• 快捷搜索
  • 全站搜索

Digital Banking in Asia

2014-02-15 16:42:35作者:McKinsey & Company编辑:金融咨询网
Digital Banking in Asia: Winning approaches in a new generation of financial services presents McKinsey’s latest thinking on digital banking. Our insights come directly from experience serving clients across Asia; in this volume, wehave focused on the essential dimensions critical to building a digital bank.

        To underpin the transformation initiative and make sure its importance is clear to the sales force, a system of incentives and requirements must be established—one that is noticeably different from what has come before, otherwise the initiative may be dismissed as just another head-office proposal that should be ignored.

3部分-图3.jpg

        In conclusion, successful implementation depends on financial-services companies being able to sustain the requirements of the five areas we outlined atthe beginning of this article. Exhibit 3 provides a checklist that players can use to gauge their level of confidence about whether their digital sales-transformation initiative is on track for success.

        While the digital sales-transformation journey has just begun in Asia, changing consumer behavior, technological evolution, and the positive business impact felt by early movers mean that it is now only a matter of time before financial players of all types are forced to rethink their sales models. Those that are willing to invest and move forward with a sales transformation stand to gain an advantage. But as players implement changes, they should continuously check their program against the pitfalls identified above to ensure that they achieve and sustain a successful transformation.

Chapter 4:Unlocking customer value with advanced data and analytics

(Christian Roland, Gunjan Soni, and Ian St-Maurice)

        Executive summary:

        ● Advanced data and analytics (ADA) is increasingly important to value creation in retail banking.

       ● Banks that have used ADA have identified opportunities across the value chain, from smarter targeting of customers and sharper risk assessment to better predictions of customer traffic in branches and call centers.

       ● Many Asian banks are in the early stages of ADA use; across countries and industries, we see significant benefits for early movers.

       ● We examine five characteristics of ADA that successful banks in Asia have applied.

        Advanced data and analytics (ADA) has become core to value creation in the retail-banking value chain. Banks worldwide are applying ADA to everything from optimizing marketing spending to making better decisions on risk, ensuring more efficient customer targeting and improving service levels in branches or call centers. For many banks, one of the main applications of ADA has helped unlock the value of customer relationships through better acquisition, development, and retention.

        In the United States and Europe, more and more banks are experimenting with ADA to kick-start lagging revenue growth. We urge senior leadership of Asian banks to consider doing the same in the face of slowing revenue growth andincreased competition in many markets across the region. We think banks that fail to pursue ADA might miss an early-mover advantage. In fact, some Asian banks are already developing ADA capabilities through investments in data infrastructure and statistical tools, as well as through hiring analytics experts.

        Most Asian banks are lagging behind on ADA and face big challenges. For example, good customer data was scarce until the recent development of credit bureaus in the region. Perhaps the biggest factor inhibiting the widespread use of ADA is the tremendous growth in consumption and in the middle class. Bank executives felt no urgency to develop their analytical muscle as volumes and market share kept growing. Those days are mostly past in this era of slower growth. Banking executives have an incentive now to squeeze more from their customer base, and ADA offers that opportunity.

        Some Asian banking executives are skeptical about the difference between ADA and the customer-relationship-management (CRM) tools that many banks have invested in over the past 10 to 20 years. ADA employs additional data and statistical techniques to derive customer insights. And those insights are not just in the marketing and sales arena but also across the value chain (Exhibit 1). Banks in Asia and elsewhere that have employed ADA have been able to identify opportunities to increase credit balances, raise cross-sell rates, and greatly improve the yields of their call-center operations. In addition, some banks have started to use advanced analytics to predict foot traffic in their branches, enabling them to match staffing accordingly or to predict daily cash demand for ATMs.

4部分-图1.jpg

        In this article, we focus on how to maximize the impact of ADA in customer relationships, including identifying the customers that are most likely to purchase a certain product, and on how banks can get started on their journey. As noted, ADA is useful across the banking value chain, but we have chosen to focus on customer relationships because this area has the most immediate impact. Squeezed profit margins argue for investment in ADA now. And we see that early movers have captured an outsize share in countries outside Asia and other industries. Additionally, Asian bankers should keep in mind that multinational banks have an advantage in ADA because they can apply ideas from elsewhere and use them to attack in the region.

        Making the case for ADA

        Many financial-services companies and other consumer-facing businesses outside Asia are successfully using ADA to unlock customer value. But some Asian bankers are asking how ADA differs from CRM before they invest in this approach.

