Unilever needed to understand complex biology, microbiome behavior, and underserved consumer cohorts more effectively so it could develop differentiated beauty and personal care products faster and with fewer physical experiments.
Reported impact
Unilever reports faster, smarter innovation, fewer required experiments, differentiated products for underserved consumer groups, premium product development, and growth in categories such as whole-body deodorant.
Unilever uses an in-house R&D stack built around AI, machine learning, big data, the Polycultural Centre of Excellence (PCOE), and measurement technology to turn biological and consumer data into new product formulations. Teams capture clinical and consumer data on melanin-rich skin tones, textured hair, sweat, wetness, and the skin microbiome, then aggregate those large datasets for analysis. AI is used to interrogate complex biology and microbiome data, while machine learning mathematically models results that would previously have required many more physical experiments. Insights from the PCOE and microbiome analysis are applied to launches such as Vaseline Radiant X, Dove and Shea Moisture products, and whole-body deodorants. Sophisticated digital tools also supported development of Dove’s MicroMoisture technology, and digitally integrated Unilever plants manufacture the resulting complex formulas on Unilever sites.
Clinical measurements and consumer-relevant observations from melanin-rich skin, textured hair, sweat, wetness, and skin microbiome studies are gathered through the PCOE and other R&D programs. Unilever aggregates these datasets, uses AI to interrogate complex biology patterns, and applies machine-learning models to infer results beyond the experiments it physically runs. Scientists then translate those insights into formulations and process designs that are manufactured in Unilever’s digitally integrated plants and released through brands such as Dove, Vaseline, Rexona, and Shea Moisture.
“At the PCOE we’re entirely focused on studying and capturing data around melanin-rich skin tones, textured hair and what consumers from these under-served cohorts really need and want from their products.”
“In the past we might have had to carry out 1,000 experiments to get the data we needed for a new ingredient. Now we can do a fraction of that – and machine learning helps us to mathematically model the rest.”
“Our plants are fully integrated to make sure they can manufacture this complex formula, and Unilever is the only place where MicroMoisture is made.”
“The AI technology we have allows us to interrogate complex biology data to find that ‘butterfly’ that allows us to invent new technology.”
“The massive amount of microbiome data we can mine using AI and other tech allows us to examine how sweat and wetness affect other areas of the body.”
“With discoveries like machine learning and AI, we can unpack and explore enormous sets of data in a way that’s never been possible before.”
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
Related
In-house Sample May 15, 2026 Unilever Foods Unilever Foods needed to keep brands visible in AI-driven search, accelerate food innovation as trends move faster, and deliver tailored recommendations to foodservice operators using real operational context rather… 10-position boost in visibility rankings AI search visibility platforms · AI-powered product development · UFS digital experience In-house Sample Feb 23, 2024 Unilever Unilever wanted to transform legal service delivery so its global legal team could operate in a more efficient, simplified, digitally enabled way and improve operational effectiveness while supporting the business. generative AI tools In-house Sample Oct 21, 2021 PepsiCo PepsiCo needed to use large, diverse data sets across farming, manufacturing, merchandising, consumer insight, and product development to improve forecasting, product quality, maintenance, sustainability, and speed of… less than 12 months 