How explainable AI can transform your collection operation

Mike leads consumer lending business for a global financial organization. He is looking to upgrade its existing systems from a legacy platform to an AI-based solution. He has heard about the power of AI and how it can help businesses. Within his organization, Mike can see its applicability in risk models and also in taking some pro-active action to check whether the system can predict customer delinquency. As Mike explores this option further, he also understands why existing risk segmentation models are rudimentary based on static rules or statistical models. However, should Mike need to change his core collection and servicing platform to get benefits of technology? He fears that a rip and replace of collection systems would take too long and his business would get impacted if he is not able to use the power of AI sooner. After double-clicking the initiative to implement AI in-house on top of the existing collection system, Mike starts to lose hope in getting faster results. He finds out, that manual data preparation, feature engineering, and model building is a time-consuming process. Also, productionizing and realizing business outcome is a massive challenge with in-house AI implementation. However, Mike also has other problems to tackle. In spite of technology coming on leaps and bounds over the past decade, AI remains a black box, and this presents a significant challenge for audit and regulatory compliance. Although Mike can choose to be guided by an AI system, if he cannot defend or explain the recommendations it makes, and this would place him in a vulnerable position in front of auditors and model risk validators. How can he balance what the business needs faster with what regulation and compliance demand? If you identify with Mike’s conundrum, read on.

Fast becoming the cornerstone of digital transformation, the proliferation of AI has evolved from automating repeatable tasks to powering human-led decision making. Its significant advantages, however, are weighed down by some very real challenges – compliance, traceability, and auditability. This absence of transparency does little to inspire trust in AI-powered decisions, especially when the human being taking the final call accepts liability for any errors. In the financial industry, the need for organizations to make regular judgments about customers exacerbates the problem. Companies need to be able to explain and back the decisions they make, while customers need to understand the systems assessing them, neither of which is possible with conventional cognitive applications. In the financial sector, from fraud detection to lending, we have seen that the hurdle of transparency offsets the appetite for AI adoption.

The alternative, however, is hardly ideal. Consider the lending sector; Data models are updated no more than twice each year, generating dated results may not correspond with the actions eventually taken on consumer accounts. They also take into account a limited number of criteria before creating a broad and inaccurate borrower risk profile. This traditional output from legacy systems offers sub-optimal segmentation, preventing businesses from understanding the risk profile of their delinquent portfolio and, more importantly, the true recoverability of overdue accounts. Intelligent systems could transform these operations by introducing efficiency and accuracy that saves millions of dollars while increasing customer satisfaction. The answer lies in a solution that offers the transformative power of AI while providing human-level contextual explainability.

That’s why we built CollectEdge. A thin layer of AI-powered intelligence that sits on existing core platforms, CollectEdge is a powerful solution that integrates smartness, transparency, and accountability into existing collection systems. CollectEdge helps organizations reduce delinquency rates, improve recoveries, and be more operationally efficient through actionable business insights backed by clear explanations. The solution offers a substantial advantage over existing systems, and its core advantage moves well beyond efficiency. Here’s why.

When you make a machine learning model, the objective is to ignore correlated information and create a feature set that is a combination of two to three data points. We understood that it was essential to develop a system that could generate smarter data models, interpret their output, and offer an easily accessible narrative. That’s why with CollectEdge we built a model that tells you the decision and then shows you which influencing factors helped arrive at the result. These insights are not available in a conventional analytical or augmented analytical system, but are offered by CollectEdge because of its ability to provide decision-level explainability. More importantly, it gives businesses a thorough and explainable understanding of the influencing factors, the reason behind their selection, and their impact on the final recommendation. Companies can also identify and utilize complex links between multiple influencing factors to inform upstream processes, future models, and collection strategies.

Every prediction that CollectEdge makes is derived using specific influencing factors. It’s important to note that these are not just analytics-driven insights. By using a wide variety of influencing factors including, for instance, macroeconomic indicators, CollectEdge offers end users powerful predictive intelligence based on a comprehensive view of the data subject. Combined with the solution’s explainable AI, this ability means that users can now see the thinking and data points behind each recommendation allowing for better auditability, compliance, and regulatory discipline. Additionally, if there is data available outside of the current base product model, it can easily be added using CollectEdge without having to repeat the model creation process.

