Bart Leurs (Rabobank) on Customer Data Collection: 'Full Transparency is Essential'

Banks manage numerous customer files. What is the biggest challenge for Rabobank when it comes to managing that data?
‘The crucial issue for me is to discover in what areas the company should excel, given the increasing amount, relevance and dependence on data. In terms of numbers: Customers visit our app roughly 2 million times a day; approximately 100 million transactions are made through the app each month; 9 million phone calls go to the bank every year. Those are all data points. We aspire to excel in 3 areas. Firstly, in data management: We must ensure that we have high-quality data that is and will remain accessible and correct. The latter is not that simple. Every company creates and uses data everywhere, but it requires a great deal of management skills to keep them up to date.
Secondly, we need to pick up speed in working with algorithms and machine learning. Companies have fewer and fewer human interactions and interventions., but this also applies banks, but also to companies such as booking.com or JustEatTakeaway. In the future, most of the bank’s core processes will run on algorithms and therefore on automated decision-making. To some extent that is already the case. Our app provides customers with insights into their spending pattern. For instance, does most of the money go to the bakery or the butcher? How much do you spend on vacation? It seems simple, but a complicated algorithm is running in the background that classifies all those expenses. The great thing is that customers can change things themselves in the app. When an expense is misclassified and “vacation” should actually be “business trip”, you can adjust this yourself. The performance of the algorithm improves from the input of the customer. The app also advises you in using the bank’s services, such as how to block your card or how take out insurance – provided you do not have one with us yet. What kind of advice the customer gets is not determined by people, but by the underlying algorithms.’  

Do you also use artificial intelligence to improve the quality and efficiency of internal business operations?
‘A great example of this is Fundr, an initiative for loans to small and medium-sized enterprises with a business account. We have put together a completely new method of credit assessment that is fully based on machine learning and data we collect from the customer. Based on that data, we can assess a credit application in a few seconds. In addition, we are increasingly using algorithms to detect fraud and money laundering. This way we discover patterns or transactions in the payment system that we would never be able to extract manually. This is a highly effective and valuable use of artificial intelligence for one of the bank’s core functions: Ensuring no fraud occurs through the system.’

For customers to trust a bank, it is essential that the bank handles their data with care. How challenging is that?
‘This is the third aspect we aspire to excel in: Data ethics. We have drawn up a manifesto in which we secure how we do or do not want to use data. In this manifesto we state, among other things, that data always belongs to the customer and that we want to be utterly transparent about what we do with it. The customer is in control and can indicate in the app in what ways we can use his or her data. For example, you can turn off an algorithm if you do not wish for it.
Besides that, we have an ethics committee that assesses whether the way in which we share or process customer data is acceptable. It must be crystal clear to a customer why he does not receive credit while his neighbor does. For example, we need to make sure that there are no elements of discrimination in the decision-making process?'

Traditionally, the banking sector has been organized around products such as mortgages, insurance and checking accounts. Digitization pushes companies to think more from the customer's point of view. How does Rabobank approach this?
‘The transition from product-oriented to customer-oriented has been going on for ten years now. In the development to our agile structure, we identified the most important customer interactions when purchasing products or services: The customer journey. Occasionally such a customer journey is still connected to a product. Whoever wants to purchase a house eventually takes out a mortgage and thus ends up with a product. Other journeys, such as simply becoming a customer, are disconnected from products.’
To make contacting the bank smoother, we then set up multifunctional, agile teams that are fully grouped around those customer journeys. Of course, we certainly still have various departments for mortgages or insurance, but this should not be a concern for the customer. We strive for our app to remain consistent in terms of its design and feel. In the past this was different. The challenge for us is to make sure the different parts of the customer journey act in consort and ensure that all data is managed centrally, so that information does not get fragmented. There is still room for improvement. However, thanks to customer interaction technology we can already see whether a caller has been active on the internet or can complete his journey later in the app.’

Digital innovation is an important part of your portfolio. What is your view on the completely digital native fintechs. Are they a threat to traditional banks?
‘A considerable amount of competition favors any industry. Fintechs such as Adyen, N26, Klarna, Revolut and Nubank keep us attentive. Due to our size and our former legacy and technology, we are not as agile and fast as those new players. The majority of our technology investments is spent on renewing the old company. We are unable to alter this. We can only make sure that our future investments are competitive and solid enough.
The most important question is whether we, as a large, unwieldy corporate, are able to respond quickly enough to what is happening in the outside world and what customers expect. Ultimately, this has a greater impact on our competitive ability than tech investments have. In that respect, our agile transformation has been key. As a result, currently 6,000 people work in teams that are fast, agile and have more responsibilities to be able to react quickly. Consequently, I would not categorize new players as a threat. Be that as it may, we should not have a complacent attitude toward it. New generations may embrace the shift more easily.’

What opportunities do you think there are for collaborating with those new players? Traditional banks have the authority and credibility toward the market, and, conversely, fintechs can assist the banks accelerate digitization.
‘Companies such as Adyen or Bunq are true competitors, as opposed to other start-ups that we have actively been working with. An example is the video platform 24sessions. During COVID, we started using their technology for remote conversations about mortgages. Now more than 90 percent of those exchanges go through their platform. Another example is SurePay, our own start-up for fraud checks in the payment system. Today all banks in the Netherlands and a number of banks in the United Kingdom use this system.
We already offer banking services at places where it was previously not possible. We call this banking as a service. To meet the needs of the platform of Bol.com, for instance, we have developed and plugged in technology that allows entrepreneurs to apply for instant credit. We have also linked our payment platform to Apple Pay, thereby integrating credit card processing within Apple. I also recognize opportunities in the field of customer due diligence: Know your customer. As a bank, we have to go through a very thorough process to ultimately onboard new customers. In an increasingly digital world, banks could offer identity checking as a service to others. After all, when companies and organizations do business on the Internet, they want to know that the other person indeed is who they say they are. In my view, there are only two authorities that are successful at doing so: The government and banks.'

As well as there are opportunities, there are risks. The more parties collaborate in an open economy, the more can fail in the areas of ​​security and privacy. How does the bank handle this?
‘Together with other banks and the government, we are constantly doing everything we can to test whether we are protected, for example by trying to break into our own house or by having others do it. To manage this becomes certainly more complex due to more interaction with the outside world. That is why it is important to have a clear overview of the deals we make with other companies about sharing data, so that our customers are confident that these companies handle data as carefully as we do. I believe the deal with Apple is very clean. Data on payments by Apple Pay never goes to Apple iCloud and only the last 10 transactions are stored on your phone. Since Apple earns money with hardware and services, we can make reliable arrangements with them.
If we took this up with Google, which makes money from data, we would have a different conversation. I feel that people themselves will increasingly make a difference between companies that do or do not handle their data with care. We live in a time where we take full advantage of the joys of technology, but also begin to see the disadvantages: Identity theft, bias by algorithms, manipulating elections… The question of who you trust with your data is becoming increasingly important for consumers – and we have been looking after our customer data with great care for 125 years.'

Where does the bank's responsibility end and the customer's responsibility begin when it comes to cybersecurity? For instance, if there is an issue with the credit card payment, is the bank held accountable too quickly?
‘I find it difficult to give a clear answer to this, considering it concerns such a broad subject. Of course a heavy responsibility rests with people themselves. Secure your phone with a code. Check your privacy settings. Perform software updates. When people give permission to share data, it is oftentimes with uninformed consent. They hardly know what the other parties do with it. We inform our customers about possible fraud scenarios and how they can be alert to cybercrime, while making sure they will never bear the full extent of the risks.’

This interview was published in Management Scope 08 2021.