A North America multi-bank study by Novantas reveals an “arms race” to build systems, applications and talent in customer analytics.
Effective application of customer analytics is fast becoming a determining factor in retail banking success, but are individual players up to the challenge? In a continued tight market, the answer could make the difference between ongoing revenue stagnation or growth in market share and profits.
While retail banks have harnessed account and customer data for years to aid in targeted sales and refined portfolio management, many have used these data “around the edges.” Principally, most have relied on street corner branches for both marketing and sales success, particularly regional players. But as more customer shopping activity and transaction volume diverts to online and mobile channels, banks increasingly will need to engage customers in impersonal circumstances, requiring advanced skills in distilling account and transaction data and targeting offers via remote channels.
This creates a pressing situation for mid-size and smaller banks, which find it harder to justify investments in sophisticated customer analytics. For any given application, the expenses and benefits will be spread across fewer households — often far fewer — echoing the scale imbalances seen with core systems technology in the 1980s, before vendors made the technology more freely available.
To define the organizational issues and the industry’s state-of-play, Novantas conducted a multi-bank survey on customer analytics. Completed in the fall of 2013, the project included 13 major North American banks — five of the top 20 U.S. banks, six regional banks, and two of the top five Canadian banks.
The findings showed that smaller players are carrying from two to five times the concentration of analytic staff per thousand households (Figure 1: Competitive Tilt). Yet despite their scale disadvantages and the shared industry challenges of talent scarcity and unwieldy data, many regional banks remain firmly wedded to building their own capabilities — often delaying needed applications rather than turning to outsourced expertise.
The situation calls for a multi-year vision and strategy to build competitive advantage, not just annual incremental improvements to support various projects and business silos. Smaller players must be particularly selective about how they place their bets. The good news is that banks of all sizes were able to show leadership on various survey metrics, demonstrating the difference that creative management can make.
Miles to Go
While there are many important details to consider with customer analytics, the bigger picture is about revenue growth. With both consumer loan demand and fee revenue growth remaining muted relative to other economic recoveries, the best path to growth is to do more business with each customer. The challenge is further complicated by the customer online migration, which undercuts the traditional branch-centric sales engine.
Customer analytics and direct-to-consumer marketing are crucial resources in this battle. Over the longer term, we believe that success with customer analytics stands to make a 1% to 2% difference in annual revenue growth. This translates into a potential 3% to 4% improvement in retail banking earnings growth.
To reach this level of breakout performance, future leaders will traverse a four-stage journey in customer analytics (Figure 2: Evolution of Customer Analytics). The future state of the art will feature integrated multi-channel information; in-depth mining of internal transaction and channel data, as well as web site and social media activity; fluid systems that radically reduce staff time devoted to data extraction and assembly; well-developed programs to recruit, train and retain analytic talent; and savvy outsourcing arrangements.
As underscored by our survey findings, many banks are hovering around what we would term “Stage II” in the evolution of customer analytics, meaning that they are getting better with the promotion and pricing of individual products, but still patching projects together ad hoc and not yet tapping the full breadth of information available on core customers, particularly multi-channel.
Analytical Priorities. Our research suggests that most banks are focused on traditional analytic priorities. These include customer profitability; price elasticity/optimization for deposits; customer lifetime value; network and marketing mix optimization; and propensity/attrition modeling. Findings are being applied across banking’s three customer objectives: acquisition, cross-sell and retention.
Given the important ground that still needs to be gained in these areas, even the most advanced analytic banks are appropriately focused on deeper mining of internal data, in particular the vast payments and transaction data from customers who hold their primary demand deposit account with the bank (Figure 3: Current Capabilities and Development Priorities). Despite industry buzz about mining unstructured data in social media such as Facebook, that is not a likely priority for most banks in the near term.
Data Management. To move ahead in customer analytics, banks must collect and comprehend data of ever-increasing volume and granularity, synthesized across multiple sources to compose deep customer profiles. The complexity of this pursuit is straining current data management infrastructures and exacerbating longstanding challenges in the way that source data are organized.
At many banks, core data systems and infrastructures are slipping out of date. Most organizations still capture and organize the customer data needed for analysis in relational warehouses or data marts. This approach atomizes customer information and absorbs substantial resources to “connect the dots” and derive generalized insights — particularly difficult when teams are forced to start over in a series of one-off projects. Many systems also suffer from the after-effects of hasty assimilation, compounded by lingering mismatches of disparate components melded during mergers and acquisitions.
These factors have created a fragmented data infrastructure where analytical teams must often employ elaborate workarounds just to assemble source information. As a result, survey respondents reported that on average, they spend more than 40% of their time collecting and assembling data.
Setting aside additional time for project preparation and planning, that leaves only about 30% of the effort focused on analysis, interpretation and execution.
Two-thirds of respondents listed difficulty collecting and assembling data as the top challenge, and all ranked this as one of their top three challenges (Figure 4: Top Challenges in the Analytic Function).
Another pressing question in customer analytics is how to acquire and most efficiently deploy talent — a scarce, expensive resource. To counter the scale efficiencies of larger banks, smaller players need to be especially savvy and creative with various resource-magnifying alternatives, including outsourcing, partnering and “renting” resources.
Human Capital. The most important resource — skilled statisticians and sharp, creative analysts who also have business experience — is in short supply. The ability of many surveyed banks to develop this kind of talent ranges from limited to nonexistent, and it is difficult to recruit externally.
