How I Set a Price With Nothing to Copy From

The products page on this site doesn’t show a single price. That’s not me being coy about the number. The real one lives on whichever storefront someone’s about to actually pay on, Gumroad or Etsy, because that’s the only place it stays current without me having to remember to update it in two places. But before any of those pages went live, there was a moment for each product where I had to write an actual number down for the first time, with nothing proven to copy it from. I’ve done that now for an Etsy listing, for six Obsidian plugins, and for a batch of boilerplates that never sold at any price I tried. Here’s the process I actually use, not the tidier version that sounds better after the fact.

The anchor I reach for first

When there’s nothing I’m selling against directly, I don’t start by guessing a number. I start from whatever the buyer is already doing instead of buying from me. For a tool that clears a backlog of unread articles or unwatched videos, the real comparison isn’t a competing plugin, it’s whatever the person is already comfortable paying somewhere else to make a different annoyance go away: a subscription, a service, anything in the same category of “something handles my mess so I don’t have to.” If my number reads like more effort to justify than just living with the backlog a while longer, I’ve already lost before the page finishes loading. That’s the anchor. Not what a similar tool charges. What the buyer already pays, elsewhere, to feel the same relief.

Where my own first guess kept going wrong

My mistake was pricing from the inside: how long the thing took me to build, how much I’d want for the hours, what would feel fair to me for the work. The boilerplates are the clearest case of this. I built real, working products over real time, and some part of me wanted the price to reflect that. None of that effort is visible to a stranger looking at a product page for four seconds. They’re not pricing my hours. They’re pricing whether the problem is worth the number next to the button, and the hours I put in don’t enter that math at all unless I’ve made the result of them obvious before the person ever sees the price.

Treating the number as a guess I’m required to revisit

I’ve already written that a price should start as a guess, not a decision made once, when I was setting up the Etsy shop. What I hadn’t worked out yet is what should actually trigger me to go back and look, instead of vaguely intending to “check back in a few weeks” and letting that intention quietly expire. Two different signals need two different fixes. If a listing gets views and almost nobody clicks through, the number has probably been read and rejected before the buyer even saw the rest of the page, and lowering the price first is a guess about the wrong variable, the real problem is likely the title or the photos. If people click through, read the whole thing, and still walk away, the price is a much more likely suspect, because by then they’ve already decided the problem is real and are only weighing whether the ask is reasonable. I was treating both of those as the same signal for longer than I should have, and fixing the wrong one more than once because of it.

What this doesn’t give me

None of this adds up to a formula, and I’m not sure one exists for someone selling things nobody’s seen before, one person at a time. What it gives me instead is a process I can actually point to the next time I have to write a number down cold: anchor to the annoyance being replaced, not to a competitor or to my own hours; expect the first guess to be wrong in a specific, traceable way rather than just “too high” or “too low”; and know which of the two signals I’m actually looking at before I touch the number again. The habit of treating price as an experiment was the right instinct from the start. This is the rest of what I needed to go with it, mostly learned from the numbers that didn’t add up to much on the products where I got the anchor wrong.