AI has become integral to our personal and professional lives in recent years. Many of us have found tools that are fun to use and others that significantly boost our productivity. Businesses, too, have recognized that AI should be part of their daily operations, yet many struggle to find the right tools or use cases. Numerous companies have experimented with SaaS products featuring AI, or even attempted to develop their own internal solutions, only to fall short of the expected return on investment. I’d like to share a story about a company and its journey to effectively use AI to lower costs while dramatically improving its customers’ experiences.

Meet InGen Technologies. InGen is a B2B services company that provides various services to other companies and then resells these services to their clients. InGen’s offerings vary by client company and often differ in subtle ways. The details of these services are documented and stored on their intranet in different folders for each client. Doug is one of the many SMEs (Subject Matter Experts) responsible for verifying these details and ensuring the documents stay current. Doug doesn’t have time to talk to clients and answer questions directly; instead, a team of customer service professionals handles questions that come in via phone, email, and their website. Because the customer service team answers inquiries about hundreds of different client companies, they always need to look up the answers in the documents stored on the intranet.

This is where things get interesting. The company’s intranet houses thousands of documents, with many versions of each. Over time, these documents have become increasingly difficult to locate, making it challenging to find the right one. The customer service team is taking longer and longer to answer questions, with the average response time rising to 25 minutes. This is longer than their clients are willing to wait for a response, leading to the hiring of more customer service team members to handle the growing volume of inquiries. The team is expanding rapidly, the average customer wait time is increasing, and they are desperate to find a way to get faster.

The customer service team realizes that if they message an SME to ask where the proper document is, they can save 15 minutes per call. Doug, our SME, finds himself receiving more and more messages. What used to be a once-a-week or biweekly request for help has become an almost daily occurrence. When Doug was asked for help five times in one day, he finally became frustrated—he wasn’t getting his own work done! Doug discusses the problem with the other SMEs, and they decide that additional training is needed. They also agree to refuse requests for help unless the team has thoroughly searched for the answer themselves.

Doug leads a training session with the customer service team, teaching them everything he knows about where the documents are stored and how to search for them quickly. This seems to help initially, but soon Doug notices that the average case times are back over 25 minutes, and customers are growing frustrated. Doug admits defeat, and the SMEs go back to answering a flood of questions. Doug ends up working extra hours just to get his work done.

Doug has an idea: What if AI could be used to help the customer service team find the answers? InGen tries several different SaaS solutions from various vendors, including one that can be added to their Office suite. However, none of these solutions consistently gets the answer right. Once the novelty of using AI wears off, the team reverts to their old ways. Doug realizes that none of these solutions understand their business domain or how the documents fit within it. He reaches out to New Resources Consulting’s AI Solutions Group for help.

NRC designs a solution for InGen Technologies tailored to their specific business and tuned to provide answers quickly. Using the latest Large Language Models from OpenAI and cutting-edge search technology from Microsoft Azure, NRC creates an AI-based search engine for their business domain documents that is more accurate than their SMEs and provides results in mere seconds. Because accuracy is crucial to InGen, the search not only provides an answer to the question, but also includes references to documents on their intranet. When these references are clicked, users are taken directly to the exact location in those documents where the answers were found. This means that the search engine not only quickly returns the needed answers, even when they are found in multiple documents, but also makes it easy for the team to validate the results.

InGen’s customer service calls go from taking 25+ minutes (and creeping higher every day) to less than 5 minutes. Customer service costs decrease, SMEs are less frequently interrupted, and clients receive quicker, more accurate answers!

If you would like to learn more about this real-world use of AI in an enterprise environment, view this case study with more details. Reach out today to learn more about NRC’s AI Solutions Group and the various challenges we can help solve in your enterprise. From AI-assisted workflow enhancements, to AI-powered customer-facing features, we’d love to help!

Cameron Vetter | New Resources Consulting