The leading ingredient transparency solution for the cosmetics industry
AI-powered mobile app that helps customers decide which beauty products fit them best
This case study shares how we built an AI-powered iOS app for a cosmetics industry that empowers consumers to easily find safe beauty products that satisfy their specific skincare needs.
Results obtained in numbers
200k+
Product formulas
benchmarked100k+
Happy customers
served100k+
Ingredients
cataloged45k+
Products were
analyzed21k+
Scientific research
sources analyzed
Benefits of the app
A custom
Skincare routine based on a user's unique skin type and needs
Thousands of brands
To compare prices and buy cosmetics from verified retailers
800,000+
Analyzed ingredients to find the one that fits a user's unique skin best
Safety ratings
Powered by real science to buy safe and effective beauty products
Have a project in mind?
Let's talk!
Business context
With thousands of beauty
and skincare products on the market it's easy to be confused about which one is the best match for your needs. With this concept in mind, our client developed the idea to create a unique skincare AI-powered platform.
The main product idea is to leverage AI technology combined with scientific research to check product formulations for toxic or unsafe ingredients
and make independently validated safety guides available to consumers.
Product scope
- Gather requirements to define the main project features and goals
- Create an intuitive admin panel to manage projects effectively
- Build a project architecture to create the highest level of app security
- Provide a fully-functional mobile app to find sustainable and eco-compliant beauty products based on the best science and real-time regulatory resources available
Solutions we provide
01
A highly-skilled team of Onix specialists followed the sophisticated process to build an iOS mobile app powered by AI and data science that helps people choose skin care products based on their age, unique skin type, needs, and lifestyle. The consumer-facing solution also recommends products to users based on their individual needs.
02
The users fill in a questionnaire, and the science-backed AI-powered recommendation engine of the app offers the most fitting products. The app lets users vet product formulations for safety and creates personalized skincare recommendations.
03
The main challenge we faced was data collection and maintaining a vast database. With the help of ML, we created and categorized all possible ingredients used in personal care products.
We made a great effort to
provide the following solutions:
An intuitive admin panel
Managing projects is now easier since all the data is in one place. We built a sophisticated administrative panel that allows administrators to easily manipulate the data, handle it in the most convenient way possible, and organize the data through one simple interface.
Enhanced project architecture
To work with world-famous manufacturers, we built the project architecture so that the app has the highest level of security, usually required by large companies.
Convenient search engine
Using this feature, users can quickly search for products within thousands of brands, compare prices, and buy desired cosmetics only from verified retailers.
Appealing website
We built a website that addresses marketing goals helping our client sell more efficiently and grow faster. The website states its main value proposition clearly and succinctly. In addition, it reinforces the homepage with a video to explain the product features and increase visitors’ time on the page.
Huge ingredient database powered by ML and Data Science
Using machine learning technologies, we enhanced one of the largest ingredient databases in the world that helps to understand products' benefits, toxicity, and other features. We developed a classification system for cosmetic products available on the consumer goods market, and numerous data scrapers focused on obtaining new information from the Internet, scientific literature, and articles in PDF format.
Core technology
Python
Flask
Angular
React
Scrappers
MATLAB
PostgreSQL
MongoDB
Redis
Elasticsearch
AWS
SciPy
TensorFlow
PyTorch
Scikit-Learn
Darknet
Jupyter
Google Colab
Pandas
NumPy
NLTK
Matplotlib
Pillow
Seaborn
And some others
Results
Solutions
Thanks to the results of our collaboration, our client now has the leading ingredient transparency solution for the cosmetics industry.
45,000
To date, the app has analyzed more than 45,000 products and is highly rated by its users.
Sales increase
Now retailers can sell products more effectively based on consumers' specific personal needs, unique skin types, and so on. This solution empowers people to discover safe products that make them look and feel their best!
Use our expertise to build a top-notch
solution tailored to your business needs!
