Analysis Paralysis? Here is a Way-Out!

Many organizations have been overwhelmed with the amount of data that suddenly became available to them and they feel a bit paralyzed by it. How do you think through the clutter and use the data in a smart way to deliver on a few critical customer, employee and internal journeys ?

A good way to start, is to familiarize yourself with a few examples of how the usage of analytics can help grow the business by properly identifying your customers’ potential, increase customer satisfaction from the ATMs usage, understanding strong and weak points of your mobile application or even predicting which branches are going to fail the next audit?

Identifying Customer Potential

A friend of mine, who lived in Asia, kept two separate banking relationships. One bank was a recipient of his monthly salary and that was also the place where he kept vast majority of his savings. He had his credit card, however, in the other bank.

In order to make the card repayments smooth and easy, he set up a standing instruction to send the money from his savings bank to his credit card bank every month and it was automatically repaid this way.

It was very obvious for anyone who would analyse his credit card transactions that he was not an ordinary mass customer. Both the amount spent on a card, as well as the types of transactions, were saying a lot about his capacity to spend money and his lifestyle which was on a high end.

Therefore, he was a perfect customer to be offered a Premium or Private bank offer, that would for sure attract more business to the bank beyond his credit card. Nobody has ever contacted him though, and after a few years he closed his credit card and stayed with just one bank.

Why does such a situation happen? The bank had all the information that was necessary to identify an affluent potential in this customer, but it had never done anything. It all boiled down to lack of knowledge on how to use the data at hand in a smart way. The definition of the “affluent customer” was solely dependent on the amount of deposits and investments kept in the bank and it never looked at any other product that the customer was using. Lost opportunity? For sure. But also, a new opportunity to explore going forward by changing the mindset and opening ourselves to new ways of thinking and using the data.

ATM Optimization

 In contrast, I saw a good example of using data to analyse the usage of ATMs, making sure they never run out of cash. One of the Asian banks always had big queues at their ATMs and often its customers were leaving them frustrated as there was no more cash available.

A group of people in Operations have been tasked to come up with a solution that would analyze the usage of each ATM, understand their daily, weekly and even hourly patterns and come up with a tailored made solution that has replaced the standard “one size fits all” solution. In the past, all ATMs were funded at the same intervals, with the same amount of cash. Employing analytics in the most powerful way, by diversifying the scheme of funding of the ATMs, has not only tremendously increased customer satisfaction. It also saved the bank a lot of money by optimizing the amount of cash kept in the ATMs versus deposited with the central bank.

Audit failure? Impossible!

Another example of a smart usage of analytics relates to predicting that a branch might fail an audit and hence taking action to fix the issues before they happen. Gathering a large number of different variables related to events in the branches e.g. employee absenteeism, cash shortages, customer satisfaction scores and many more resulted in identifying a strong correlation between those different data points and the historic issues with controls and audit failures in those branches.

When such a correlation has been found, it immediately helped the bank raise red flags before a failed audit took place.

 This example can be replicated in many other places that are big enough to gather a reasonable number of data points and the size of database that will enable drawing conclusions from identifying trends.