Danske Bank Maximizes The Value of Data With Decision Intelligence
To extract the full value from their data pools, enterprises need to have the right IT infrastructure in place. Here’s how Danske Bank is resolving legacy challenges around data, improving collaboration between business and IT units, and leaning in to the future of cloud adoption.
Industries worldwide have undergone rapid digital transformation, driven in large part by the recent pandemic, which forced organizations in every sector to find new and innovative ways to protect their vast data resources against acts of fraud while finding new ways to provide better services for their customers.
The banking sector was among the early adopters of digital technology to:
Shield their customers and investors, who were having to adapt quickly to digital services, from criminal organizations bent on extracting dollars from them through acts of fraud.
Investigate illicit acts perpetrated against the banks themselves by bad actors seeking to hide funds earmarked to support terrorism or to hide illicit profits made through human trafficking, drug dealing, or other crimes.
To meet these growing threats, banks needed a solution that would provide a complete view of their customers and counterparties to mitigate fraudulent acts while providing accurate, actionable data to meet government compliance standards.
Danske Bank was one of the leading financial institutions in transforming anti-money-laundering (AML) investigations by utilizing advanced entity resolution to more deeply understand the relationships between its customers and the counterparties with which they were transacting.
By automating what were previously manual processes to gain actionable intelligence from data and using artificial intelligence and machine learning to create more reliable data streams, the investigative process has been greatly streamlined, providing significant cost savings in terms of human resources needed to analyze data.
A Challenging Landscape
A recent study by IDC summarized some of the key challenges banks are facing in the race to digitally transform their operations:
Anti-money-laundering (AML) investigations within financial services organizations (FSOs) are often inefficient and ineffective.
Manually-intensive processes contribute to rising labor costs each year, but investigators’ productivity is hampered by poorly integrated data sources, requiring inefficient data collection and analysis. FSOs have indicated that swivel-chair investigations cause major effectiveness and efficiency challenges.
FSOs often struggle to produce a consistent, holistic view of counterparties and the direct and indirect risks that they present to the institution.
Investigations are hampered by huge volumes of false-positive alerts generated from rules-based transaction monitoring systems that look at the volume and value of anomalous money movement transactions without surrounding context.
IDC identified a number of challenges specific to Danske Bank including:
Significantly bolster the bank’s ability to detect, prevent, and investigate financial crime
Build customer trust
Assess risk across geographies and off-board those that don’t fit with the risk appetite
Improve efficiency and effectiveness of manual processes
As the Marjo Pikkukangas, Head of SAR Analytics and Oversight for Danske Bank described, “We needed the ability to know that our customer is who they say they are and understand with whom our customers are transacting. Our challenge was to identify technology that could help us do it faster and with more efficiency, effectiveness, and depth.”
In a recent discussion with Quantexa CEO, Vishal Marria, COO Technology & Services for Functions & Data at Danske Bank, Bo Svejstrup, went even deeper, pointing to historic challenges that many banks including Danske are facing:
Underlaying data problems
Siloed data resources
Potentially missing data
Problems revolving around data lineage, which is especially problematic when working in a regulated market
Pulling all of the disparate data resources together
Inability to properly analyze datasets in context to render clear, actionable data
Lack of agility to respond to rapidly evolving challenges and customer needs in the marketplace
How Decision Intelligence is helping
As Bo was quick to point out, “Cloud technologies can help us to have a smarter view of how to deal with some of these traditional data challenges in terms of automating some of the control processes, and also using analytics in order to get insight into your data and deal with some of the lineage issues that you have in our regulatory environment.”
But cloud technologies alone are not the silver bullet to meet every challenge the banking sector faces.
Speed and productivity are crucial elements to ensure banks are not just gaining access to data, but that they are able to access the right data with the right quality. In short, it’s about realizing the value that comes from bringing the IT and business side together.
As Bo explained, “That can be from a regulatory point of view, being in control of your data in order to deliver professional risk management services, or it can be from a commercial point of view, in terms of ‘how do you deliver better services to your customers,’ and ‘how do you drive those use cases together?’ That’s becoming more and more important to understand the end-to-end value chain, or how you bring that out.
And while machine learning, AI, robotics, and automation, can all help expedite the process, if you don’t have the context right about your customer and you’re just doing automation on its own, or applying machine learning to old data in a very new world without context, you are going to miss a lot of valuable information.”
This is why gaining the full view of customers and their counterparties through Decision Intelligence is so important in providing the context upon which data can be more fully utilized to drive better decision-making.
Building that context, using both internal data and the external data, helps to manage risk, but it also provides a roadmap to the future in terms of how to better serve customers. This is what Bo refers to a true digitalization.
“It comes back to taking the insights of your business context,” he said. “The business knowledge that we have built up literally over hundreds of years in the bank about banking and combining that with the very big volumes of data that we have about our customer base, our customers, and our business in general. And putting that in front of our customers in a digital, scalable way, almost like on a one-to-one basis, and thereby serve our customers with personalized services.