We believe this eliminates the need for organizations to choose between a less elaborate method with a high level of explainability and a sophisticated tool with little or no transparency. To make it easier for financial institutions to adopt AI, we integrated explainability into CollectEdge in a way that feels intuitive and actionable to end users. While AI-based systems are usually a tradeoff between accuracy and clarity, CollectEdge’s model-agnostic approach provides enhanced explanations. These are provided at the local level instead of model-level recommendations, allowing teams to extract actionable business insights that can be fed upstream into the process and, in turn, deliver real downstream impact.

The best part of the solution is that it can deliver sizable improvement in business performance, including reduced charge-offs, savings on call investments, improved CSAT scores, and near real-time models. To allay any implementation concerns about the magnitude of effort or investment, CollectEdge sits unobtrusively on top of existing systems without any need for ‘rip and replace’ allowing organizations to experience the power of AI adoption in production.

With CollectEdge, companies can now adopt a future-proof AI-led decision-making process that their leadership, regulators, and customers can trust. This addition of efficiency, intelligence, and clarity frees up resources and investments that could drive significant business growth. Now that needs no explanation.

To experience the power of CollectEdge, schedule a demo today.

How NOT to make debt collection a nightmare for your customers

Imagine a scenario where Keith (customer) and Jason (debt collector) are having a conversation:

Keith – Well yes, I know I have a balance past due, but I hope you understand this is not a good time. I am at the office right now.

Jason – Oh! I am sorry to catch you at the wrong moment; can I call back in the evening?

Keith – Well I moonlight as a Radio jokey. I can talk between either my day break 12-1 PM or 6:30 PM – 7:00 PM.

Jason – Very well, let me make a note of this……

While this looked like a perfectly executed collection call, there is a fundamental flaw in this scenario. The problem is that these notes may not be referred ever again and hence a crucial piece of information about the borrower’s communication preference would unfortunately be lost forever!

In today’s world where terabytes are no longer considered absurdly large and gigahertz is easily available on a laptop’s processors, for the lenders, not being able to use every bit of available information might mean losing out to the competition. Customer experience in debt collection is no longer just a good to have.

While the banking industry has seen a dramatic shift in the past decade or so – with online banking and app-based services that enable customers to make transactions anytime, anywhere and across devices, the debt collection process has somehow lagged behind in this customer-centric approach. Most outsourced agencies still have traditional call-center setups, which use rudimentary risk segmentation method without any regards to behavioural aspect of the customer. The lack of empathy in the collection agents’ tone coupled with untimely follow up calls may not only cause grave inconvenience to the customer but may also diminish chances of amicable collections altogether and any possibility of resurrecting customer’s loyalty towards the bank.

What banks need today is a customer centric approach to debt collection. And this is where Artificial Intelligence comes in by enabling debt collectors to effectively plan and execute a well-crafted customer segmentation strategy by utilizing data. The availability of huge volumes of historical transactions of customer data can be utilized effectively to analyze and predict patterns and create accurate risk segmentations by taking into consideration customer behaviors and outcomes.

Read our detailed whitepaper on Debt Collection made intelligent with Artificial Intelligence

In this blog I would like to propose a ‘3 RIGHTS’ strategy that uses AI to create customer friendly collection strategies.

Conclusion

A considerable shift is the need of the hour to look at debt collection from a customer-centric perspective to bring about a change that will not only improve collections but also ensure an improved customer experience. Many banks today, are shifting to technology to ensure a customer-first approach to debt collection, and Artificial Intelligence (AI) is leading the way in ensuring effective steps to make this process efficient and profitable while ensuring higher levels of customer’s satisfaction.

Existing processes can be improved to deliver better value with the application of AI by analyzing data and making recommendations based on usage patterns. Objection handling and language used by successful agents can be analyzed to provide insights and best practices to improve performance and productivity of other agents.

Businesses thrive when customers see value in their services. A great experience with a brand is what helps customers stick. The banking industry has been actively progressing with a customer-first approach and reaping the benefits that come along. It’s time debt collection adopts this approach too, to deliver an experience that is in the interest of customers while constantly improving collection in a structured manner.

EdgeVerve’s CollectEdge is a data driven intelligence application powered by advanced machine learning that helps reduce delinquency rates, boost recoveries and improve operational efficiencies, all-the-while delivering a great customer experience.

