A UK consumer-facing company is working with us to understand how AI technology can work for their business. They're a big company with big numbers on their balance sheet. ubisend had been working with them on internal communication solutions, and it was time for them to test how their consumers interact with the technology.
The plan was to deploy an on-site AI-driven chatbot that sold just one product. It was a complicated product, available in different configurations and specifications. Often, a sales or customer service team member needed to speak to the customer to guide them through the selection process. The cost of the product didn't justify spending much time on the sale.
They needed a logic-based sales decision tree with an output that talked like a human. ubisend to the rescue!
Production ready, niche and highly targeted
The chatbot lived on just one page of their website, the individual product page. It was trained to know everything there is to know about that product, from dimensions and delivery times to costs and specifications. The user could talk to the chatbot and ask any questions they had.
When the user was ready, and at timely parts of the conversation, the chatbot would nudge them into a sales funnel. The sales funnels were customised based on the user's requirements of the product, location and availability. Seamlessly moving from an FAQ-style interaction into a logic- and a flow-based conversation was important for conversion metrics, the user must not feel they were being sold to.
10 days and £3.1 million
The chatbot gathered the user's requirements and guided them through the configuration of the product. It was connected to stock data, and so instantly knew what was/wasn't available and lead times where appropriate. The client was able to prioritise and have the chatbot offer discounts on configurations that had high stock levels.
When the user was happy, the chatbot seamlessly took payment or handed over to the salesperson to take card details over the phone.
The goal of this chatbot was for the client to test if consumers were ready to make purchases by talking to a machine. They quickly found out the answer was yes. We are now working through the rest of the low-hanging fruit in their product catalogue.
Due to the confidential data we help our clients manage, we often work under hefty NDAs. If you’re interested to know more about this project, get in contact, as we may be able to introduce you to the client for a chat.