article

Machine learning project aims to find out what angers customers

The project, which is being funded by Touchpoint, will use srtificial intelligence on big data to better understand how organisations can improve customer experience

A new artificial intelligence project aimed at identifying the brand behaviours and practices that irritate customers the most is being launched in Australia and New Zealand. Called Radiant, after a supercomputer created by sci-fi author, Isaac Asimov, in the 1950s, the six-month project is being positioned as one of the largest machine learning projects of its kind. It will simulate millions of customer interactions using big data to better understand what company behaviour patterns are leading to negative customer experiences. The project is being backed by a $500,000 investment from customer engagement consultancy and software vendor, Touchpoint Group. The company has also spent the last two years building the massive data set on customer interactions required. Touchpoint said Radiant will create virtual customer discussions using thousands of different experience variables and a data set based on millions of real life, anonymous interactions between irate customers and large Australian companies. Data scientists in Australia and New Zealand will generate these different discussions using the big data sets to develop more accurate predictive customer behaviour models. “Radiant will be searching for behavioural patterns that typically suggest moments of risk or opportunity when interacting with customers,” explained Touchpoint Group’s CEO, Frank van der Velden. “Effectively, it will be constantly running ‘what if’ scenarios, to see if a particular scenario is likely to enrage or benefit the customer.

“The problem with analysing both staff and customer behaviour is that there are so many different variables that could come into play. Many businesses are often left scratching their heads wondering what went wrong, let alone how to fix it. The end goal of Radiant is to automate identification of these root causes and related issues, and to prioritise and recommend actions across different areas of a business.” Van der Velden told CMO Touchpoint is partnering with a large bank on the project and is looking to do two things: Firstly, to work through what is technically possible, and secondly, to understand commercially where the real value lies. “This is a very practical exercise – it’s about where we can make the greatest gains around improving customer experiences,” he said. Van der Velden cited rising interest across A/NZ organisations in how to lift customer experience. “The challenge for companies with large customer bases and lots of room for improvement is that it can take quite a while – years – to turn the company operationally to achieve an optimal level of performance,” he said. “We’re trying to help tackle a problem for these companies, which is loads and load of data, and use machine learning to come up with recommendation engines they can use, especially at the front-line, to respond to systemic issues quickly.” The more transparent the information is to the frontline and staff, the most things can change, van der Velden added. As well as what angers customers the most, the project will also look at where positive experiences have occurred to better understand instances where companies have delighted or surprised consumers.

“These are often the seeds of opportunities where a business can differentiate itself in the market by consistently delighting customers in ways no easily visible to competitors,” van der Velden said. “One of the key objectives of radiant is to automatically detect these situations, and to allow both risks and customer opportunities to be quickly acted on.” Touchpoint is hoping to use the research findings to extend the capabilities of its enterprise customer experience software offering, TouchpointCX, as well as to deliver non-industry specific benchmarks and best practices around customer engagement. Radiant takes its name from the ‘Prime Radiant’ supercomputer created by Asimov more than 60 years ago, that could predict the future behaviour and development of humanity through the analysis of history, sociology and mathematical statistics. Touchpoint is a NZ-based business that launched in 2001 with a focus on customer experience software and consultancy services. It opened its doors in Australia in 2007 and has a client base including ANZ, Spark, AA Insurance and Westpac.