Know More

Enabling smoother change management when implementing automation: Part 1

Change is hard. In an enterprise with multiple stakeholders, hierarchical levels of employees and complex processes, even smaller changes could be highly demanding. Effective change management is needed at all levels to ensure the success of any undertaking.

Implementing Automation across an enterprise is one such challenging task. Even though RPA has been known to provide considerable value to businesses, ineffective change management can significantly reduce the ROI from implementation.

There are multiple factors that have to go right for an enterprise to plan, implement and scale Automation. But the following are known to cause the maximum strife to an organization in its digital transformation journey:

Ineffective planning

In a McKinsey Global Survey on organizational transformation, it was found that transformation initiatives can achieve a 79% success rate by taking rigorous planned actions and took them to completion. Only 26% of all respondents had reported success rates in the survey. This underlines the need for effective planning when starting a change initiative.

Even well-planned processes in a modern enterprise grow complex over time with user customizations and workarounds. A theoretically simple process of invoice handling also falls prey to potential pitfalls like human error, human bias and incomplete documentation. Extrapolated over the divergent workforce practices and the sheer number of incoming invoices in a large enterprise, even smaller mistakes and biases tend to shipwreck an RPA implementation.

Underestimating the time and effort involved

While a well implemented RPA could deliver a significant return on investment for an enterprise, achieving that goal takes equally significant time and effort from everyone involved. Managers tend to underestimate the effort involved and overestimate the expected benefits at the start of a project. This in turn leads to unexpected delays in implementation, additional costs and even decrease in morale of the workforce.

For sales and customer support teams, time spent in working with an automation consultant is time away from reaching their targets. Employees do not take kindly to a decrease in their productive hours during the process mapping phase. Due to these reasons, manual or intrusive process mapping techniques like Process Mining take longer than expected, which in turn delays the full automation implementation. Because of this accuracy gets hit.

Inherent resistance to change

Employees are inherently resistant to change. Any change that is not comprehensively planned and backed by exhaustive research and analysis, tends to be met with questions and differing opinions. A full-fledged successful RPA implementation requires whole-hearted support at every level and skepticism means that implementations are not provided with support that is needed.

Human bias and inefficiency- The common factor for failure

Most enterprises are not successful in implementing major transformative change. Bain and Company found in a study that 20% of major change efforts fail completely, and only 12% meet their goals. But the good news is that the underlying risks are known and manageable. The solutions recommended by Bain and Company are about getting buy in from the employees and on effective communication and clarity provided by the leadership.

If the answers are so obvious, why are organizations still failing at large scale change management? It is because organizations do not realize they are making these same mistakes. Leaders and employees are prone to all the cognitive biases that afflict other humans as well. This inherent bias is the common denominator in the three factors discussed above.

Thus, for a change initiative to be successful, the first step should be to remove the human bias and errors that arise due to it. An automated, objective and data driven execution will definitely lead to a successful transformation.

Balancing Risk, Effectiveness, and Customer experience in debt collection

“I wish the collectors would understand me better. Being a customer for over 15 years should mean something. Money is tight just now but it’s not like I’ll never pay them back. I wish they could respect my schedule and send me an email instead of the incessant calls when I’m at work. Is that too much to ask?”

“I wish the business understood that my primary responsibility is to collect on the portfolio. There are roll rates and charge offs to manage. Yes, customer experience is important, of course, it is! But, I manage a portfolio with millions of customers and don’t have the tools to personalize my approach. My team needs help.”

If this sounds familiar, you’re not alone. Creating a great customer experience in lender-debtor engagements, especially delinquency, can be difficult. On the one hand, the traditional approach to collection results in inefficient outreach since it doesn’t take into consideration the customer’s channel preference. On the other, this lack of focus means the costs of staffing and account management are through the roof. For a customer, the absence of nuance in the debt recovery process can be distressing. In the case of collections, specifically delinquency, the focus on risk over customer experience can also cost companies valuable relationships.

In order to address the problem, we first had to develop an intrinsic understanding of the context. Financial institutions have to balance a number of often conflicting priorities such as capitalizing on credit appetite, mitigating risk, and minimizing bad debt all while ensuring a quality customer experience. Legacy systems continue to inform collection strategy, applying a one-size-fits-all approach devoid of insights into an individual customer’s often unique payment situation. Also, risk segmentation remains rudimentary, based on a myopic view of the debtor generated by a limited selection of variables such as payment history, DTI, and credit scores. As a result, every year, sub-optimally prioritized collection efforts informed by a limited view of the borrower cause companies to spend hundreds of thousands of dollars on call investments covering even low-risk profiles. These well-intentioned but misdirected efforts can cause significant customer distress, damage the reputation of the business, and, in some cases, result in lawsuits.

