The Same Script, Two Quotes

I wrote the same kind of script twice, and quoted it two different ways.

The first time was for a coworker. She had to pull reimbursement receipts from a few different channels into one spreadsheet every day, checking them by hand, and asked if I could script it. I did the math in my head: writing the code, handling the formatting, testing edge cases — two or three hours, easily. Going rate for freelance work like this was two or three hundred yuan at least. Before leaving on a Friday I mentioned it, "probably two or three hundred." She paused, then said, "it's just a script, doesn't so-and-so on your team know how to write those too?" I didn't push it. Bought her a bubble tea and let it go.

The second time was for a friend who runs an online shop. She has to reconcile inventory every week, and right now two part-time workers spend three hours each doing it by hand every weekend — call it under a thousand yuan a month in labor cost. This time I'd learned something: before quoting, I asked, "who's doing your inventory checks now, and how much does it cost you a month?" Once I had that number, I didn't price it off my own coding time. I priced it off "how much labor cost this saves you in a year" and quoted two thousand. She didn't negotiate. The transfer came in faster than I expected.

Both were similar automation scripts, and they took about the same amount of time to write. One got called expensive, the other got paid instantly. What was different wasn't the difficulty of the code — it was what I was pricing against.

The first time, my anchor was "how much time did I spend": two or three hours, two or three hundred yuan, seemed reasonable enough. But really I was treating my own cost as what the other person owed me. To my coworker, how long I spent writing code had nothing to do with her — what she cared about was whether this fell into the category of things she should have to pay for at all. Once something gets filed under "just doing me a favor," two or three hundred starts to feel unnecessary.

The second time I asked a different question first: what does this cost you right now, before I even think about my number. Same two or three hours of code — attached to "saving a coworker some manual busywork" it's worth two or three hundred, attached to "saving an online shop owner nearly ten thousand yuan a year in labor" it's worth two thousand. Ten times the difference, and the gap is in the method, not my skill: my hours are my cost, not the other person's benefit. Those are two completely different ledgers, and I used to price the second one off the first.

This isn't a new idea in the industry, either. Patrick McKenzie (known online as patio11), who's written a lot about pricing for developers, wrote a well-known post about twenty years ago pointing out that engineers habitually price based on how long something took them, when what a client is actually willing to pay has nothing to do with your hours — it's entirely about what the outcome is worth on their books. I've been in this line of work long enough that the idea itself isn't news to me. But when it's my own quote on the line, I still catch myself defaulting to counting my own hours first.

This approach isn't a license to charge whatever you want, either. The precondition is that you actually find out what the other side's cost was to begin with — guess without asking and you can miss in either direction. Guess low and you end up pricing off your own hours anyway, which defeats the point of asking. Guess high and throw out a number that scares them off, and you'd have done better just quoting your hours honestly. If I hadn't gotten that "two part-timers, under a thousand a month" number out of my friend, I probably would have defaulted back to two or three hundred on that job too.

There's another boundary that money alone doesn't fix: once someone files you under "just doing me a favor," it's very hard to walk that back later and start charging. My coworker never filed those two or three hundred yuan under "things I should pay for" in the first place — it's not that she was cheap. So whatever I quoted at that point, I'd be arguing with her mental accounting, not with a price. The same piece of work sits on wildly different ledgers depending on who's holding it — someone will happily pay two thousand for it, someone else won't pay two hundred for it. That has nothing to do with who's generous and who isn't; it's that the two ledgers were never the same ledger to begin with. Try to satisfy both at once and you usually end up collecting money from neither side in a way that feels right.

Now when someone asks me to help with something technical, I don't lead with "this'll probably take me a few hours." I ask first: "how are you handling this right now, and roughly how much time or money does it cost you?" Once that question is out, the anchor for the quote has already moved from my side of the table to theirs — and what I'm pricing is what it's saving them.