Reaction diary that links skin flare-ups to the ingredients you applied
TL;DR Patchlog is an app for people whose skin reacts to products. You scan what you apply and log flare-ups with a photo, and the app finds the ingredients common to the products that came before each reaction and guides one-change-at-a-time trials. Users pay a small subscription. It sells no products, so its advice has no stake in the answer.
About one in three Indians says their skin is sensitive. A national phone survey of 3,012 people found 36.7% of women and 27.9% of men reporting sensitive or very sensitive skin, and 14.7% of that group named cosmetics as a trigger[1].
When a reaction does turn allergic, the cause is often a single ingredient. At one Indian hospital, 78.2% of patients with cosmetic-linked pigmented dermatitis tested positive to at least one allergen, most often a hair-dye chemical[2].
The trouble is the stack. A typical routine mixes five or six products, people add new ones every month, and who remembers what they put on their face two days before a flare? Nobody.
What it does, in regulatory terms: a personal product and symptom log with ingredient matching. It does not diagnose, grade or treat skin conditions, which keeps it outside the software provisions of the Medical Devices Rules 2017[3]. Confirm that reading with a regulatory adviser.
You scan barcodes or ingredient lists for each product and tap what you used each day. When your skin flares, you log it with a photo and a few tags (itch, bumps, darkening). After several flares the app ranks the ingredients that keep showing up beforehand and suggests a slow trial - drop one product for two weeks, or test a new one on a small patch of skin before full use. A clean export lets a dermatologist see months of history on one page.
I've reframed the seed. It promised continuous AI personalisation of skincare, but quiz-based "custom" regimens already exist, and an AI that judges skin from photos edges into diagnosis. Patchlog keeps the AI to pattern-finding in your own records.
The user pays, with a free log and then ₹149–299 a month for matching, trial plans and exports (my estimate, to test). Today the same buyer spends on products instead, and one customised-skincare brand's serums and regimens run ₹299–999[4]. A second stream is a flat fee from dermatology clinics that want patient histories in a standard format, but only once users start asking their doctors for it.
Ingredient literacy is spreading through free ingredient-decoder sites, which explain what each listed ingredient does[5]. Cosmetics sold in India must follow the Cosmetics Rules 2020, notified in December 2020[6], so ingredient lists are on packs to scan. The DPDP Act now sets clear consent rules for health-adjacent photos[7], which a privacy-first product can lean on.
No public figure exists for a skin-reaction tracking market in India. The bottom-up number is firmer - about a third of adults report sensitive skin in the national survey[1], skewing younger, which matches smartphone users. Treat that as the pool and not as demand. The real test is how many people keep logging after the second week.
Customised-skincare brands run a skin quiz and sells a matching regimen with a free swap after 10 days[4], but they sell the products they recommend. Ingredient decoders explain ingredient lists and have no routine or reaction log on its main pages[5]. Dermatologists do patch tests in clinic, though they rarely see what a patient used in the weeks before. The gap is a neutral diary that connects products, ingredients and flares over time.
| Difficulty | Moderate (32/100) |
| Starting capital | ₹40–75 L |
| Time to launch | 5–7 months |
Building the app isn't the hard part. The real work is an Indian product and ingredient database accurate enough to trust, and a habit loop that keeps people logging when their skin is perfectly fine and they've got better things to do. Wording every screen so it never reads as a diagnosis takes care too.
Probably no device licence, as long as Patchlog stays a log and never names a condition. That reading needs confirming with a regulatory adviser.
Building this outside India? The categories above usually have a local equivalent, so check with your own regulator.
Don't let AI judge skin from photos or name a condition. That edges into diagnosis, which is the line Patchlog is built not to cross.
Want to work out the AI layer for Patchlog? Talk to Yash →
I think this suits a small product team that can build a clean database and write calm, careful copy. It is a poor fit if you plan to make money selling your own skincare. Before you raise a rupee, prove that 30% of new users are still logging at day 60.
Plan for roughly ₹40–75 L, most of it for building the ingredient database, the app and careful regulatory advice.
Probably not, if it stays a log that never names a condition. Get a regulatory adviser to confirm it sits outside the Medical Devices Rules, and follow the DPDP Act for skin photos.
Yes, the problem is global. Check how your country treats health software and get advice before you make any claim about skin conditions.
The free Patchlog validation kit has the line-by-line capital breakdown, the effort scoring, the cosmetics and data-protection checklist, a 14-day validation sprint with pass/fail targets, a 90-day launch plan, 10 interview questions for users and dermatologists, and 5 assumptions to test. Download the kit →
Thinking about building Patchlog or something close to it? The make-or-break work is keeping users logging long enough for the patterns to mean anything, and that's a problem I enjoy chewing on with founders. 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.