That’s why we built CollectEdge – a thin layer of AI-powered intelligence designed to help banks and other lending institutions make faster, smarter, and more effective business decisions. CollectEdge is an intelligent bolt-on that sits on top of existing collection systems, instantly transforming their performance by adding efficiency, perspective, and explainability. Currently, the time-intensive nature of manual model creation means that they are only updated once or twice a year, resulting in dated information that may not be reflective of a borrower’s circumstances. Up to 40% faster than traditional manual segmentation efforts, CollectEdge enables organizations to frequently create risk profiles in line with a customer’s actual financial ability and propensity to pay. Companies can now automate the effort of data cleaning, feature engineering, and model creation, saving millions of dollars in resource and process time. If you think those efficiency gains are substantial, the application’s intelligence is truly transformative.

The intelligent insights from CollectEdge are derived from a holistic customer understanding generated by a wide variety of data sources – credit history, customer behavior across channels, both internal and external, and macroeconomic factors like GDP and employment rates. The product is able to deeply analyze a wide variety of structured and unstructured data including customer calls, emails, texts, public posts, blogs, tweets, and comments, enabling it to segment delinquent portfolios by relative risk based on intelligence at an individual level. Furthermore, this mechanism allows the segmentation to be based on a unified view of the customer across products and channels, as opposed to a product-specific view of the customer that never offers the whole picture. In addition to unearthing new influencing factors through this comprehensive approach, the product also discovers complex linkages between these factors, empowering companies to extract actionable business insights for process improvement and customer excellence.

CollectEdge then offers account-level treatment plans and collection strategies with each recommendation substantiated by a list of influencing factors, augmenting business processes with human-level intelligence at scale. With a deep view of the customer and their preferences including the frequency of contact, mode of outreach, time to reach out, and the tone of communication, the outreach team on the floor can now personalize collection strategies with unprecedented specificity. The result – genuine business impact. CollectEdge’s implementation model includes a commitment to demonstrate business results such as a 25-50 basis point improvement in reducing charge-offs, better CSAT scores, and even a 10% increase in call savings.

While the industry has started to explore best-in-class technology products and AI-enabled tools, most of these still remain in the PoC phase. And since core systems are procured and implemented at significant capital investment, a rip-and-replace approach isn’t just expensive, but also fraught with risk given the absence of demonstrable results in production. The ease of implementation and configuration means CollectEdge helps businesses develop a competitive advantage by shifting their thinking from a reduction of cost and effort to an increase in better decision making and customer satisfaction. The application’s vertical-specific offerings easily feed intelligence to existing systems generating updated segmentation for users within weeks of implementation. This makes CollectEdge a sizable opportunity for financial institutions to test the impact of AI in production before committing to larger digital transformation efforts.

In a highly competitive landscape where each basis point counts, the seismic shift from reactive damage control to predictive analytics and segmentation allows organizations to balance retention and customer experience efforts with risk management. Click here to discover how CollectEdge can help you create a better experience for your collections teams and customers.

References:

https://www.nclc.org/images/pdf/pr-reports/report-analysis-debt-coll-ftc-data.pdf

https://libertystreeteconomics.newyorkfed.org/2019/02/just-released-auto-loans-in-high-gear.html

https://www.kansascityfed.org/en/publications/research/mb/articles/2018/auto-loan-delinquency-rates-rising

https://www.cnbc.com/2019/03/16/robocalls-about-your-bills-can-pour-in-every-day-all-day.html

https://www.experian.com/blogs/ask-experian/what-do-americans-hate-most-about-debt-collectors/

https://www.consumerreports.org/consumerist/new-rules-would-require-debt-collectors-have-proof-you-actually-owe-money/

https://www.pymnts.com/news/artificial-intelligence/2018/ai-debt-collection-consumer-behavior-brighterion/

https://www.insidearm.com/news/00043910-artificial-intelligence-can-it-help-colle/

https://www.mirror.co.uk/money/debt-collection-firms-creating-more-13951289

https://www.nclc.org/issues/consumer-debt-collection-facts.html

ARE YOUR DEBT COLLECTION SYSTEMS UP-TO-DATE?

