I came to Lovi sideways. I was looking for a non-Yuka scanner that did not also try to sell me a multivitamin or push me toward an affiliate brand line, and the indie-app community kept pointing here. Two weeks in, here is what I learned, and where the gloss on the App Store stops matching what the app actually does.
TL;DR
Sharp scanner, opinionated Fit Score, calm chat assistant, occasional database gaps. The lack of brand partnerships shows up everywhere once you start looking. Worth installing if you want a second opinion on your shelf without an algorithm trying to redirect you toward a sponsored serum.
What Lovi is
Lovi is an iOS app built around three pillars. A face scanner that uses computer vision to flag concerns by facial zone, sebum, redness, texture irregularity, dehydration, post-inflammatory marks. A cosmetics database of more than 500,000 products with a per-product Fit Score that scores how well a formula matches the skin profile you have built inside the app. An AI assistant trained on medical board content that answers ingredient and routine questions and cites back what it pulled the answer from. There is also a face yoga and facial massage video library, which I expected to ignore and ended up using twice a week.
Who it’s for
If you are building a slow-skincare cabinet and want an app that scans your face and your shelf in the same session, yes. If you have specific reactive ingredients you want flagged across new purchases, the saved skin profile makes that fast. If you are post-procedure or in a retinol reintroduction window and want to track barrier recovery against zone-by-zone scan data, this is one of the better apps for that.
Skip it if you want a single red or green verdict on every product. The Fit Score is a percentage, not a binary, and the database explanations are long. Skip it if you are Android-only at the moment. Skip it if you want telederm consults built in, that is a different category of app.
Features that matter
- Face scanner with zone-by-zone reads. The scanner splits the face into regions and scores each one independently. My T-zone and my left cheek got different recommendations across the same scan, which matches what a dermatologist would actually say in clinic.
- Fit Score against a saved profile. The profile asks about reactive ingredients, current concerns, and routine maturity. Scanning a product pulls a Fit Score that weights formula context against the profile, not against a generic safety hazard list.
- AI assistant with citations. The chat layer cites the ingredient functions and cosmetic literature it is pulling from. I asked it about niacinamide and vitamin C together and got a calmer, more accurate read than the version of the question that does the rounds on TikTok.
- 500K+ product database. Wide coverage of Korean, Japanese, US, and European brands. Some indie Korean lines were missing or sparsely populated, which the app handles by letting you submit photos to expand the database.
- Face yoga and massage videos. I expected to dismiss these. I did not. The lymphatic massage series is short, calm, and useful during a fluid-retention luteal week.
My contrarian take
The 100 percent independent claim is the editorial spine of this app and also the part the App Store screenshots bury. Most scanners route you toward a sponsored brand line at the end of every scan. Lovi does not. There is no affiliate link in the product page, no promoted serum at the top of the search results, no nudge toward an in-app store. After two weeks of side-by-side use with Yuka and Skin Bliss, the absence of that nudge is the single biggest reason I would recommend Lovi to a slow-skincare reader. The cost is that the catalog is built on community submissions and brand cooperation rather than retail data feeds, which means coverage is uneven in some indie niches. I would rather have uneven coverage and editorial independence than complete coverage and a referral fee in every scan.
Real-world test
I tested Lovi for 14 days starting in early May, during a planned retinol reintroduction after a six-week pause. I had been off tretinoin since a March flare, and I wanted to track whether the cheek erythema was actually fading week-over-week or whether I was wish-thinking it. I scanned my face every morning before any product, in the same north-facing bathroom, with the curtains half-drawn so the light was diffuse rather than direct.
The scanner picked up a clear downward trend in the redness score over the first nine days, then a small spike on day ten that lined up with the first 0.025 tretinoin night. The Fit Score caught one product in my cabinet that the saved profile would have flagged at purchase if I had used the app then, a benzoyl peroxide spot treatment whose pH was incompatible with the niacinamide serum I had been layering. I asked the AI assistant to explain the conflict and got a clean three-paragraph answer with citations. The face yoga sessions, twice a week, were a real surprise. I do not know if they did anything biomechanically. The fact that they made me sit still for eight minutes did do something.
How it compares
Yuka is louder, more app-store-popular, and has been criticized by cosmetic chemists for the same hazard-score methodology that gets covered in the Cosmily review. Lovi’s Fit Score is closer in spirit to Cosmily’s compatibility checker than to Yuka’s hazard read, which is the right direction. Skin Bliss is the closest direct competitor in the AI-skin-analysis bucket, with a similar scanner-plus-database structure, and the comparison there is largely about catalog coverage versus editorial independence. Skin Bliss has the bigger user base. Lovi has the cleaner business model. If you want a teledermatology consult inside the app, neither of these is the right pick. Look at First Derm or Skinive instead.
FAQs
Is the face scanner medically accurate? The scanner is trained on dermatology-board content but is not a diagnostic tool. Use it as a tracking instrument, not a verdict. For suspicious moles or rashes, see a clinician.
Does the Fit Score replace an allergy test? No. It weights known reactive ingredients you have flagged in your profile, but it cannot predict a new sensitization. Patch test new products regardless.
How big is the database actually? Lovi cites 500K plus products. In practice, mainstream Korean, Japanese, US, and European brands are well covered. Some indie lines have sparse ingredient data, and the app lets you submit photos to fill those gaps.
Is there an Android version? Not at the time of this review. Lovi is iOS-first and the Android roadmap has been vague for a while.
Is the AI chat assistant safe to trust for routine advice? For general ingredient context and routine sequencing, yes. For prescription decisions or anything involving a flare-up, get a dermatologist in the loop. The chat itself surfaces this caveat when you push it on clinical questions.
Tool: niacinamide vs vitamin C — which one to pick, and whether you can layer them.