Framing automation ROI: the cost-of-an-employee math

Most pricing conversations stall because the agency talks about hours saved and the client hears a vague promise. The fix is to reframe the whole pitch around a number the client already knows cold: what it costs to employ someone to do the task by hand. A repetitive coordination role, a data entry job, a scheduling function. Every business owner has a rough figure for that in their head, whether it is a full salary, a contractor rate, or the outsourced fee they already pay. Your job is to put your price next to that number and let the gap do the selling.

Start from their number, not yours

Before you quote anything, ask what the task currently costs to run. Sometimes that is a headcount fraction, sometimes it is a monthly outsourcing bill. One coach's client was paying a fixed monthly fee to an outside team just to handle quoting. That number became the ceiling for the whole conversation.

If your solution can't beat it, don't build it. If it can, you now have a client-supplied anchor that makes your price look small by comparison, and you never had to invent a hypothetical.

Convert the task into a fraction of a role

Most automations don't replace a full employee, they replace a slice of one. A project manager spending a third of their week chasing vendor updates by email is effectively renting out a third of a $70k salary to a job an inbox workflow can do. Say that plainly. "This is roughly a third of one salary, recovered" lands harder than "this will save you time." It also sets up the next move naturally, because a fraction of a salary is a real dollar figure you can put a price against.

Price under the wage, not under your effort

Once you have that fraction, price meaningfully below it. The automation has to look obviously cheaper than the alternative of hiring, training, and managing a person to do the same job. A common structure ties fees to a percentage of the savings the client actually books, often in the 10 to 15 percent range, rather than to the hours the build took. That keeps the price tethered to their outcome instead of your labor, and it survives scrutiny because the client can check the math themselves against payroll.

Show the payback window, not just the total

A number alone doesn't move budget. A timeline does. Frame the investment against a payback period measured in months, not years. Something like a three to six month payback reads as low risk even to a cautious buyer, because it fits inside a single budget cycle. If you don't have hard data yet, say so and use a conservative range, a 20 to 40 percent time saving is defensible even from an educated guess, and admitting the estimate is rough builds more trust than pretending precision you don't have.

Don't force ROI onto tasks that don't have it

Not everything is worth automating just because it can be. If a task is already cheap, already outsourced efficiently, or too tangled with edge cases to model cleanly, the ROI case will be weak no matter how you frame it. Skip it. Chasing automation for its own sake burns trust faster than saying no to a bad fit.

Before the next proposal goes out, ask the client one question first: what does this task cost you today, in a number you already track. Anchor everything else to their answer.