Elevating levels of debt magnify the chances of delinquency, which can considerably reduce profit margins and surge costs for lenders.

Furthermore, since lofty debt levels render the economy to extreme vulnerability, inability to check this stalemate in time can also give way to a global economic meltdown.

Thus, with the amount of global debt accumulating every year, traditional processes have been able to do very little to arrest the problem, creating the need for advanced systems that can inhibit galloping delinquency rates by also enhancing customer experience.

PROBLEMS IN TRADITIONAL COLLECTION PROCESS

In 2017, the global debt reached an all-time high of $184 trillion in nominal terms, the equivalent of 225% of GDP. On average, the world’s debt exceeded $86,000 in per capita terms, which is more than 2½ times the average income per-capita.1

To recover loans, traditional systems deploy time-intensive manual correspondences that impede proper resource utilization. Also, they only analyze recent internal transaction data and lack accurate risk segmentation models. Regular updates of data since collection or loan approval are also not available with these traditional systems.

The use of AI and machine learning (ML) has become obligatory for businesses to beef up collections and skirt the initial roadblocks presented by legacy systems.

AI IMPROVING PERFORMANCE RECORDS

When it comes to collection strategy, AI-powered machines emulate cognitive human behavior to solve a host of issues discussed below:

Powered by advanced ML capabilities, CollectEdge is a data-driven intelligent application designed to help lending and collecting organizations reduce delinquency rates and boost recoveries. CollectEdge can assist companies that want to make existing debt collection systems more intelligent and seek a balance between mitigating loan losses and enhancing customer experience.

COLLECTEDGE

CollectEdge studies data across channels and looks up statutory/regional regulations to recommend appropriate channels and time for contact. It understands personality traits, negotiates and connects with borrowers based on e-mails, texts, public posts, blogs, tweets and comments. In addition, it offers preventive insights, risk predictions and ML-based queue prioritizations that focus on profiles with high chances of recovery and personalized repayment plans for delinquent customers.

Apart from major perks like lower delinquency rates, reduced charge-offs and higher operational efficiencies, the software offers a unified view of customer accounts and can be synced with existing core collection systems. With CollectEdge, companies are not required to rip and replace their existing legacy systems and train their talent resources on new technology. Easy implementation and faster time-to-market, unbiased data model and easy access for audit teams are factors that can make the product popular among companies concerned about value cycles and understandability among auditors.

Sources:

New Data on Global Debt, https://blogs.imf.org/2019/01/02/new-data-on-global-debt/, January 2, 2019

IOT: Adds life to things

I vaguely remember reading about classification as living and non-living things in my childhood days, but now with IOT, we can visualize redefining them, and comfortably say, I can have a chair that communicates, shoes that connect to a network, door lock that can sense me walking close by. This technology enables the things around you to sense, communicate, network and produce new information, thereby adding life to things. It marks a crucial turning point for future in the same way as invention of computers and Internet did a few years ago.

IOT is still yet to get the limelight it deserves. Reasons could be many, and this raises the inquisitiveness to delve deeper into what goes into creating these thinking and talking things. There are number of questions like how can things talk, how do they connect to network, how are they controlled and how secure are they? IOT as I understand, extends connectivity to everyday objects. Going a little deeper, it also helps in monitoring, analyzing and controlling these devices.

IOT Ecosystem: Smart device in Smart space

  • The first element in the IOT Ecosystem is the smart device. Everything in this space has some information to give. The otherwise traditional device turns into smart device with the help of sensors. Sensors detect, measure and respond to physical parameters and convert them into electronic signals. Best example is biometric sensors.
  • The second element is the Microcontrollers – which give brain to the device. We can call them tiny computers with small memory, processor and programmable input/output pins. These can have programs to store data generated by a sensor. It can do math on the data collected and control the device.
  • Gateway: a piece of networking hardware that has a node which can network with other network devices using GPRS, Wi-Fi or Ethernet.
  • Cloud: The whole lot of data collected and processed is stored on the cloud for further processing and analysing. (Ex: google drive).

