Home People
Business analytics: the intelligence behind customer loyalty

Business and marketing analytics are powerful tools for organisations who want to keep customers and keep ahead of the competition.

One of the cornerstone rules for any business is to “keep the customer happy”. 

It may be a well-worn motto, but the reality is that if you keep your customers happy then you’re likely to keep your customer.

And from an economic perspective at least, that’s crucial because the cost of attracting new customers is significantly more than retaining existing ones.

Deakin researcher Dr Ali Tamaddoni is a senior lecturer in the Department of Information Systems and Business Analytics in the Deakin Business School (DBS) and he’s an expert in using business analytics to help organisations hold on to their clients.

With research published in Industrial Marketing Management, Journal of Service Research, and Decision Support Systems, he specialises in customer churn management, customer analytics and social media analytics.

Dr Tamaddoni’s expertise lies in the ability to look into the future – not with a crystal ball but with the application of business analytics that can predict customer attrition and provide strategies to prevent it.

Drawn to the area via an academic pathway that includes an undergraduate degree in engineering, a masters’ degree in marketing and e-commerce, and a PhD in quantitative marketing, he says the ability to ‘foresee the future in any area’ is fascinating.

‘I was looking for a topic for my masters’ thesis and came across an article explaining how data mining methods can help us predict customers’ future behaviour (including churn). I was very interested in the topic and with my supervisor’s assistance, went on to design my master thesis around it … since then it has been my main area of research interest,’ he explains.  

The term “customer churn” describes the time that customers decide to stop being customers and predicting this is a crucial measure for any business because it provides the information needed to retain customers.

While it’s true that no business is immune from losing some customers along the way, a sudden or significant downward spiral can create ruinous financial and competitive fallout for an organisation.

‘Being aware of a customer’s churn likelihood  allows a business to keep focused on customer retention as a top priority but implementing an effective retention campaign is complex because it’s dependent on firms’ ability to accurately identify both at-risk customers and those worth retaining,’ says Dr Tamaddoni.

One of Dr Tamaddoni’s projects, now published in the Journal of Service Research, focuses on churn prediction methods, their accuracy, and different models of retention campaigns.

It draws on empirical and simulated data from two online retailers and evaluates the performance of several prediction techniques to identify the best modelling approach.

‘The results show that under most circumstances, a nonparametric method – known as the boosting technique – delivers superior predictability. Further, in cases where churn is rarer, the logistic regression model is most successful. Finally, where the size of the customer base is very small, we found that parametric probability models outperform other techniques,’ he explains.

In a second project currently underway, Dr Tamaddoni has developed a new predication model based on the argument that existing forecasting methods tend to identify churn rather than predict it.

He says that the field of business and marketing analytics has gone through significant changes in the past decade.

‘The analytics and business analytics buzz words have led to many companies investing in, and developing, both commercial and open- source software that makes complicated methods palatable to many marketing/business analysts, even those without strong quantitative background.’ 

While the challenge of business analytics research drives the core of Dr Tamaddoni’s career, he balances it with departmental responsibilities (he is director for Deakin’s Master of Business Analytics program) and enjoys applying his research in the classroom.

‘I teach marketing analytics both at the undergraduate and postgraduate levels. For my classes I use topics related to my research and bring examples and cases from the projects I work/have worked on. This includes bringing real-world data from my projects to design my assignments and class activities.’

Forecasting the future, he believes, is critical across the business context as it helps organisations proactively approach ‘at-risk’ customers with marketing strategies designed to promote loyalty.

‘Such an approach is very important because the cost of acquiring a new customer can be up to five times higher than retaining an existing one. This research helps businesses, including government organisations and charities, to retain the customers and donors more efficiently and effectively thus saving on marketing expenditure.’