Technologies used

Google Dialog Flow

Google Dialog Flow

Google Cloud Platform

Google Cloud Platform

Google App Engine

Google App Engine

Google Cloud Storage

Google Cloud Storage

Google Cloud Functions

Google Cloud Functions

Google Compute Engine

Google Compute Engine

Cloud VPC

Cloud VPC

Stackdriver

Stackdriver

Pub/Sub

Pub/Sub

Docker

Docker

Node.js

Node.js

Kubernetes

Kubernetes

Chatbase

Chatbase

Financial Services

Customer service chatbot optimization for a leading bank

Impact by the number

 

50%

improvement in customer journey completion rate

80%

improvement in intent understanding accuracy

Financial Services

Customer service chatbot optimization for a leading bank

The challenge

A large well reputed bank wanted to improve the operability of their customer service chatbot. They wanted to implement changes to it that would result in improved conversation flows, better integration with internal systems and better scalability. Their current chatbot was not properly optimised and was resulting in performance losses and low NPS scores from customers for the bank.

The solution

Addo assessed the requirements listed and presented a solution that was in line with our client’s expectations. The solution consisted of 3 parts:

  1.  We first conducted a current state assessment of the chatbot to gauge what changes were needed and how we were to optimize it for our client.
  2. We proposed an optimization strategy for their chatbot which focused on addressing the major pain points of the bank and ensuring easier management for both technical and business users, making it less error prone in terms of conversational flow.
  3. And finally, we provided a Future State Architecture, to give an overview of what the future state of the chatbot would look like and to describe its salient features.

Our most vital solution component, which was the optimization strategy, consisted of two major streams of work:

  • Dialog Flow Optimization: This involved a complete optimization of the Dialog Flow based chatbot implementation including the redesigning of conversational flows according to the Dialog Flow best practices, documenting them through the use of flow diagrams with the business team, and providing training to the business team on how to create detailed conversational flow diagrams for Dialog Flow by making use of best practices.
  • Backend and Code Optimization:  This consisted of enhancing the application’s architecture to include logging and monitoring in order to provide visibility into the chatbot performance to business, documenting and standardizing coding practices to ensure a consistent code base, enhancing security posture of the chatbot, and finally upskilling and training the chatbot’s technical team.

For this project, a team of highly qualified NLP Engineers, Data Scientists, Infrastructure Engineers and Cloud Engineers at Addo was engaged.

AI Techniques: Natural Language Processing, Sentiment Analysis

The results

  • Improved customer satisfaction
  • Increased user engagement
Looking for a
similar project?
Let's talk

Related case studies

Financial Services

Micro-lending using alternative data for a multi-national bank

Read more

Financial Services

Prediction of risky driving behaviour in the elderly

Read more