Case Study: Complex Conversational Platform


A leading insurance company wanted to increase their volume of sales without expanding their current team. They discovered that their sales team were spending an estimated 30% of their time working on operational queries from customers. Therefore, the client wanted to reduce non-sales queries going to the sales team whilst still getting customers what they needed quickly.

This year, Salesforce found that 69% of consumers prefer chatbots for quick communication with brands. The business realised that AI was the solution to their problem and approach Heron to run some workshops to help to define and design the the AI powered solution.

How did Heron approach this?

To address their complex issue and improve customer interactions, Heron set out to create a chatbot-fronted tool that could use time as efficiently as possible to retrieve answers to customer queries accurately and consistently.

We took their unstructured call and email data and ran it through a combination of AI data analysis tools to review the customer and user feedback to derive patterns and insights from their data. This provided an understanding of the pain points and gaps in the as-is process that could be addressed by AI.

The insights from the data was used to inform the design of a conversational platform. A full review of the user experience was performed to design customer-led interactions, rather than strict linear conversational flows. The platform was iteratively tested and validated with the end users to help drive a high adoption rate.

Heron worked alongside the client team to deliver demos of new customer-led interactions and new AI technologies that would address pain points, which Heron researched and tested with user groups in order to iterate and demonstrate the art of the possible with AI, whilst upskilling their teams internally on how to design and test AI solutions.  

Technologies Used

  • Decision Management
  • NLP/NLU Chat interface
  • Machine Learning

The Outcome

We combined an automated decision-making model with NLP technology to create a tool capable of handling operational queries from customers. This combination of NLP and automated decision-making is able to process both simple queries and more complex cases, and defers to human intervention where necessary.

The client team members were upskilled in the processes and technologies used throughout the process. We also defined a clear AI roadmap to guide the client on next steps in order to add value quickly and efficiently using existing and new capabilities.

The outcome was that the tool freed up time for the sales team time to focus on what they do best: selling their product to market. The sales team are now able to focus on making sales and increasing revenue for the company.