改善
What we changed, and why
Three revisions to the 500x Tremella Moisturizer that nobody asked for, what prompted each one, and the one change we tested and rejected.

Most brands reformulate quietly. The batch changes, the label does not, and you are left wondering whether the thing you liked is the thing you just bought. We would rather write it down. Here is every change made to the 500x since the first production run, in order, with the reason attached.
Revision 01 — the fragrance came out
The first run carried a trace of fragrance. Not much — well under a tenth of a percent, a soft green note meant to cover the faint cereal smell that rice ferment brings with it. It worked. It also meant a formula aimed at people whose skin is already dealing with late nights and dry offices contained an ingredient with no function beyond how it smelled for four seconds.
We took it out and solved the smell the harder way, by adjusting the ferment fraction and letting the tsubaki oil carry the scent instead. The current formula smells faintly of nothing, which is the correct amount for a product you put on before bed.
Revision 02 — it was too thick, and we had been defensive about it
The original was a firm cream. It scooped cleanly, it photographed beautifully, and internally we liked it. Then we ran a longer wear test through a Tokyo August and the feedback stopped being about the first minute and started being about the twentieth. People were using less than a full application to avoid the residue, which meant they were under-dosing a product built to be used properly.
A texture people ration is a texture that fails, no matter how good the ingredient list underneath it is.
We dropped the viscosity by roughly a fifth, moved a portion of the shea into squalane, and rebalanced the gum system so it still holds a scoop shape in the jar but breaks down faster under a fingertip. Same ingredients doing the same work; less of a film left behind at minute twenty.
Revision 03 — the jar
Not a formula change, but it belongs in the log because it changed the product. The first jar was wide and shallow, which looked good and buried a third of the contents in a rim you could not reach. The current 40mm jar is narrower and deeper, has an inner seal that survives a suitcase, and empties properly. We also stopped shipping the plastic spatula, because everybody used a finger anyway and it was one more thing to throw away.
And the change we rejected
We tested a version with a low-dose retinal added — 0.05%, well inside the range that is comfortable for most people. On paper it made the range stronger. In practice it broke the thing the 500x is for.
The moisturiser is the product you can use every single night without thinking about it, including the night after a peel, including the week your skin has decided to be difficult. Putting an active in it means it needs a frequency, a warning, and a conversation about what else you are using. That is a different product. If we build it, it will be its own jar with its own instructions, and the 500x will stay the one you never have to think about.
Referenced in this note
More notes
覚書


