A paying user made me rethink what generation credits are actually for
I recently noticed an interesting pattern in AIMakeTattoo.
A user kept working on the same tattoo idea across multiple generations. They were not just pressing generate and hoping for a better random image.
They were changing one part at a time: the main subject, the colors, a supporting element, or the overall proportion.
That made me rethink what people are actually buying when they purchase generation credits.
I used to think of credits as paying for more outputs.
But in this case, the value seemed to be having more room to think through a decision. The user already had a direction. They just had not settled on the exact version yet.
It also made one weakness in the current workflow more obvious.
When someone only wants to change one detail, the whole image is generated again. They might like the composition but want a different color, or like the main subject but want to remove one supporting element. Right now there is no simple way to say, “Keep this part, but change that part.”
So I am starting to think the next useful improvement may be a lightweight variation flow rather than adding more styles.
Something simple that lets a user keep the current idea and ask for one specific change.
I am not rushing to build a full history or versioning system based on one user. But it was a useful signal.
In generative products, maybe credits are not only paying for outputs. Sometimes they are paying for enough attempts to figure out what the user actually wants.
Have you seen something similar in your own product?
Hi Warren, Your AIMakeTattoo note about users buying room to explore, not simply more outputs, shows a concrete product insight turning into a next iteration. I’m building Build Before 2030: a manually reviewed public record where makers claim a Founding 100 number, set one measurable 30-day milestone, and later add proof of what shipped. The first cohort is free. AIMakeTattoo feels like a strong fit. Interested? https://buildbefore2030.com
I like the distinction you're drawing here.
It's easy to think users are buying the output, when they may actually be buying the freedom to explore until they're confident in a decision. Those are two very different kinds of value, even if the interface looks the same.
Exactly. The interface may look like it sells more outputs, while the real value is confidence gained through exploration. That distinction is changing how I think about the product.
That’s the interesting shift. Once users are paying to explore until they feel confident about a direction, the product question becomes less about increasing outputs and more about understanding which part of that exploration process creates the value.
Curious what you discover as more users show this same iteration pattern.
That’s what I’m trying to separate now: healthy exploration versus friction from not being able to preserve the parts users already like. I’m watching the next few user sessions before deciding what to build.
That distinction is probably the key one to watch. The interesting part is whether users are spending more time exploring because they enjoy the process, or because the current workflow forces them to restart when they already have something close.
Curious what patterns you notice from those sessions.