This editorial was first published in our Web Fraud Prevention and Online Authentication Market Guide 2017/2018. The Guide is a complete overview of the fraud management, digital identity verification and authentication ecosystem provided by thought leaders in the industry from leading solution providers (both established and new players) to associations and experts.
The rapid adoption of mobile banking presents many opportunities for the financial services industry. However, fully embracing this digital transformation comes with numerous challenges — which require a fully adaptable approach.
It’s not easy to add and maintain support for mobile and desktop users, yet alone build and preserve a competitive edge over multiple newcomers in the digital world. Thriving in mobile banking requires supporting technologies that are constantly changing and evolving, and there’s pressure to add new services such as remote chat and video-based advisory amenities. Furthermore, consumers are demanding a faster and smoother process when creating new accounts and performing transactions, and that adds additional risks and complications.
Detecting fraud without adding user friction
Perhaps the biggest challenges to fully adopt mobile banking are the added security and fraud risks. The many technical complexities of digital banking introduce a whole new array of security vulnerabilities and their associated risks. Furthermore, opening the door to remote customers also opens the door to cybercriminals and fraudsters.
Adding layers of security is necessary, but it can have a negative impact on users. For instance, banks are faced with the dilemma of how to accurately validate the identity of remote individuals applying for new accounts or performing transactions — without annoying legitimate customers with a multitude of security questions and hurdles.
This task becomes even more difficult when multiple channels are involved. No financial service markets are exclusively digital — a personal, human touch will remain critical for years to come. That means banks must support in-person, fully-digital, and a hybrid of both types of banking. Consider a situation where a customer uses their mobile phone at night to begin a loan application, but finds it necessary to ask a few questions via the call centre the next day, and then finish the process in-person at the branch. For both the bank and the applicant to have a smooth and efficient experience, information from all three channels (the mobile app, the call centre, and on-site) need to be fully integrated. To quickly and effectively validate the identity and data provided by the applicant, the loan officer needs a consolidated view of data and risks from different sources, including:
Data entered via the bank’s mobile app
Mobile phone fingerprints
Device cross-checks with blacklists and whitelists
User location
Information gathered via the session with the call centre
Data from the on-site application process
Third party validation of the individual’s name, address, and other personal data
Regardless of which channels are used, for a bank to evaluate and approve new accounts or any transaction, the bank must ingest data from multiple, disparate sources and quickly analyse, calculate, and present the risks so a decision can be made.
Omnichannel banking requires an adaptable fraud prevention solution
Today’s banks not only need to gather fraud data from multiple, different and often incompatible sources, but this data is constantly changing — both in format and in source. Successful fraud mitigation must constantly adapt to varying amounts and types of unstructured data, evolving threats, and continuously-changing regulations and policies. This requires a system that is specifically designed to automatically adapt to this ever-changing environment.
The success of such an integrated and adaptable approach rests on four pillars:
1) Enriching the data lake with pertinent information to create a customer 360-degree view
Since the goal is to create a 360-degree view across products and channels, the fraud solution must ingest structured and unstructured data from any channel and user device—in real time, as well as, batch mode. It must be possible to gather data from home grown and third-party systems along with mainstream tools. Furthermore, the solution should include smart-ingest capabilities that simplify the incorporation of data from any source or format.
2) Leveraging data analytics to build fraud indicator models
Banks have many areas where machine learning (ML) and data analytics can efficiently detect fraud. Superior fraud detection solutions can run multiple supervised and unsupervised ML models in parallel, each tuned for specific use cases such as account takeover, new account fraud, or payment transactions. This level of granular adaptability is necessary for a comprehensive and effective solution.
3) Insightful, visual dashboards and reporting
Efficient case analysis requires intuitive visualisation tools that can provide fraud analysts with comprehensive and relevant data. For analysts to quickly adapt to evolving scenarios and make a decision regarding each transaction under review, the system must provide both summary and detailed information of all events.
4) Enabling continuous refinement and improvement with dynamic ontology
Finally, the solution should give omnichannel businesses the opportunity to set their own rules and configurations to manage fraud movement across multiple channels. This adaptable approach ensures that the fraud detection solution’s performance continues to improve as the business evolves or as threats mutate. Also the addition of new data sources requires a flexible modelling capability that can evolve quickly to incorporate the new data sources.
Conclusion
To stay competitive in today’s digital and omnichannel world, banks must increasingly use evolving technologies that contain vulnerabilities and increase the risk of fraud. Fortunately, there are tools to help mitigate the risks. Organizations in the financial services industry need to carefully evaluate fraud detection solutions they put in place. The choice solution must adapt to an ever-changing environment, be affordable to sustain, and effective at identifying and stopping fraud.
About Rahul Pangam
Rahul Pangam is the Co-Founder and CEO of Simility. He is an industry veteran, with impressive experience from Google, and is dedicated to empowering fraud fighters with the most adaptable, scalable, and accurate fraud analytics platforms and solutions.
About Simility
Simility provides intelligent fraud prevention that grows with you. Our flexible platform ingests data sources in the public or private cloud or on site. Plus, you can easily bring in new sources (whether structured, unstructured, or data lakes) as you grow. Without having to write a single line of code, your analysts can quickly and accurately identify evolving fraudulent tactics across silos and create appropriate rules, thanks to a powerful combination of human intelligence with Similitys self-optimising machine-learning models. Simility helps you spot and stop fraud in real time while providing greater fraud intelligence with fewer false positives.
Every day we send out a free e-mail with the most important headlines of the last 24 hours.
Subscribe now