LONDON (Reuters) – From sending special offers on restaurants to burger-loving current account holders to selling anonymised credit card records, banks are racing to monetize the huge troves of data they hold.

Wall Street trails Silicon Valley in using customer information to boost revenue but with tech giants such as Amazon and Google wading onto their turf with forays into lending and payments, banks including JPMorgan, HSBC and Barclays are moving to narrow the gap.

Mining mountains of trading data to predict stock moves; partnering with retailers on marketing campaigns and using artificial intelligence (AI) tools to try and speed up credit decisions are some of the areas banks are focusing on.

In the digital era, knowing how much people earn, where they spend it and what they buy – information some wouldn’t divulge to their closest confidants – is valuable, particularly when banks’ earnings from lending and trading are under pressure from persistently low interest rates and tougher regulation.

“We are now seeing some amazing uses of data in banking, and the reason is pretty simple: they know their clients better than anyone, they have a name and address, information about what you’re buying and once you have those you can do so much,” said Craig Macdonald, head of data monetisation at Accenture.

The surge in data mining is happening against a changed regulatory backdrop. New European Union (EU) rules introduced last year allow technology companies to access banks’ customer data if they have customers’ permission.

The EU has also toughened its privacy laws. Companies now have to get permission before they can collect and use personal information gleaned online from people living in the bloc.

But even with the extra protections, sensitive data is still at risk of being exploited because many people are not aware of how they can shield themselves.

Less than a third of Europeans were aware of all their data rights and only 13 percent said they read privacy statements fully, according to a poll this year of 27,000 EU citizens.

Banks do not disclose how much they earn from analysing and marketing customer data or other ways in which they monetize the information they hold. But, in comparison to the billions earned from lending and trading, the amounts generated are likely to be small.

“If there was a gold mine people would probably have found it by now,” said Benjamin Ensor, an analyst at Forrester. “But if you can generate some marginal incremental revenue at relatively little cost why wouldn’t you do that?”

CUSTOMER 12345

Tie-ups with retail firms is one way banks are monetising their data.

Customers of Britain’s Lloyds and Spain’s Santander can get special offers from a range of retailers after the banks joined a digital loyalty scheme run by US-based data advertising firm Cardlytics.

The scheme uses spending data to give customers discounts at shops they already frequent or which are in their neighbourhood. So, burger-aficionados get deals at local burger restaurants and fashion fans get ads about discounts at clothing stores.

The banks get a percentage of any purchase that results and Cardlytics gets insights on consumer behaviour which help the retailers tailor and fund the offers and discounts.

Cardlytics, Lloyds and Santander declined to comment on what cut the banks get from purchases made through the deal.

“We leverage transaction data that’s created every time the card is tapped, every time a direct debit is made by a customer, in an anonymised way,” said Campbell Shaw, London-based head of bank partnerships at Cardlytics.

“We only need to know it’s customer 12345, we don’t need to know the name of the customer for any reason.”

Bank clients have to enrol in the rewards programme.

A spokesman for Santander said their customer spending data was only shared with Cardlytics if customers choose to receive retail offers. The bank said the information was shared on an anonymised basis meaning the customer’s name is replaced by a unique identifying number.

Lloyds declined to comment on the specifics of the deal. Its privacy policy said the scheme would use customers’ mobile location data only with their permission.

Even with the tougher regulations around big data, privacy experts warn there is still scope for abuse, for example, if highly-indebted people are targeted with unsuitable offers for high interest loans or credit cards.

“If you can use data to get a customer to buy something that they otherwise wouldn’t, it’s good for the bank but not necessarily for the customer and the potential for misuse is significant,” said Paul Bernal, an expert in data privacy at University of East Anglia.

Ashok Vaswani, global head of consumer and payments at Barclays, told attendees at AI conference CogX in London this month that the bank would crunch data in an ethical way.

“We’re going to do it in a transparent and understandable fashion,” he said. “If I can’t explain it [to a customer] I’m not going to offer it.”

Like many banks, Barclays markets anonymised spending data to a range of businesses including mall operators who can see from the information which retail chains attract the most customers and are therefore worth targeting as tenants.

Barclays said it doesn’t share personally identifiable information and it sends privacy notices to customers through a combination of email, text, post and via mobile apps. It also has a page on its website explaining its data privacy policy.

DATA LAKES

Using data to improve risk analysis, make faster credit decisions and anticipate customer needs is particularly appealing for banks looking to cut costs.

HSBC plans to use AI tools to rake through its 10 petabytes of data – roughly equivalent to the storage capacity of 2 million DVDs – from investment banking clients in 66 countries.

Europe’s largest bank has struck a deal with Element AI, a Canadian company, to help it tap this so-called ‘data lake’.

JPMorgan, meanwhile, is developing a raft of AI applications to better predict stock moves and to map and mine 3 billion transactions it handles annually.

The bank hired Manuela Veloso, the head of the machine learning department at Carnegie Mellon University, to be its head of AI research last year.

In comparison to newer, tech-focused companies, banks are often at a disadvantage when they look to extract value from their data – they lack inhouse experts and their businesses are often siloed with legacy IT systems.

To speed things up, lenders are set to spend $26 billion on big data and business analytics this year, according to analysis by International Data Corporation, up from $23 billion last year and $19.7 billion in 2017.

Hires for senior leaders with digital experience at financial firms have doubled year on year for the last five years, according to London-based headhunters Heidrick & Struggles.

“These skills are now a necessity within senior leadership teams,” says Marcus De Luca, UK financial services practice leader at the recruiter.

“We are often asked if there is someone who works at Amazon, Google, Netflix, or Facebook who could be tempted to join.”