Product descriptions:
For 135 years, Carhartt has remained steadfast in their mission to serve more industries than any other workwear brand. Guided by their commitment to hard work, fair and ethical business practices and environmental responsibility, Carhartt was focused on improving product quality, enhancing customer experiences and streamlining operations. By adopting the Databricks Data Intelligence Platform, Carhartt streamlined data pipelines with Lakeflow Jobs and transformed their approach to analyzing customer feedback with GenAI. Through automated orchestration of data workflows, Carhartt has seen an 80% reduction in setup and maintenance efforts compared to previous orchestration tools.
Facing data challenges amid shifting consumer expectations
Carhartt has long embraced innovation to uphold its mission as the leading workwear brand. As consumer expectations evolved, they saw the need to better understand and respond to customer sentiment. With over 180,000 reviews on its e-commerce site, Carhartt aimed to leverage natural language processing (NLP) to identify sentiment trends and analyze feedback on fit and quality. The company sought to optimize marketing strategies by ensuring communications aligned with customer shopping preferences.
Despite Carhartt’s commitment to innovation, harnessing customer data at scale proved challenging due to its sheer volume and complexity. Fragmented across multiple systems and formats, this data made it difficult to create a unified customer view. Additionally, Carhartt’s machine learning workflows were slow and inefficient, requiring manual intervention to train and update models. “The whole process was very manual,” explained Christopher Mooney, Manager of Data Science at Carhartt. “With numerous custom sentiment and topic models, we struggled to extract meaningful insights from reviews and understand how shifting consumer demands can impact all aspects of our business from marketing and customer support to demand forecasting and inventory management.”
To address these constraints, Carhartt began exploring platforms like Databricks that boasted seamless data integration, automated workflow management, efficient model deployment, and actionable insights to meet ever-changing customer expectations and operational goals.
Transforming data analytics to improve innovation with Databricks
Carhartt implemented the Databricks Data Intelligence Platform. With Delta Lake at its core, data was centralized for analytics and machine learning through a single interface. This technology investment served as the foundation for Carhartt’s analytics transformation. Carhartt’s product review analytics tool, imported customer review data into Delta Lake for preparation and analysis. With Databricks Lakeflow Jobs layered on top, Carhartt streamlined orchestration, from data preparation and ingestion to processing, with greater efficiency and speed. “The ability to orchestrate workflow runs using the Databricks API has enabled us to automate workflows and integrate with other systems,” said Christopher.
With these new tools, Carhartt could more easily perform sentiment analysis, emotion detection and customer feedback classification using the Llama-3.1-70B model, deployed through the Mosaic AI Foundation Model API, alongside dashboards tracking key metrics. By integrating AI into their analytics pipelines, Carhartt created a continuous feedback loop for data refinement. Stakeholders then accessed these findings through self-service dashboards that visualized trends, such as fit feedback, product quality metrics and sentiment patterns.
For instance, the Product Safety & Compliance and Quality Issue Resolution Team leverages insights from customer reviews to better respond to product issues or potential safety concerns. “The powering of our NLP-based dashboards has made a game-changing impact,” said Shiloh Devore, Associate Product Safety & Compliance Specialist at Carhartt. “We can monitor keywords, identify emerging trends, and mitigate risks more efficiently—enabling quicker responses to product concerns and boosting operational efficiency across Quality, Regulatory Compliance, and Customer Service.”
By adopting the Databricks Platform, Carhartt bridged the gap between their data aspirations and operational realities. “With the help of Databricks, we have fostered a culture of collaboration and innovation across our various business teams,” said Christopher. “With unified insights, these teams can more easily anticipate customer needs, refine product offerings and maintain their commitment to delivering top-notch workwear.”
Meeting customer needs faster with data and AI
Carhartt’s investment in a combination of advanced analytics and GenAI yielded measurable business impact, driving efficiency, cost savings and improved stakeholder satisfaction. With Databricks Lakeflow Jobs, they achieved significant operational improvements, including a 50% reduction in code review time, an 80% decrease in setup and maintenance effort compared to other tools, and the complete elimination of pipeline failures. These gains enabled Carhartt to accelerate feature delivery, reducing the time-to-market from months to just two weeks—allowing for faster innovation and more responsive customer feedback integration.
Carhartt has continued to expand the value of these solutions with their product and marketing teams. The marketing team can now tap into enhanced sentiment and feedback insights to refine campaigns and adjust product descriptions. On the other hand, the product design team can proactively address fit and quality issues — that lead to returns, which are historically a huge burden and cost on retailers.
Carhartt’s advanced analytics capabilities are paving the way for a broader, AI-driven transformation across its operations. With Databricks as the backbone of its data infrastructure, Carhartt is now well-equipped to harness AI and analytics to deliver greater customer value while strengthening its legacy as a leader in the workwear industry.