AI Case Study

Creating an AI Agent To Enhance Efficiency

Luis Lopez Garay
| 7/9/2025
Creating an AI Agent To Enhance Efficiency

See how our AI transformation team helped a distributor enhance the efficiency and accuracy of its sales team’s part-finding process.

Company

A global distributor with multiple locations throughout the U.S. and internationally sought to enhance its processes.

Challenge

With a small sales team managing hundreds of orders per day across hundreds of thousands of SKUs located in various locations, this distributor faced a variety of challenges, including:

  • Lost sales because of missed or mis-parted orders
  • Customer complaints due to prolonged quotes or misquotes
  • Less time for salespeople to sell because of excessive quote time
  • Prolonged onboarding needed for new hires to navigate products 

Support

In fewer than 20 days, the Crowe AI transformation team went from discovery to deployment using the following timeline:

  • Discovery: Engaged stakeholders across the business and received data
  • Building: Developed and iterated a Microsoft™ Copilot AI agent based on feedback
  • Testing: Engaged IT and the sales team for user accessibility testing
  • Deployment and transition: Provided a working AI agent, published and documented in Microsoft Copilot Studio 

Solution

Our team employed various Microsoft AI tools to build an easy-to-use product search chatbot in the Microsoft Teams™ platform by loading and interpreting both structured and unstructured data from the company.

The chatbot operates in three easy steps:

  1. A salesperson queries for a part in a Teams chat based on specifications needed.
  2. The chatbot returns a list of parts that fit the query with the part number, description, and source link from tens of thousands of PDFs.
  3. The salesperson selects the appropriate part.
Explore AI transformation services
Boost your business with quick wins and tailored AI solutions from our AI transformation team.

Increased efficiency in quoting

Our solution helped the team go from about 15 minutes per search to fewer than two.

Millions of dollars in potential revenue gain

By nearly eliminating missed or mis-parted orders, our solution established a significant potential for revenue gain.

Less onboarding time​

By reducing the reliance on institutional knowledge and addressing a long learning curve, our team helped the company save significant time onboarding new salespeople.

Challenge details

The company’s process for locating a quote or order was inefficient and prone to delays, which resulted in a lack of productivity and a decrease in customer service.

Part information was fragmented across multiple resources, including both structured datasets from the enterprise resource planning (ERP) system as well as unstructured documents such as specification sheets and manufacturer catalogs.

When a customer submitted a quote or order, the sales team began its search in the ERP system. If team members couldn’t find the part, they often resorted to an internet search. However, that approach bypassed the wealth of information available in those specification sheets and manufacturer catalogs.

Solution snapshot

To address the inefficiencies in the part-finding process, the Crowe AI transformation team used the company’s existing Microsoft technology stack and integrated generative AI to identify, extract, and search part information from unstructured documents.

The company provided an export of the parts listed in its ERP system, as well as a variety of PDFs with product information. Our team then used generative AI to extract information and create search indexes. These indexes allowed salespeople to efficiently search for all data in one place.

The results

To provide users with an intuitive and accessible way to search for parts, our team developed a Copilot agent and deployed it on the Teams platform so it would be easily accessible across the organization and could help streamline the part-finding process for the sales team. The agent’s higher search accuracy rate significantly increased efficiency, and sales team members were able to redeploy their time to more value-added activities.

Takeaways 

With help from our AI transformation team, the company:

  • Significantly increased efficiency in the part-finding process
  • Located millions of dollars in potential additional revenue
  • Decreased onboarding time for new sales team members
Microsoft and Microsoft Teams are trademarks of the Microsoft group of companies.

Contact us

Our AI transformation team delivers tailored AI solutions that can help companies solve problems, optimize performance, and transform their business. Contact us today to start your company’s AI transformation.
Luis Lopez Garay
Luis Lopez Garay
Partner, AI Transformation