The situation
Customers landed on the company’s site wanting one thing fast: how much is this going to cost? The answers lived in a big internal price book, so visitors either called and tied up the office or gave up and went to a competitor. The price book was solid. It just wasn’t reachable from the website.
What I built
I built a price-lookup tool that sits on the site and answers pricing questions in plain language, pulled straight from the company’s real price book. When someone asks, it reads the question, searches the price book by meaning rather than exact wording, and returns the closest real price. Every lookup gets logged, so the company can see what people are asking about.
It runs on a proper retrieval pipeline rather than a chatbot that guesses, so the answer always comes from their actual pricing and stays accurate as those prices change. I also built in cost and abuse controls, so a public AI tool doesn’t quietly turn into a surprise bill at the end of the month.
The result
Visitors get an instant, specific answer instead of a phone-tag loop or a bounce, and the office fields fewer repeat pricing calls. As a bonus, the company now has a running record of exactly what customers want to know, which is its own kind of market research.
Curious whether AI could pull real weight in your business? Get in touch.