We Built Droplet to Reduce Skincare Trial and Error
Most skincare recommendations start with the product.
Is it popular? Does it contain a trending ingredient? Does it have good reviews?
But they often ignore the person using it.
The same product can work well for one person and cause irritation, duplication, or unnecessary spending for someone else. Skin type, sensitivities, goals, current routine, previous reactions, and even the other products being used all affect whether something actually makes sense.
That is why we started building Droplet.
Droplet is a personalized skincare companion that helps people understand products before they buy or add them to their routine.
Users can scan a product label or ingredient list, understand what the formula is designed to do, check whether it fits their skin profile, compare it with their current routine, and identify possible irritation, conflicts, or duplication.
Droplet also remembers previous products and reactions, so recommendations can improve over time rather than treating every decision as completely new.
We have launched Droplet on iOS and Android, and we are continuing to improve how the app explains formulas and makes recommendations without presenting skincare as an exact science.
One of the hardest parts has been balancing useful personalization with clear and responsible explanations. We do not want to simply label products as good or bad. We want to help users understand why a product may or may not fit their specific situation.
We are currently focused on improving onboarding, product analysis, routine compatibility, and the overall experience of moving from a product scan to a confident decision.
We would love feedback from other founders on:
How would you explain personalized recommendations without overwhelming users?
What would make you trust a skincare analysis tool?
Which part of the product would you test first for retention?
You can learn more about Droplet at:
Skincare is a strong vertical, but also very crowded — most tools compete on recommendations.
The ones that stand out usually tie into:
• personalization depth
• or ongoing routines/results
Right now it feels like a useful checker, but not yet something users rely on long-term.
Curious — are people coming back regularly, or using it once?
Happy to help shape this into something stickier if you’re exploring that.
The trust question seems especially important here because the product is helping someone make a decision, not just giving them information.
Since launching, where have you actually seen users hesitate most — before trusting the analysis, or after getting it when deciding whether to act on it?