“At the end of the day that’s probably where some of the changes will lie in the future, so that we can be more relevant for the individual customer, using our knowledge and the data, and combining that with technology in the center of that.”
Another key part of the quality data equation involves the human factor. When you put the customer at the center of your actions, being able to combine human intelligence with artificial intelligence and machine learning is key. For example, your Relationship Manager, who sits astride information services department and your business-unit or function, or your credit decisioning officer will know certain things because they’ve had experience with data that no machine has touched.
As Vish explained, machines can take that human knowledge many steps further. “The machine can scale that decisioning, and also find anomalies and find patterns that simply the human brain cannot touch, and it’s always about combining them both i.e. ‘How do you get the best of the human and the best of the machine and bring that together to best service your customers?’”
Key benefits of Decision Intelligence
The rewards for being able to access the right data with the right quality are truly transformative.
One dataset for multiple use cases: One of biggest benefits in being able to use these enriched data resources across different use cases within your organization.
As Vish explained, “Getting the right data, those are also use-case dependent. If you’re looking at a KYC use case or an AML use case in comparison to a revenue, sales, or marketing use case, the critical data elements for those different use cases would be a bit different.
“There will also be certain data attributes that you would want to give to an application in AML or transaction monitoring that you wouldn’t necessarily give to a sales and revenue application.
“So having that control, but also having that accessibility and almost on-the-fly applications to pull those datasets and come up with those insights, becomes even more important in today’s technology so that you’re not continually copying the same data multiple times across your enterprise.”
Efficiency and cost-savings: As IDC described, “Danske Bank was able to strategically change its approach to handling AML alerts and investigations with the implementation of key technologies, including Quantexa.
“Entity Resolution technology provided the framework to conduct AML investigations in more efficient and effective ways. Manually intensive investigations involving investigator searches in legacy systems to collect relevant data were replaced with automated data integration and analysis tools.
“As a result, Danske Bank has substantially reduced the average cost of an investigation while greatly improving the quality and consistency of investigations. These improvements help Danske Bank to ensure that its customers can transact safely and securely.”
ROI gains: Danske Bank saw material reduction in time to determine salient risks and related investigation costs based on average case handling time. The reduced cost of maintaining infrastructure by streamlining the use of investigative tools from many to a few, as well as infrastructure savings, continue to be realized as the use of Quantexa is extended to other use cases and areas of the bank.
Other AI benefits: Investments in AI, including Entity Resolution, will help to reimagine how AML investigations will be executed. For example, AI combined with insights derived from Entity Resolution will enable straight-through processing of low-risk alerts. Level 1 investigation automation will free up resources to focus on more complex L2 and L3 investigations.
Benefits of building a holistic view of customers and counterparties
By adopting Decision Intelligence solutions, Danske Bank investigators now have the ability to conduct context-based investigations that provide a holistic view of the inter-relationship of customers, entities, and transactions across a network.
This network view provides context to individual transactions and customers under investigation. An investigator can often make determinations about the whole network, not just a single customer. This results in:
A lower volume of investigations and higher value output
Decisions can be made to off-board entire networks versus individual customers
Vastly improved investigation quality through true-positive identification of risk. Since implementation, more cases have been determined to be high quality through quality assurance measures
False positives have been reduced and continue to improve
Investigators now have a standardized way to identify risk and have access to a standard set of data with simple access to it
Contextual investigation technology provides investigators with suggested actions and assessment of risk, mitigating inconsistencies that naturally occur in any manually intensive process
At the end of the day that’s probably where some of the changes will lie in the future, so that we can be more relevant for the individual customer, using our knowledge and the data, and combining that with technology in the center of that.
COO Technology & Services for Functions & Data, Danske Bank
What’s next for Danske Bank?
As he looks to the future of the financial sector and Danske Bank in particular, Bo identified several key trends regarding data analytics and the Cloud that will be critical business drivers:
Ecosystems of data used for real-timing or streaming data. “The use of ecosystems in the sense of joining multiple data sources—your own with external data sources—and using that in your company, but potentially also exposing some of these insights for the world to use.”
Technologies that provide different types of insights at a fast pace will be key. “Systems and technology that support real-time creation and access to data will become more and more important in decision-making.”
The Cloud will provide scalability. “I think this is where you make things scalable, and, as I said before, you don’t have to think too much about how you use these technologies, because they’re available, they’re speedy, and they’re productive for the platforms that you choose to work on that provide the insights for you with the different types of technologies that you need, whether it’s advanced analytics, machine learning, or whatever it might be.”
As recommended by IDC, “AML investigation transformation will increasingly be a strategic priority for many leading FSOs over the next few years. To stay ahead of regulator expectations and competitors, FSOs must take strategic actions to transition to next-generation, contextual investigations.”