Young couple checking smartphone

Accessing the best quality and accurate data available is just the first piece of the puzzle

Through applying advanced analytics modelling techniques, lenders are able to deploy data to enhance decision making with accurate results. With more technology now available than ever, how can lenders use analytics to grow?

At Experian, we talk a lot about the power of analytics and their ability to give lenders greater insight into customers and more power to make decisions. In a recent podcast, we spoke to long-standing Experian experts Paul Russell, Director of Customer Analytics, and Chris Curtis, Head of Analytics, about how analytics have changed – and why quality data is more important than ever.

Today we’re bringing you some of the key points from that conversation. If you’d like to hear the podcast in full, you can here:

More data, more power, more insight

In some ways, the principle of gathering data on customers and then applying techniques to draw insight is no different today than it was 30 years ago.

But what’s changed is that there’s now much more data, vastly superior computing power to process it and a range of algorithms that can be applied to spot trends and extract meaning. It’s this combination – not just analytics alone – that gives credit businesses more insight to make decisions and better service the needs of customers in the right ways.

While different kinds of data allow you to build a more complete picture, it’s analytics that sharpen the focus on that picture.

Lenders are moving away from regression techniques and into those driven by machine learning

While scoring is still popular, it has its limits in the decision-making process. Machine learning can be applied to data to improve customer modelling. Whether you’re operating at the margins, wanting to improve customer profitability or looking to cross-sell products, machine learning can provide greater insight and inform your decision-making.

However, it’s important to remember that machine learning is purely a mathematical technique applied to data to deliver an outcome. It doesn’t matter how good your algorithms are, if the data is poor or biased then so too will be the results. The reason that sometimes machine learning gets a bad reputation is that the data it’s based on is not good in the first place.

Ultimately you need to know what you want to achieve and have the right data inputs. If you have, then analytics will help you implement faster decisions at scale.

Advanced analytics can help you build pre-emptive strategies

The more data you have about aspects of a customer’s behaviour, the better placed you are to anticipate their actions in the future. That’s where open banking data, alongside traditional bureau data, can help. For example, you can identify customers who are at risk of defaulting earlier, if you can use analytics to see the trajectory of cashflow going through their account.

Of course, you can never see into someone’s personality to know if they’re going to be a bad payer, but you can piece together disparate data sets to build a more detailed picture. Once you have that, there are complex and intricate algorithms to help you maximise the potential of your decisions and refine your risk models.

Consumers are less patient than ever before

Over the last year, the pandemic has accelerated the growing trend towards greater digitalisation. As such, more customers are demanding quick, seamless credit decisions. If customers run up against too much friction at any point in the journey, they can easily drop and switch to competitors. Analytics and machine learning can give you insight into why they may be doing this at different stages.