To summarize the flow – the sensors generate data, controller collects the data, sends to the cloud through Gateway. The data is analyzed over cloud; the analyzed output is sent back to another controller which further triggers the device to perform an action. What makes this process intelligent is that these smart devices can now take part in your business processes and they use intelligent interfaces to connect with the rest of the world.

IOT Security: An extra eye and an extra ear right in your home

With almost everything on the Internet, the risk of personal information getting leaked is high. The threat may not be only to invade public spaces but can also influence decision making. While the policy makers, encryptions and firewalls would do their part to secure IOT, we also need to own this piece. There needs to be a limit on whether we really need everything on the Internet or just a few things.

IOT and Banking

Banking is just entering into the space of IOT. Currently there are use cases like tracking ATM usage, tracking crop yield, tracking vehicles sold on loan and this gives a lot of data to the banks. Once banks convert the IOT data into useful information, they can engage with their customers better and increase their market share. The core component in banking is money which is already being used digitally on mobile, online and other channels. However, we are currently using indirect applications of IOT. It can be of more significance if used in branches where the customers’ inflow and outflow can be tracked. IOT can also prove to be a major differentiator to improve branch staff and field staff services where sensors can be used to track the nature, extent, and quality of services provided.

Exciting times ahead

It might take a while before the presence of IOT is felt in our daily lives but thanks to the underlying technologies like advanced wireless networking, cloud computing, standardized communication protocols, smaller size silicon chips and finally falling prices of computing resources, it won’t be long before we start seeing various corporate IOT applications deployed in our surroundings.

Application design for the changing context

Mass customization in application design is the concept and act of designing interfaces that make every user believe that a particular interface is designed exclusively for them. Personalization at scale is the key driver for mass customization. Any task or activity, persona regardless, spans various stages such as interaction, operation and business process. In most of the cases, it cuts across multiple application boundaries as well. Personalization should span all these dimensions of interaction, operation and business. Context plays the biggest role in determining personalized behavior.

Key elements required for context include:

  • Where (location, branch, ATM, kiosk, Facebook, Twitter, Online banking, Mobile banking etc.)
  • When (time of the day, day of the week, week of the month, month of the year etc.)
  • What (business moments, payee, amount, biller, etc.),
  • Technical context (browser type, browser version, mobile OS version, signal strength, battery strength etc.)
  • Who (Teller, RM, bank customer, segment, user group, persona financial behavior profile etc.)

Please note, the above list is not exhaustive.

All dimensions of context are essential for an interface designed for personalization. In this article, we look at one of the dimensions – “where” – and explore it to figure out if there exists a pattern.

Access phase which followed automation phase in the banking industry focused on enabling the customers to carry out banking transactions using digital interfaces like ATM, online banking and mobile banking. All the controls and personalization were designed around these consumption points or touch points which are “Technical” in nature like ATM, browser, mobile, kiosk etc.  For example, banks can setup different limits (which can be personalized) for browser, mobile and ATM.

During the next wave, banking services were made accessible through social media applications like Twitter and Facebook. Now, Facebook can be accessed using a browser on a desktop or through a mobile application. Hence “Application” provides context, with design controls for suitable behavior to drive personalization. In this wave, “Technical context” was immaterial. Hence the controls and personalization are around “application context” irrespective of their access points like browser or mobile.

During the next wave banking services were made accessible through third-party application through open banking APIs. This is a very interesting aspect since the provider of the banking services is neither aware of the technical context (can be browser or mobile) nor aware of the application in which banking service would be consumed. Here the context would be “Third party”.

This changing context is captured with examples in this table:

ChannelSupported ByApplicationTechnicalityContext to be considered
Online BankingBank’s own channelOnline BankingBrowserTechnical
Mobile bankingBank’s own channelMobile BankingMobileTechnical
ATMBank’s own channelATM bankingATMTechnical
FacebookFacebookFacebookManaged by 3rd partyApplication
TwitterTwitterTwitterManaged by 3rd partyApplication
AISPThird partyManaged by 3rd partyManaged by 3rd partyThird Party
PISPThird partyManaged by 3rd partyManaged by 3rd partyThird Party

Banks should recognize the various dimensions to provide personalized experiences amid changing context, and make sure that the right applications, advice, and assistance are made available. Designing such as experience cuts across interaction, operation and business layers.

