Two numbers show up next to almost every name in a payroll file, and most tools collapse them into one. That's a mistake — the gap between them is often the most interesting part of the record.
An employee's salary rate is their official, budgeted pay for their position — the number tied to their title and pay grade. It's stable, it's what appears in a job posting, and it's what most public salary lookups show you. It's also, on its own, incomplete.
What someone actually earned in a given year can diverge from their rate for entirely ordinary reasons: overtime, shift differentials, a mid-year promotion that only applied for part of the year, or a temporary assignment to a higher-paying role.
A sanitation worker whose rate hasn't moved in two years might still have earned 30% more this year than last — and that's a completely different story than a raise.
This distinction becomes critical the moment you try to compare two employees, two departments, or two years. If one department relies heavily on overtime and another doesn't, comparing "rates" alone will understate the real cost difference between them.
Consider two employees with an identical posted salary rate. One works a department with minimal overtime; the other works a department that regularly requires extra shifts. Their "rate" is the same on paper. Their actual annual compensation could differ by a significant margin — a difference invisible to anyone only checking the posted pay scale.
Every profile in CompensationAtlas shows both figures side by side, explicitly labeled, for every quarter on record — flagged whenever they diverge by more than a small margin.
Collapsing rate and actual earnings into a single figure is the single most common way public payroll data gets misread. Keeping them separate, and explicit, is a deliberate design choice, not an accident of the data.
It depends on your question — use rate for comparing official pay scales, and actual earnings for understanding true cost or take-home pay.
This varies by role and pension plan; CompensationAtlas shows what was earned, not how it's treated for benefits purposes.
Usually, but not always — shift differentials, temporary assignments, and retroactive pay adjustments can also contribute.
Back to 2016, for every employee and department covered in the platform.