Most AI acne apps in 2026 are running computer vision pipelines that nobody outside the developer has audited. That is the baseline. Spotscan+ Coach is the exception. It runs the GEA scale, which is a clinically published acne severity grading method used in dermatology research. The model was trained on more than 6,000 multi-ethnic photos, and La Roche-Posay has put dermatology working groups behind the development. That changes how the score lands. It also changes nothing about the fact that the app’s coaching layer points at LRP products on every recommendation card.
What Spotscan+ Coach is
It is a free iOS and Android app from La Roche-Posay (a L’Oreal Group brand) that asks you to take three selfies (front, left, right) and returns a GEA acne severity grade. The grade slots into a personalized coaching program that mixes daily check-ins, expert-led content (videos, podcasts, articles, more than 500 of them), and tailored product recommendations from the Effaclar and Lipikar lines. The score is not a self-assigned wellness number. It is a clinical scale that dermatologists use in published research, and the app makes a credible attempt to reproduce that scoring at a distance.
Who it’s for
Anyone with moderate acne who wants a baseline severity number they can show a dermatologist without feeling like they pulled it out of an Instagram quiz. Anyone tracking acne treatment progress over months and wanting a consistent grading system across check-ins. Anyone already in the La Roche-Posay ecosystem (Effaclar, Cicaplast, Lipikar) who wants a routine that maps to that brand. Not the right tool if you find product recommendations from a single brand house claustrophobic, or if you want a neutral severity check with no commercial layer. The app is genuinely well-built. It is also unmistakably a brand asset.
Features that matter
- GEA severity scoring. The Global Evaluation of Acne scale is the published clinical reference. The app’s reproduction of it is the closest thing to a defensible AI acne grade currently available to consumers for free.
- Multi-ethnic training data. 6,000-plus photos across skin tones. This matters because most AI dermatology tools have documented failures on darker skin tones, and LRP is one of the few brand-backed apps that has publicly addressed the training set composition.
- Coaching program. Daily prompts, content library, and check-in cadence. The structure is useful even if the brand-tied product recs are not for you.
- Product recommendations. Pulled from the LRP catalog. Effaclar Duo, Effaclar Adapalene gel (where available over the counter), Cicaplast, etc. Useful if those are already in your rotation. Not neutral.
- Educational content library. 500-plus expert-led pieces. Higher signal-to-noise than I expected, especially the videos with practicing dermatologists.
My contrarian take
The GEA score is the genuine asset. The coaching layer is a marketing funnel and it should be read as one. La Roche-Posay is part of L’Oreal Group, which is the largest beauty conglomerate in the world. The Effaclar line is solid in the over-the-counter acne category, but it is not the only solid line, and the app will never recommend a CeraVe SA Cleanser or a Paula’s Choice 2% BHA, both of which would be equally defensible suggestions for the same severity grade. The right way to use Spotscan+ Coach is to take the score, ignore the product recs unless they match your existing routine, and treat the coaching content as broadly useful dermatology education. The wrong way is to let the funnel rewrite your cabinet.
Real-world test
I ran Spotscan+ Coach for 21 days starting in early April, across a luteal phase that produced two predictable jaw breakouts and one less-predictable forehead flare after a week of unusually high pollen counts in my city. The three-selfie capture flow took under 90 seconds each session. The GEA grade came back consistent across check-ins (mild on baseline, moderate during the jaw flare, mild again by day 18), which matched my own visual read.
On day 9, I tested it against a controlled comparison: I took the same three selfies in two different lighting conditions ten minutes apart. The score held within the same GEA grade band, which is the kind of consistency SelfieLog’s lesion count cannot match. The product recommendations were exactly what you would expect: Effaclar Duo, Effaclar Adapalene gel (UK availability noted), Lipikar for the cheek dryness I had logged. I already use a different adapalene script from a derm, so I ignored that card. The content library surfaced a video on cyclical acne that was genuinely well-produced and not a sales pitch. The check-in cadence is the friction point. By week three, the daily nudges felt heavier than the underlying utility justified.
How it compares
SelfieLog wins on photo alignment via Ghost Mode but has no clinical scoring credibility. Spotscan+ Coach wins on severity scoring and educational content but loses on photo consistency and brand neutrality. MDacne is the aggressive funnel: AI grading flows straight into an MDacne-branded treatment kit subscription, which is a different commercial model entirely. TroveSkin tries to score wrinkles, spots, pores, and habits at once, and the acne signal gets diluted. If you want one clinically defensible number for your acne, Spotscan+ Coach is the answer. If you want photo continuity for your own gallery, pair it with SelfieLog. If you want a brand-neutral product layer, pair it with Cosmily for ingredient checks.
FAQs
Is Spotscan+ Coach actually free? Yes. No paywall, no subscription tier I could find. The commercial layer is the product recommendation, not a fee.
How credible is the GEA scoring? GEA is a published clinical acne grading scale used in dermatology research. The app’s AI reproduction is the closest free consumer equivalent I have tested. Not a substitute for a derm visit, but a reasonable baseline.
Does it work for darker skin tones? The training set explicitly included 6,000-plus multi-ethnic photos. This is better than most AI dermatology apps, which have a documented bias toward lighter skin. I cannot personally verify the score quality across all Fitzpatrick types, but the composition of the training data is at least transparent.
Do I have to use La Roche-Posay products? No. The recommendations are LRP-only, but the GEA score itself is independent. You can take the grade and apply it to a brand-neutral routine.
Is my photo data shared with L’Oreal? LRP’s privacy policy applies. Worth reading it directly before uploading clinical-grade photos. The app is well-built but it is still a brand asset, and brand-asset data practices vary.
For the routine layer that pairs cleanly with a clinical acne grade, the Cosmily review covers ingredient compatibility in a way Spotscan+ Coach deliberately does not. The full concern-trackers hub has the rest of the acne and eczema apps tested in this round.