        For many banks, CRM was about an IT solution as a repository for all relevant customer data. Some banks went further and applied heuristics for customer campaigns based on the bank’s data. Classic applications of CRM in banking include cross-selling to existing customers and proactive and reactive churn programs. ADA pushes this further, for example, by using sophisticated next-product-to-buy algorithms based on Bayesian models to ensure that cross-sell offers are done with the right product at the right time. Some ADA techniques are possible today thanks to advances in computing power and a lower cost of data storage. While ADA builds on CRM, it differs from it in four areas:

        ● Using new forms of data.Today, banks mainly use internal structured data like balances, transactions, product holdings, and customer demographics. In the future, banks will use both internal and external data (such as from loyalty programs of partners), as well as structured and unstructured data (for instance, call-center records). Additionally, increased use of social networks creates more marketing and sales opportunities (where privacy laws and other regulations permit).

        ● Employing statistical models instead of heuristics.Many banks now use heuristics for their campaigns. For example, based on the experience oftheir salespeople, they may assume that one customer segment, say, urban males between 28 and 35 years old with an income of $1,500 to $3,000 a month, are more likely to buy a certain product. So, in a sense, banks are using experience-based decision rules with simple demographics and information. In the ADA world, banks are employing a hypothesis-driven approach to create hundreds and, in some cases, thousands of new variables (such as the velocity of balance changes or a concentration of locations of withdrawals) that they then test with statistical models to enhance their ability to locate the most attractive microsegments. The result can be a doubling or tripling—or even more—of product-purchase hit rates.

        ● Speeding up response time.ADA decreases response time by identifying and computing in real time the variables that matter to make tailored offers. Using this, banks are now able to make almost instantaneous offers to customers when they call or when they visit specially designed landing pages on Web sites.

        ● Increasing the level of tailoring and personalization.While marketers have long talked about personalized offerings and interactions, ADA allows banks to embrace these, using better computing power and more sophisticated models. This is a step change from marketing to relatively large subsegments such as “affluent rate hunters” or “young families.”

        Getting started in ADA

        Large full-service retail banks across Asia have access to transaction-level data and could use them for cross-selling life insurance, mortgages, and other financial products that are intrinsically linked to changes in life stage. For the most part, retail banks in Asia find themselves unable to pursue this approach aggressively because the data tend to sit in product-based silos (for example, the credit-card group uses only credit data, or life insurance looks at only its own data). Although the data could be of immense value to the insurance and secured-credit departments, there are few organizational incentives for cross-silo data sharing. Still, some Asian banks are breaking down these barriers to get the full power of ADA.

        One Asian bank used ADA to assess its credit-card business. The bank faced an increasingly competitive unsecured-credit landscape and sought to understand how it could promote market-share growth. The bank was in a privileged position—it had access to more data from more people in more products than most of its competitors. Until then, however, opportunity assessments had only used data from within the credit-card department. Using ADA, the bank worked to create a 360-degree view of customers, aiming to delve deeply into theirbehaviors across all product and service categories. The approach effectively tripled the amount of information used to identify business opportunities through the integration of insights across silos and pointed to an opportunity to more than double net receivables.

        Our research and client experience suggest that Asian banks that succeed in applying ADA share five characteristics. 

        Clarity on the source of value

        It is easy to get excited about the potential of ADA. However, ADA is not a panacea for the problems of an organization. Each project should start with an articulation of the business need, how analytics could help address the issue better than today’s tools do, and the expected impact. Executives should avoid falling into the trap of considering ADA as a technology or pure data initiative. There are multiple examples of large investments in infrastructure with little to show for them because they focused on technology rather than impact, building large infrastructures that aim to do it all. These systems take significant time and resources to build and often don’t perform as well as more targeted solutions that are focused on the biggest sources of value, which will differ from bank to bank.

        As mentioned earlier, ADA has used cases across the banking value chain—some banks have prioritized enhancing their risk models with new forms of data, while others have focused on using analytics to enhance branch service levels by predicting customer footfalls per hour for each branch. On the customer side, the source of value can also differ widely. For example, one bank has prioritized payroll and unsecured consumer lending as penetration rates in the customer base were significantly below the level of peers. Other banks have focused on identifying customers that are willing to buy insurance products.

        An integrated data strategy

        Once banks determine the source of value, they should identify the required data and how to obtain them. In our experience, banks use only a limited amount of the available data, typically working with internal structured data like demographic information, product-holding information, channel usage, and transactions. Banks should try to look at the full range of what’s available, including structured and unstructured data, as well as internal and external data, to develop a 360-degree view of the customer.

        Examples of internal unstructured data include call records and behavior in the bank’s channels. Telecommunications companies, for example, are analyzing internal unstructured data like call records to reduce customer churn. Thisincludes monitoring call records for phrases by call-center agents such as, “Sorry, I can’t help you with that,” which are early indicators of churn.

