Recently, The Estée Lauder Companies Inc., a leading U.S. beauty giant, announced its adoption of the enterprise edition of generative AI tool ChatGPT Enterprise. The initiative aims to inspire employee creativity and extract valuable insights to better serve today’s consumers.
Establishing a Cross-Functional GPT Lab
As the global beauty industry evolves, Estée Lauder’s product line adjustments are increasingly influenced by vast consumer data gathered from research and clinical trials.
Jane Lauder, Chief Data Officer, Executive Vice President of Enterprise Marketing, and a member of the company’s board, stated, “We are excited to collaborate with OpenAI. ChatGPT Enterprise allows us to safeguard our most valuable asset—over 75 years of data.”
With ChatGPT Enterprise, Estée Lauder’s teams can process and analyze data at scale. The tool has become an integral part of the company’s workflows. By leveraging over 240 customized GPT models, employees gain more time and insights to focus on their expertise—developing new products and promoting existing ones while staying aligned with emerging skincare and beauty trends.
Raheel Khan, Senior Vice President of Future-Focused Growth Intelligence, shared, “When OpenAI launched ChatGPT, we asked our employees how they might use it. Over 1,000 ideas were submitted.”
This enthusiasm led to the creation of the GPT Lab, a cross-functional team dedicated to experimenting with and developing ChatGPT-based business solutions, referred to as “custom GPTs.”
Charmaine Pek, Estée Lauder’s ChatGPT Enterprise Program Lead, explained, “Our job is to identify high-impact use cases that deliver value across brands and regions. The GPT Lab aims to recognize meaningful application patterns and scale these successful experiences to more brands and regions.”
Through this approach, Estée Lauder drives innovation in AI applications, enabling different departments to optimize workflows, enhance efficiency, and foster innovation, creating greater value across multiple markets and brands.
Above: Insights generated by a ChatGPT model based on data provided by Estée Lauder.
Unlocking Consumer Insights with ChatGPT
In just 10 weeks, members of the GPT Lab developed several customized GPT models, each serving unique purposes and application areas.
One notable application is the “Fragrance Insights GPT.” This model assists the fragrance foresight team in extracting insights from consumer surveys. By analyzing large datasets, it uncovers trends and preferences to help Estée Lauder design products tailored to diverse consumer groups.
Yuan Zhan, Director of the Fragrance Foresight Team, stated, “Previously, we spent hours manually cleaning and organizing data to find insights. Now, with the Fragrance GPT model, we can ask simple questions in plain English and receive immediate answers from the data.”
Additionally, the lab introduced the “Clinical Trial Data GPT,” which quickly extracts insights into the efficacy of skincare products. For example, a simple query can reveal data such as the immediate hydration improvement percentage of Estée Lauder’s Advanced Night Repair Serum from thousands of clinical trial reports.
Other custom GPT models include:
- Copywriting GPT: A tailored writing assistant for various brands, generating detailed, meaningful, and on-brand content for multiple platforms.
- Supplier Snapshot GPT: This model consolidates key information about each supplier, including profiles, Estée Lauder’s procurement history, and other relevant details.
Through these customized GPT models, Estée Lauder efficiently extracts valuable insights from vast datasets, improving workflows, enhancing understanding of consumer needs and market trends, and aiding in the precise development of market-responsive products.
Product-Oriented GPT Model Development
Estée Lauder adopts a rapid iteration and experimentation approach for GPT model development.
In this model, teams create and test prototypes through short, efficient work cycles. This enables them to quickly identify success factors and make adjustments, laying a foundation for scalable applications.
Kingsuk Chakrabarty, Director of Enterprise Architecture, AI, and R&D, explained, “We evaluate GPT models based on the value they bring to the organization and the work required to implement them. Priority is given to high-value models that can be quickly built.”
Within the GPT Lab, each “team” is composed of three key roles: a business user, a subject matter expert (SME), and a technical lead, working together to ensure each idea is grounded in both impact and feasibility:
- Design: The business user defines the purpose, scope, and target audience of the GPT model in a concise, two-page use case brief to ensure a clear direction before development begins.
- Preparation: The SME collects and organizes relevant data to prepare the use case, adhering to best practices for GPT model development.
- Build and Test: The technical lead constructs the GPT model using the dataset and conducts rigorous testing to evaluate its accuracy and consistency.
- Deployment: The full team deploys the GPT model and publishes user guides to facilitate adoption across teams.
- Refinement and Scaling: Using feedback loops, the team iteratively optimizes the GPT model based on its performance, preparing it for broader application and scalability.
Enhancing Creativity and Market Responsiveness with ChatGPT
ChatGPT has not only boosted employee efficiency at Estée Lauder but also unlocked greater creative potential by reducing manual tasks:
- Time Savings: Within R&D and marketing teams, ChatGPT has improved response times by over 90%.
- Faster Market Response: By accelerating data analysis, the company launches products more quickly, staying ahead of rapidly shifting consumer trends.
- Internal Adoption: GPT acceptance is growing internally, with more teams requesting AI integration into their workflows for innovative solutions.
This fast-iteration, high-value-focused GPT development approach enables Estée Lauder to respond flexibly to market demands while unleashing creative potential and expediting product launches.
| Sources: OpenAI, Estée Lauder
| Image Credit: OpenAI, Estée Lauder
| Editor: Liu Jun
