Every headcount plan starts the same way: a spreadsheet of who's here, what they cost, and how that's changed. Most of that is already in NetSuite, if the account runs payroll, and the rest is in the employee records. What's usually missing is a consistent way to turn it into the numbers a plan needs.

This is the second of two posts on the workforce prompts in my NetSuite AI Prompt Library. The first covered benefits, compensation, expenses, and turnover. This one covers the other four: headcount planning, payroll cost, time and attendance, and productivity.

If you're new to the library, it's a set of 150 prompts for NetSuite finance work that I released in January. Each one is a structured system prompt that you paste into Claude or ChatGPT along with your NetSuite data, and it turns the model into a specific kind of analyst with a defined method, a required output, and rules about what it's not allowed to make up.

Headcount Planning Tool

The headcount prompt turns the model into a workforce planning analyst and works from the employee master outward: a current headcount summary, distribution by department and by job title, a trend from hire and release dates, span of control by manager, and tenure distribution. Six queries, one for each of those.

The thresholds are the ones a planner uses. A span of control of six to ten reports is healthy, over fifteen or under four is a concern. New hires above 35% of headcount is a concern, because it means the organization is mostly people who haven't finished learning the job. A vacancy rate over 10% likewise. The deliverable ends with scenario modeling, which is where the prompt is strict: a scenario is a calculation from stated assumptions, and the assumptions have to be printed with it.

Payroll Cost Analyzer

This is the finance side of the same data. The model becomes an HR finance analyst and breaks payroll cost into base compensation, benefits, payroll taxes, and variable pay, by department and over time. It ships with five queries: cost by department, cost per employee, overtime, benefits breakdown, and trend.

The variance thresholds are tighter than for most prompts in the library, because payroll is the largest expense line in most businesses and a small percentage is a large number. Month-over-month movement over 10% is high risk, budget variance over 7% likewise, and a change in cost per employee over 15% is a flag. The output is a cost breakdown, a trend, a variance analysis, and a forecast that has to show its assumptions.

Time and Attendance Analyzer

The time prompt reads time entries and looks for the things that create cost and compliance risk: hours by employee, attendance patterns, overtime, paid time off, and weekly hours compliance. Its thresholds are about people, not money. Weekly hours over 55 are high risk. Overtime above 20% of hours is high risk. Ten or more consecutive working days is high risk.

I think that this is the prompt in the set where the human-review rule earns its keep most clearly. The data can show that someone worked 58 hours for three weeks running. It can't tell you whether that's a problem, a project, or a mistake in the time entry, and the prompt is written so that the model reports the pattern and stops.

Workforce Productivity Analyzer

The last prompt puts the workforce against the business. Revenue per employee, labor cost as a share of revenue, transactions per employee, and department-level comparisons, with a trend over time. Five queries, all joining the employee data to the transactional data that the rest of the library works with.

Its thresholds: revenue per employee growing more than 10% a year is strong and shrinking is a concern; a labor cost ratio under 30% is strong and over 50% is a concern; transactions per employee above 100 is strong. Those are generic, and the prompt says so. The useful version of this analysis is the trend within one business, not the comparison to a number from somewhere else.

A Note on Sample Runs

As with the first four workforce prompts, I haven't run these against my test account, and the reason is the same. The account has employee records but no payroll transactions and no time entries, so three of the four prompts would report that their required data is missing, which is what they're supposed to do and isn't worth a picture. The productivity prompt could run on the revenue side, but dividing revenue by 31 demo employees produces a number that means nothing. If your account carries payroll and time in NetSuite, you have what these need.

Who They're For

Finance leads building the annual plan. Operations managers who own overtime. HR partners who want the numbers before the conversation. And anyone who has been asked whether the company is over- or under-staffed and would like to answer with something other than a feeling.

All four are in the NetSuite AI Prompt Library, under HR and Workforce Analytics.