        A much bigger data universe opens up for banks that look at external data. This can include information on property ownership and credit scores. As part of their data strategy, banks should also assess if it is worthwhile to partner with retailers and telecommunications companies to get access to information from sources such as loyalty programs. If data sharing is possible, this can help banks better identify customer life stages and assess true customer value. For example, a universal bank in Latin America created a joint venture with a local supermarket, giving it access to data from a customer-loyalty program. The data set included 1 million customers, 700 million transactions, and more than 100,000 products. Access to the data permitted, among other things, the creation of a risk model that could be used for prescreening and selective preapproval of credit products. Some Asian banks have turned to direct-marketing vendors, such as Yellow Umbrella of India, for similar data.

        An approach for modeling insights

        Once the business problem, sources of value, and required data are defined, banks must choose the analytical methods to crack the problem. Many banks are starting to explore the use of “Amazon type” next-product-to-buy (NPTB) algorithms.

        Typically, the NPTB recommendations for retailers are based on basket analysis. For a book retailer, the basket could contain all book purchases in the last year. The decision logic then is, “Customers who are reading Book A also read Book B.” For grocery retailers, a basket may consist of all the purchases in one trip. The decision logic then is, “Customers who are buying Product A and B together in one trip.” In banking, the definition of a basket needs to be changed to apply the concept; here it would contain a collection of customer-specific data that could include sociodemographics, product portfolio, and transaction behavior. The decision rule then consists of a product recommendation and certain criteria that need to be fulfilled.

        In our experience, the results of the modeling can be enhanced by creating new variables based on business insights and adding them into the NPTB models. A large social-media player in Asia, for instance, was assessing how to identify customers for account upgrading. One hypothesis it had was that people who were rapidly increasing their usage of the site would be more likely to want to upgrade. To identify these subscribers, a new variable was created that measured how much each subscriber was increasing his or her account activity over a four-week period. This was found to be highly predictive of the take-up rate for upgrade offers. Similar triggers can be identified in financial services—forexample, observing a spike in withdrawals could be an indicator of higher need for liquidity, which could be used to target individuals for personal loans.

       The ability to move from insights to frontline delivery

        Generating the insights is not enough—the front line needs to use them. Banks must ensure that the insights flow, without leakage, from analysis to campaign design, target-list creation, contact with the customer, and then to closure. The process is easier said than done, as the leads that emerge from analytics can be lost little by little during the multiple required handoffs.

        To reduce the chance of this happening, we recommend that banks begin with small pilots and scale up over time. Too often, we see banks biting off more than they can chew by building large, fully automated systems that take a lot of time and resources before the value of ADA to the institution can be proved.

        A way to make change happen

        Once the technical elements of ADA are in place, banks must address how to motivate the front line to properly use this new resource. We have seen three big challenges arise.

        First, how can banks encourage product groups to collaborate and share data? Second, in the target-setting and budgeting process, if customer value is managed at the overall bank level, how can different product groups be motivated to cooperate, since each will want to maximize its own business? And third, sometimes the most valuable product to offer to a customer might be one that salespeople don’t believe in or are not well trained to sell: How can you motivate them to buy in and try recommending these products so that the value of the insights is maximized? Some banks are trying to address this situation by running ADA training academies for frontline staff and, in some cases, offering certification (for example, blue-, yellow-, and black-belt levels of achievement) that can be earned by not only attending training but also successfully completing real business projects. Still, training may not be enough if the incentives to cooperate aren’t present. At one bank, although different ADA product recommendations were visible to the call-center service team, frontline staff would still pitch personal loans because the related business unit was sponsoring an incentive-laden campaign.

        As is the case in all change-management situations, there’s not an easy answer. At the highest level, all these issues require a shift of mind-set. In retail banking, analytics are too often equated with reporting. So it’s critical to identify what we call the bilinguals—those employees who can speak boththe language of ADA and of the businesses they support. Then, a choice must be made about where to put these people. Should they be embedded in the businesses to get them closer to the P&L or aggregated in analytics teams to ensure better sharing of best practices and ideas, as well as “analytical economies of scale”?

        Several Asian banks have aggressively pursued ADA, and others are starting to experiment. In other industries, early movers in ADA have achieved higher revenue and profit growth than competitors. Banks that are seeking a competitive advantage should seize the opportunity that ADA presents.

Chapter 5:Making the digital bank work end to end: Digitizing the operating processes

(Robert Feeney, Andy Holley, and Sasi Sunkara)

        Executive summary:

        ● Tech-savvy digital-banking customers expect a flawless experience, so banks must move toward end-to-end process optimization with them in mind.

首页 上一页 2 3 4 5 6 7 下一页 尾页

扫码即可手机
阅读转发此文

本文评论

相关文章