Photo-and-quiz skincare matcher for chemists and beauty counters
TL;DR Skinbrief is a tablet and phone tool for independent chemists and multi-brand beauty counters. A shopper answers a short quiz and takes a face photo, and the tool suggests a simple routine from the shop's own shelves and flags anything that needs a dermatologist. Shops pay a monthly fee, and it only works if it lifts their skincare sales.
Skincare advice at the Indian shop counter is often wrong, and sometimes it does real harm. In a 2026 study of 96 patients with steroid-damaged facial skin in western Uttar Pradesh, 64.6% had used steroid creams without a prescription, and pharmacists were the advice source for 33.3%[1]. Almost all of them (95.8%) did not know what was in the product, and melasma was the most common reason for use[1].
That is a hospital sample and not a population rate, but it shows you exactly where advice goes wrong.
Demand isn't slowing down either. The head of one Indian consulting firm values Indian beauty and personal care at $21 billion, growing 10–11% a year, with younger buyers putting skincare first[2]. A good chunk of these buyers still shop at the neighbourhood chemist, where the staff have no training in skin types at all.
First, the important bit - Skinbrief does cosmetic product matching only. It does not diagnose, grade or treat any skin condition. Under CDSCO's finalised guidance, software with a medical purpose counts as medical device software while general wellness tools fall outside it[3], so the product has to stay on the cosmetic side of that line.
The seed pitched AI skin analysis to fix skincare mismatches. Global vendors already sell selfie analysis to big brands, so I narrowed it to a brand-neutral counter tool for the shops where bad advice actually happens.
Here's how it works. The shopper answers eight questions on skin feel, routine and budget, and takes one photo in guided lighting. Skinbrief sorts them into a cosmetic skin type and suggests a cleanser, moisturiser and sunscreen from products the shop stocks. If the shopper reports patches, sudden pigment change or steroid cream use then the tool shows a plain "see a dermatologist" card and suggests nothing for that concern. Staff see the same screen, which quietly trains them too.
The user is the shopper, often a woman in her twenties or thirties. The payer is the chemist or beauty-store owner, at ₹1,500–4,000 a month per counter (an assumption you'll need to test). A cosmetic brand wanting fair placement data could be a second payer, but brand-paid ranking would wreck trust, so keep it out of the recommendations.
CDSCO finalised its medical device software guidance in July 2026, which draws a clearer line between wellness and medical tools[3]. Indian dermatology journals published a run of 2026 studies on steroid misuse, so the issue is sitting right in front of regulators[1]. And phone cameras and on-device models now make guided face photos workable on cheap tablets.
There's no public figure for in-store skincare advice tools in India. The nearby signal is the $21 billion beauty and personal care market growing 10–11% a year[2]. The narrower one is footfall - the 2026 study found 1,731 tubes bought by just 96 patients[1], which tells you how much product moves on counter advice alone.
One Estonian skin-tech firm sells selfie skin analysis with more than 20 skin metrics and lists large global skincare brands and beauty retailers among users[4]. Large Indian beauty retailers and brands run their own online quizzes. None of them is built for a brand-neutral chemist counter or for tier-2 stores, and none puts referring red flags ahead of selling.
| Difficulty | Moderate (41/100) |
| Starting capital | ₹60 L–1.1 Cr |
| Time to launch | 6–9 months |
The software isn't the hard part. Photo models have to work across Indian skin tones and dim shop lighting, which means your own labelled images and a dermatologist adviser. Selling to thousands of small chemists is slow and painful, so distributors or pharmacy chains matter far more than a sales team.
No licence is needed to operate if the tool stays a cosmetic matcher, but staying there is the whole job.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't trust AI to make any medical judgement here. Skinbrief does cosmetic matching only, and it must never diagnose or treat a skin condition.
Want to work out the AI layer for Skinbrief? Talk to Yash →
Here's what can go south -
I think this is a reasonable bet for a founder with pharmacy retail or beauty distribution contacts and a dermatologist willing to advise. If you want to claim the tool "diagnoses" skin then walk away, it's a poor fit. Prove one thing first - that 40 pilot counters raise skincare sales by 10% or more in two months.
Plan for roughly ₹60 L–1.1 Cr over 6–9 months, including labelled images across Indian skin tones and a dermatologist adviser.
Not if it stays a cosmetic matcher. Once it names or grades a skin condition it can fall under CDSCO's medical device software rules, so keep it on the cosmetic side.
Yes, but the line between wellness and medical software is drawn differently in each country. Check your local device and cosmetics rules before you launch.
The free Skinbrief validation kit has the line-by-line capital breakdown, the effort scoring, the cosmetics and software rules checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for shop owners and shoppers, and 5 assumptions to test. Download the kit →
Thinking about building Skinbrief or something close to it? The make-or-break work is staying clearly on the cosmetic side of the rules while still giving advice people can use, and that's the kind of line I help founders walk. If you need a hand, you know where to find me. Talk to Yash →
For illustration only. Names are fictional and ideas hypothetical; sourced facts are real but may be out of date. Nothing here is legal, financial or professional advice, and Yash Khatri is not personally liable for any use of it. Read the full disclaimer.