In these increasingly digital times, interactions will originate on different channels and will also be influenced on these channels. And many of these channels will not be owned by banks. Context at all times, irrespective of the channel, will be the cornerstone of personalization in this digital future.

Banking on Public Cloud

Customer delight is the key to success in business, and cloud is one of the foundational building blocks and enablers for the digital technologies like analytics, machine learning and artificial intelligence that can serve to enhance customer experience significantly. Instead of spending time on concerns about infrastructure and resources, enterprises can shift focus on their core business by using the public cloud. As per the sixth Cloud Native Computing Foundation survey, the top three benefits of cloud native technology include faster deployment time, improved scalability and cloud portability.

The public cloud is defined as computing services offered by third-party providers over the public Internet, making them available to anyone who wants to use or purchase them. They may be free of charge or sold on demand, allowing customers to only pay as per usage for the CPU cycles, storage or bandwidth they consume. Multi-clouds or multiple public clouds are also being used by banks to avoid vendor lock-in and reap the benefits of competition among cloud providers.

The main benefits of using a public cloud for banks are:

  • Reduced cost of ownership by avoiding investment in and maintenance of on-premise resources and switching to an affordable subscription model. It is prudent to invest for average peak load, and manage unusual capacity with cloud bursting.
  • Scale up and scale down the right capabilities for flexible and efficient workload management thereby handling unusual peaks efficiently. There is minimal to no lead time with instant provisioning of resources in the cloud.
  • Agility in application delivery to offer engaging experience and innovative products to customers.
  • Readily available ecosystem ‘connect’ with multiple public cloud providers, FinTechs with cloud native applications and experienced cloud service providers. Communication and information exchange through APIs allows banks to innovate with partners, FinTechs and extended developer ecosystems.
  • Operational efficiency by automating the development lifecycle, provisioning etc. on the cloud.
  • Localized data centers are being set by public cloud vendors which ensures regulatory compliance.
  • Enhanced encryption practices and security controls of vendors such as AWS are far superior to bank’s own infrastructure and data center security. What’s more, AWS also provides the necessary tools to assess and meet regulatory compliance. The vendor is also working with regulators in various countries to advance the case for public cloud adoption.
  • Support from regulatory bodies to embrace cloud with a number of guidelines emerging in countries around the world to guide and support banks in their journey towards public cloud adoption.

Security concerns with respect to adoption of public cloud is misplaced. With inbuilt identity and access management and other design considerations for securing applications and secure DevOps processes as part of public cloud infrastructure, bank users and customers don’t need to worry where the application is deployed regardless of location. An end user advocacy group body called Cloud Customer Standards Council has come up with a reference architecture for securing workloads on public cloud which can standardize security on cloud.

Banks need to keep the following aspects in mind while assessing readiness towards Public Cloud:

  • Cloud vendor selection and working with them
  • Requirement for upskilling, training and support
  • Evaluation of the application portfolio and decision on what to re-architect, replace and directly source from cloud as a service (SaaS)
  • A thorough understanding of how cloud will be deployed, managed and used within the organization

Banking Visionaries’ Council in a paper on Banking on Public Cloud – Reimagining Business Agility, lists the key cloud migration strategies and Cloud Governance and operating model that can be adopted by banks. Cloud migration strategies that can be adopted by banks include Rehost, Replatform, Repurchase, Refactor, Retain, Retire where “Rehost” can be used when change required is minimum and “Replatform” or “Rearchitect” when change required is significant. According to the paper, a bank must design its cloud architecture to be flexible enough to support both its infrastructure and banking applications at different stages of evolution, with the bank using public cloud and SaaS for its IT needs in its final stage of cloud journey.

The journey towards cloud and cloud native technologies makes a real difference to the profitability of a bank. Keeping an open mind towards these technologies ensures business success.

References: https://medium.com/aws-enterprise-collection/6-strategies-for-migrating-applications-to-the-cloud-eb4e85c412b4

Analytics vital for Customer Delight

We all receive recommendations on online portals or apps for movies, books, music, videos, food etc. based on our history of online transactions. Similarly, for transactions performed over the counter we receive recommendations on products, information about discounts and offers on our preferred brands, based on the history of transactions performed using a privileged membership card. All these insights empower us to make purchase decisions.

Organizations are doing everything possible to offer products that customers want to buy based on insights about their lifestyle using advanced analytics on transaction history data. This allows them to improve the ROI on their promotions by eliminating irrelevant offers.

