Hiring vs. Automation: The $75,000 Crossover Point
Executive Summary
"Automation carries a high upfront cost and low marginal cost. Humans are the reverse. Here is how to find the crossover point for a specific role."
"Should we hire someone or automate this?" sounds like a philosophical question. It's actually an arithmetic one — a matter of comparing two different cost curves and finding where they cross.
The Fully-Loaded Cost of a Hire
A $50,000 salary never actually costs $50,000. Once payroll taxes, benefits, equipment, software seats, management overhead, and recruiting cost are factored in, a commonly used rule of thumb multiplies base salary by roughly 1.25-1.4x to estimate the true annual cost — putting that $50,000 hire closer to $62,500-$70,000 in fully-loaded terms. Humans also come with a relatively flat marginal cost: the tenth task of the day costs about the same as the first, in terms of paid hours.
The Automation Cost Curve
Automation — whether a workflow tool, a script, or an AI-agent-based process — has the opposite shape: a real upfront cost to build and configure it, followed by a much lower ongoing marginal cost per task once it's running. The upfront cost includes the build itself, testing, and the time spent training the system or writing the rules it follows. The ongoing cost includes subscription or API fees, and — this part gets skipped too often — the human oversight still required to catch errors, handle edge cases the automation wasn't built for, and update it as processes change. Automation is rarely "hire and forget."
Finding the Crossover Point
The crossover point is where the automation's upfront-plus-maintenance cost, amortized over time, drops below the fully-loaded cost of a human doing the same work. High-volume, repetitive, rules-based tasks tend to cross over quickly, because the upfront build cost gets diluted across a large number of repetitions. Low-volume or judgment-heavy tasks may never cross over, since the build cost isn't spread across enough repetitions to pay for itself, and the human-oversight cost stays high regardless.
Illustrative Worked Example
A role handling a high-volume, repetitive task costs roughly $75,000/year fully loaded. An automated workflow to replace it might cost $60,000 to build and roughly $12,000/year in software, API, and oversight costs. Year one looks close to a wash — $72,000 in build-plus-run cost against $75,000 for the hire — once the build cost is counted. From year two onward, with the build cost already paid off, it's saving roughly $63,000/year relative to the hire — assuming the task stays repetitive enough that the automation doesn't need constant rework. This is an illustration of the shape of the math, not a claim about any specific role.
The honest failure mode on both sides is worth naming: over-hiring for genuinely repetitive work wastes money, and over-automating judgment-heavy work creates quality problems that cost more to fix than they saved. The crossover math only holds up when the task itself is the right kind of task.
Calculated an impact?
Share this article with your team or planning committee.
Continue Reading
Recommended Next