Using GenAI innovation to help keep industries up and running

Experiencing roadblocks with B2B e-commerce search

When it came to executing their e-commerce strategy, Grainger found itself in the same boat as many B2B distribution companies. Their extensive product catalog of 2.5 million products was challenging for sales teams and customer service agents to navigate. With millions of customers, the distribution giant needed product search capabilities that could understand a buyer’s role as well as the part’s industry application. For example, an electrician searching for “clamps” expects different results than a machinist. It was also a challenge for Grainger’s IT teams to maintain and locate the detailed information needed for each product listing.

Aside from supporting buyers across different industries, Grainger also handles a diverse range of customer interactions. Since their customer base includes non-specialists who may lack detailed technical knowledge, their e-commerce search engine must interpret a wide range of queries, including incomplete queries or ones lacking context. Given the vast array of product options, liability is also a concern — Grainger doesn’t want to be held accountable for a mistaken purchase that doesn’t perform as expected. They knew that the complexity of their distribution model would require advanced technologies for managing an extensive inventory with millions of SKUs.

Grainger wanted to implement an intuitive search system that enhanced the ability of call center agents to quickly and accurately support customer requests. According to Ranga Raghunathan, Director, Applied Machine Learning at Grainger, ”If you have different purchase personas, you could be serving a customer who doesn’t know how to describe the technical specifications of the product. So, when they call in and want to make a purchase, they can’t necessarily look up the specifications on our website, which adds to the workflows of our sales and customer service teams.”

The scalability of Grainger’s current search system and real-time data syncing also posed challenges, especially with over 400,000 product updates occurring daily. They wanted to ensure that inventory data presented to customers was consistently accurate to maintain trust, build brand loyalty and avoid potential issues caused by outdated information.

Using Databricks Mosaic AI to turn product discovery into a customer-focused channel

Databricks provided Grainger with a data intelligence platform that transformed the approach to managing their vast product catalog. By enabling Grainger to customize their AI and machine learning pipelines, Databricks Mosaic AI tools provided the foundation of the MRO distributor’s new search function, powered by retrieval augmented generation (RAG). They could now retrieve relevant results for call center agents even with the diverse range of customer queries the company receives daily.

With Databricks Vector Search, Grainger had secure access to comprehensive data engineering and management features specifically designed to support RAG applications within a data intelligence platform. Stated Raghunathan simply, “We chose to work with Databricks Mosaic AI because it gave us flexibility in how we do vectorization and embedding.” The Databricks Data Intelligence Platform facilitated an efficient workflow that streamlined the entire data management process and minimized errors — from data extraction and cleaning to transformation and loading to vectorization.

Databricks Model Serving, a unified interface for managing multiple large language models (LLMs), enabled Grainger to easily switch between different LLMs and query them through a single API. The flexibility to experiment with and optimize these models in real-time applications allowed Grainger to significantly improve the efficiency and accuracy of their GenAI-driven search applications.

Explained Raghunathan, “We want to select the best component for each stage, from the best LLM that works with the task, to the best ETL to the best vector index. Having all this orchestrated within Databricks makes it very easy for me to do.”

Another benefit of deploying LLMs within the Databricks environment was performance. Raghunathan explained, “Most vector search services have some latency. We haven’t seen any of those challenges with the Databricks Platform so far. Being with Databricks means that I have unfettered access to the platform at all times. With the scale of products and daily product updates we deal with, you can imagine the benefits of a unified solution.”

The Databricks suite of RAG tools has refined the interaction between Grainger’s LLMs and their extensive product database to deliver high-quality, accurate responses. Now, with Vector Search automating the synchronization of product data from the source to the search index, Grainger can support high volumes of product embeddings and real-time queries. Additionally, the automation of data vectorization and embedding processes ensures product indices are always accurate, eliminating the need for manual updates and complex data pipeline maintenance. By integrating contextually relevant data into the model outputs, Grainger’s RAG application leverages LLMs to ensure that customer inquiries are met with precise, contextually appropriate responses.

The GenAI capabilities of the Databricks Platform also enabled Grainger to enhance product discovery even further through conversational interfaces. Even while supporting multiple search modalities and thousands of real-time queries, Grainger’s GenAI models provided accurate and near-instantaneous results. Furthermore, the Databricks security and governance framework seamlessly integrated with Grainger’s existing protocols, safeguarding sensitive data while maintaining compliance with enterprise-level standards.

Equipping customer service and sales teams for success

Grainger’s deployment of the Databricks Data Intelligence Platform dramatically reshaped their approach to managing an extensive product catalog. The impact of this integration has been profound, with significant advancements in search recall and discoverability across the company’s 2.5 million products. The solution further empowered sales teams and call center agents with faster and more accurate product retrieval capabilities, which saves time, reduces errors and lets employees assist customers more efficiently.

Raghunathan emphasized the capabilities of the Databricks Data Intelligence Platform, noting, “Grainger’s website is high traffic since we are a go-to for many companies. Even with our strong sales force of 4,000 professionals, e-commerce is a main pillar for our business, and that means we have to continuously improve. Updating our tech stack is a must and aligns us with modern methodologies — that’s where Databricks helped us the most.” With Databricks, Grainger enjoys ongoing improvements and customization of their AI and ML applications, keeping Grainger’s e-commerce search at the forefront of the B2B distribution industry.

The scalable nature of GenAI tools from Databricks supports Grainger’s handling of substantial daily updates and changes across hundreds of thousands of product entries, meeting their large-scale enterprise requirements. This ensures that Grainger’s product database remains up-to-date — crucial for maintaining customer satisfaction and responsiveness to market demands. Through these advancements, Databricks gives Grainger the advanced tools they need to make sure buyers can quickly get the parts and supplies they need — no matter the industry.