Similarly, banks have a wealth of data at their disposal from multiple touch points combined with customer interactions on social platforms to arrive at recommendations. With the help of all the data they have, banks can understand the latent need of their existing customers, and contextualize their recommendations, offer pre-approved personalized financial products through their digital channels and encourage customers to buy them, thus considerably increasing the possibility of success. The cost for the bank this way will be a fraction of what it can offer through any other mode, thereby increasing the profitability and competitiveness.

A data-driven approach to technology allows banks to improve business outcomes and innovation. Adoption of analytics solutions provides an opportunity for banks to:

  • Integrate data models and open source technologies to rapidly develop and deliver actionable insights to supplement the decision making process.
  • Use existing data stored on cloud and supplement it with semi-structured and unstructured data from various channels including social media, discussion forums to arrive at customer’s interest and needs, segment customers to tailor product offers.
  • Drive product and service improvements using analysis of customer opinions for any product or service on social media to arrive at better offerings.
  • Arrive at actionable insights for customer acquisition using analysis of customer opinions about their competitor’s product or service.
  • Predict strategies for customer attrition and retention as well as cross-selling and upselling using actionable insights on customer satisfaction.
  • Seize on market opportunities by providing relevant offers to customers through effective customer communication.
  • Innovate using open banking initiatives where data can be securely and efficiently transferred using APIs on a service agreement to offer more value to customers.
  • Drive risk, fraud and compliance management which are essential due to the associated costs involved.
  • Teach AI platforms so that machines can learn from experience/data, find patterns and respond accordingly and perform tasks such as customer call routing.
  • Deliver multichannel experience by gathering real time data and build consistent view across channels.

It is essential that analytics capabilities be accessible to all functions within the organization to empower employees to take decisions based on actionable insights.

Offering the customer convenience, control and choice using data analytics does not only make the customer happy but rewards the business with repeat purchases and hence customer retention. Data security also needs to be factored in to avoid loss of trust of stakeholders involved. Thus a robust framework of security guidelines and protocols need to be in place in this journey towards true customer delight.

Private vs Public in the Digital World

The title of the blogpost may lead you to think that I am going to compare private enterprises vs. public enterprises. Or from a digital standpoint: private network vs. social network. From a proprietary standpoint: private and confidential information vs. open source content. From a data standpoint: private cloud vs. public cloud. From a personal identity standpoint: privacy vs. publicity!

We should not ask the age of a female and salary of a male was the saying in the past. But things are changing in the digital age where both are at par. And moreover, few things are really private now.

It is essential to be part of a network to share information, to learn from others experiences, to stay connected and be up to date about what is happening around. Think for a while: Are the people who were initially a part of your network still required to be part of the network? Especially when it comes to a private network with high turnaround and people moving between projects, roles and careers, is the “private” information being shared still private?

We love to share our feelings, check-in details, pictures on social network as some people say there is nothing to hide or that is what social media is meant for. In fact, the more social we become the more we will be targeted with better product and service offerings by the digital giants who are always listening. But for choice and convenience are we giving up control over our information. The answer lies in having a control over our personal data – what we consent to, who we give our consent to and what information we want to share.

In certain cases, the data is used for research purposes, to come up with a point of view and arrive at trends so that users of this information can make a better decision. All good intentions! But in most cases the information is used for monetary benefits by taking advantage and misusing genuine customer consent.

Privacy is your right to control your personal information such that you can decide how much others should know about your personal information. It is something very fundamental to trust. Private and sensitive information may act as a competitive advantage for companies to differentiate themselves. But for individuals, information such as their health condition, for example, they may not want to reveal to anyone other than their doctors and near and dear ones. Being open about your condition with your doctor is good to get the required assistance but that information seldom needs to change hands any further. Even this data is being bought and sold openly in this digital age to generate revenues because of its sheer value which is an unintended consequence of further information exchange.

Privacy and data breach is considered to be a standard business practice to offer better service to customers at lower costs and not something which should needs to be prevented. Overlaid with digital technologies like cloud, mobility, analytics, social, we continue to aid the proliferation of more and more information.

Protection of privacy lies with each entity. Business entities need to understand whose data they are trying to monetize and what the right way to use is. Individuals need to be careful and choose what to share, what not to share and have control over their personal assets i.e. information in this case!