In the latest blog of our Datawatch series, we explore how the beauty industry uses data science to personalise customer experiences, and even save lives.
The beauty industry has always been focused on personalisation. When in-store experiences dominated the retail sector, beauty sales relied on personal recommendations from shop assistants and in-store treatments, with unique experiences for every customer that walked through the door.
Now, every element of the beauty industry is increasingly online, from new digital-only storefronts like Glossier dominating the market to an ever-growing culture of beauty influencers on social media and YouTube.
It’s a shift that’s forced brands to adapt their operating models and explore how they can continue offering personalised experiences in a digital industry. And for many, data science is the answer.
Beauty products shaped by data
Some of the most innovative beauty brands are using data science to guide every part of their products’ journeys, even down to the manufacturing stage.
Using data science, manufacturers can generate the granular insights they need to make more successful products. This includes identifying what fragrance combinations will be popular without testing, and finding the right pigment for skin tones according to ethnicity, age, and structure — and much more.
French cosmetics giant L’Oréal processes more than 50 million pieces of data per day to generate valuable insights for its research and innovation (R&I) department. These insights inform the characterisation and physiochemical definitions of formulas and raw materials for the brand’s products, helping its R&I team closely meet the needs and desires of its customers worldwide.
Other brands such as Charlotte Tilbury are using big data to better understand customers’ behaviours, gathering crucial insights on trends across regions. This intelligence is shared across the organisation to help ensure its supply chain is equipped to meet demand as trends fluctuate, and all its storefronts — whether physical or digital — are optimised to create effortless customer experiences.
The use of data science goes even further than operational optimisations. In some cases, brands are going to new lengths to meet their customers’ needs, using AI models to create hyper-personalised products.
One-of-a-kind products delivered with AI and data science
At first, the use of AI and data science in the beauty industry was limited to luxury brands and products.
OPTE’s Precision System Handheld Makeup Printer was one of the first AI-based products to break headlines. The $600 product uses a blue LED light to pick up post-inflammatory hyperpigmentation, age spots, and sunspots on a user’s face, processes the data using an intelligent AI model, and prints a serum that contains the perfect mixture of pigments for the user’s skin tone.
With such a high price tag, the handheld makeup printer was only available to a limited number of consumers in the beauty market. But since its launch, AI has become far more commonplace in the industry — with consumer brands using the same technology to create more affordable yet equally innovative services.
One of the best examples is Boots’ and No7’s AI-powered Digital Beauty Advisor. The service asks the user to upload a photo of themselves and uses AI modelling to analyse data from their facial features, including radiance, wrinkles, and dark spots. Using these insights, the service generates personalised advice and recommendations for No7 products.
Other consumer brands have even based their entire operating models on AI-powered personalisation. Dcypher is a digital cosmetics brand that offers custom-made foundations for users’ unique skin tones and types — all based on data gathered by AI-powered video analysis.
Life-saving data science innovations
Data science in the beauty industry goes far beyond helping consumers improve their appearances, too. In some cases, data science and AI modelling are saving lives.
While some brands have been using data science and AI to create personalised skin care routines, Google has taken the technology a step further, creating an app to identify users’ skin conditions.
DermAssist asks users to take three images of the skin condition they’re concerned about. The data in these images is then analysed and compared against a database of 288 skin conditions to create a list of possible matching conditions. With this list, users can then seek a medical diagnosis, or treatment for their condition.
Another similar life-saving innovation is L’Oréal’s My UV Patch. This self-adhesive UV sensor uses photo-sensitive dyes to capture data about users’ UV exposure. Worn over five days, the patch transmits data to the product’s companion app on the user’s smartphone, which advises them on how they can stay safe in the sun.
As data science continues to become more widely used across the beauty industry, innovations like these will continue to grow — and ultimately, consumers will win. They’ll receive even more personalised products that enrich their beauty lives, and they’ll gain tools that can help them spot and prevent life-threatening conditions.
For more fascinating insights into the role analytics plays in managing real world problems, check out our Datawatch and Inside Analytics blogs here.