RESULT Does every title fit a family and level? If not: Ask the HR partner and remap using their answer.Back to step 3. 6 USES A TOOL Check pay spread within each mapped title 7 CHECKS THE RESULT Is the pay spread within 40 percent? If not: Flag the group for review and split it if dutiesdiffer. Back to step 4. 8 DOES Build the mapping table with before and after counts 9 YOU APPROVE HR approves each mapping before systems change 10 RESULT Approved mapping table
Read the steps as a list
  1. Cleanup cycle begins
  2. Load titles, descriptions and pay data
  3. Group near-duplicate titles
  4. Map each group to a family and level
  5. Does every title fit a family and level?If not: Ask the HR partner and remap using their answer. Back to step 3.
  6. Check pay spread within each mapped title
  7. Is the pay spread within 40 percent?If not: Flag the group for review and split it if duties differ. Back to step 4.
  8. Build the mapping table with before and after counts
  9. HR approves each mapping before systems changeThe agent waits here for your OK.
  10. Approved mapping table

How it decides

It merges titles with matching duties and level, and flags groups where the pay range spread exceeds 40 percent.

  • Merge titles only when duties and level match
  • Split titles used for different jobs
  • Flag pay spread over 40 percent
  • Never map across job families without HR review

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Pay spread limit (default 40 percent)
  • Architecture framework (default the company's family and level scheme)
  • Review owners (default HR business partners)
  • Cleanup cycle (default annual)

What keeps you in control

It always asks you first

  • HR approves each mapping before any system change

Hard limits

  • Never change titles or pay in any system
  • Never merge on title text alone
  • Keep before and after records

It stops when

  • Done: Mapping approved and ready for system updates
  • Stop: Architecture framework is missing, so hand to HR

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensFrom 1,214 titles the agent groups 380 near-duplicates into 96. 41 titles do not fit a level, and the HR partner clarifies 29. The pay check flags Analyst II in two teams with a 52 percent spread; the agent splits it by duties. The mapping reduces titles to 620. HR approves 590 mappings and holds 30.

More agents for compensation analysts

Research · Decisions

Salary Benchmark Research Agent

Give the client a defensible salary range with the reasons behind it.

9 steps1 check2 jobs
Decisions · Risk management

Pay Equity Check Agent

Pay changes that do not widen unexplained gaps and stay within budget

9 steps2 checks6 jobs
Creating · Executives and strategy

Total Rewards Statement Preparation Agent

Build accurate total rewards statements for every employee that tie to payroll.

10 steps1 check7 jobs
Decisions · Analysis

Compensation Band Maintenance Agent

Propose band updates for lagging roles that fit within the compensation budget.

10 steps1 check6 jobs