Headcount is usually the largest number on the plan and the one modeled with the least care. For most companies it is somewhere between sixty and seventy percent of operating cost, and yet it is often planned as a single blunt calculation: a headcount number times an average salary. That shortcut is where most headcount plans quietly go wrong.
Why "FTEs times average salary" breaks
An average hides everything that actually matters about people cost. It assumes a role approved in the plan starts on day one of the year, when in reality it starts in May, if the requisition fills at all. It treats a two-hundred-thousand-dollar staff engineer and a sixty-thousand-dollar coordinator as interchangeable units of "one head." It ignores the loaded part of the cost, the payroll taxes, benefits, and fringe that turn a salary into the real number that hits the P&L. Plan on averages and you will be wrong on both the amount and the timing, which are the only two things a headcount plan is for.
A headcount plan is not a count of heads. It is a schedule of specific people, each with a start date and a fully loaded cost.
What a real headcount model tracks
The teams that plan headcount well stop counting heads and start scheduling people. Every planned role carries a start month, so cost lands when the person actually joins and not on January 1. Every role carries a fully loaded cost, salary plus the taxes and benefits on top, so the number in the plan is the number that hits the budget. Backfills and attrition are modeled as their own events, because a role that turns over in September is not the same as a role that sits filled all year. And an open requisition is treated as what it is, a plan, distinct from a person who has actually started, so your headcount count and your headcount cost never quietly disagree.
What headcount typically represents for a mid-market company
The largest line on the plan deserves more than an average. Small modeling errors here move the whole forecast.
Why the timing part matters most
Of the two errors, timing is the one that compounds. A plan that assumes every approved role fills immediately overstates cost early and then looks great when hiring lags, which trains everyone to distrust it. A plan that models realistic start dates, with the gaps between approval and a filled seat, gives you a forecast the board can actually believe. The point is not to plan more pessimistically. It is to plan for how hiring really happens, which is unevenly and later than the plan hoped.
Getting this right in a spreadsheet is possible and painful, because every start-date change means re-rippling loaded cost across twelve months by hand. Making that automatic, per position, is a lot of what Novaplan's headcount planning does, and it is worth seeing against your own roster.