Consumer banking is but one of a growing set of information-based industries vying for a limited supply of qualified analytic talent. All will suffer from a shortage of skilled data analysts and seasoned managers.
To further exacerbate the problem, people with promising analytic talent often see banking as a less glamorous industry, presenting fewer opportunities to flex analytical muscle. Many find more attraction in Internet startups, big data firms and even academia.
These factors often leave banks starved for talent, struggling to get by with teams that are both inexperienced and overburdened. Most analysts tend to be generalists, focused on business interpretation supported by more basic data queries, with few capable of more complex or sophisticated analysis.
Clearly, banks will have to join the many other industries that put serious energy into creative programs to recruit, develop and leverage analytical talent. Current bank recruiting programs tend to focus on candidates who already have analytic training and experience, with 85% of survey respondents requiring prior experience. This is understandable but unrealistic. Organizations that insist on hiring only individuals with prior experience while resisting the need to “make” new analysts will find themselves in chronic short-supply; there simply isn’t enough seasoned talent to satisfy demand.
Scale and Outsourcing. Some of the largest banks have hundreds of dedicated analytic staffers focused on retail banking, able to justify the expense because of the vast household customer base over which marketing and sales models can be extended. This provides a material competitive advantage.
As community banks long have known, outsourcing can be an important leveling factor in competition with larger institutions. But smaller and mid-sized regional banks have been slower to embrace this truth, as underscored by our survey, which shows they lag the largest players in outsourcing analytic functions. Interestingly, the larger banks generally view outsourcing as an opportunity to obtain skills and services they cannot easily replicate internally; smaller banks tend to view outsourcing as a cost to be avoided.
While banks of many sizes share a similar percentage of total capabilities built in-house, larger banks work with outside vendors on many more pieces of the analytical function. By contrast, small to mid-size banks tend to delay or even forgo some important emerging applications if they cannot be developed internally, magnifying their disadvantage. Overall, our survey suggests that most banks need to take a more strategic view of when and what to “build versus buy.”
Our summary conclusion from the study is that customer analytics are increasingly foundational to retail banks. It is a two-fold challenge as the industry transitions from branch-centric banking to a more active multi-channel marketplace while seeking to leverage the information advantage that goes with the primary checking relationship.
A pressing issue is expertise, with talent demand outstripping supply. Banks of all sizes face challenges in building the right analytic capacity, and small to mid-size regional players will need to be particularly thoughtful and creative in finding ways to offset the scale disadvantage of bigger banks.
One priority is reducing the amount of time spent by valuable and costly analytic talent on routine, low-value data manipulation tasks. The key is to restructure analytic databases to assure their flexible support for a variety of projects, particularly the more complex models that dig into customer-focused, transaction-level data to derive behavioral insights.
Likewise, banks will need to more proactively manage their human capital by developing robust programs to develop analytic talent internally. Cultivating and retaining sophisticated talent entails a multi-year commitment that includes structured, formal training and on-the-job mentoring.
Governance issues also must be addressed in order to improve talent retention. Too often, analytic teams, managers and individuals become stretched to the breaking point as they are constantly redeployed on an ad hoc basis to varying parts of the organization. Along with degrading performance, this undercuts job satisfaction as well, hastening analytic staff turnover at a time when banks need to slow it down.
Given the value and scarcity of expert staff resources, banks should become much more deliberate in providing clear goals and responsibilities, along with mapping out inviting career paths and growth opportunities. This will help to keep analytic teams invested in their work.
Banks will also need a more deliberate and strategic approach to outsourcing. Often, acceleration in customer analytics will depend on outsourcing select functions and tasks. This especially applies to mid-sized and small regional banks, which need to stay abreast in their analytic capabilities, yet must operate with tighter resource constraints.
Each bank should consider its core strategy — price leader, relationship intimacy or product innovator, for example — and clarify the corresponding required customer analytics (e.g. targeting vs. pricing, or profitability/lifetime value modeling). Core analytical capabilities are the prime candidates to consider developing internally; most everything else probably should be outsourced over time. A discerning ability to focus on the right skills (and avoid attempting to build out all the capabilities) will become increasingly critical.
Finally, with respect to management structure, banks will need to evaluate alternative organizational models. The largest banks (at least $200 billion of assets) tend to favor a distributed analytics model that places individual systems and teams within various management silos, justified by sufficient scale within product and channel silos. But this can work against data connectivity across the bank and foster multiple approaches to business problems.
By contrast, the smallest banks (less than $15 billion of assets) tend to have centralized analytics groups. This permits a consistent approach and good leverage of resources, yet it also can create friction and disconnects between the central analytics team and the client business units, particularly as organizations grow larger.
Regional banks often favor a hybrid model, blending centralized marketing and customer analytics with various product/pricing and distribution analytics spread among key business units. This can provide the best of both worlds: central consistency and efficiency plus specialized expertise where needed. But it can also touch off political battles as influential business silo leaders lobby for resources.
Many challenges in customer analytics have a structural component. It is important for executives to consciously recognize the strengths and weaknesses of the particular organizational model in use at their bank. The right decision for any bank entails a tradeoff between scale and agility in meeting the demands of internal “clients.”
Level No More
For decades, the street corner branch was a great leveling factor in retail banking competition. Even the smallest community bank could hope to dominate a few select locales, given the influence of branch presence.
With the emergence of multi-channel banking, however, the advantage increasingly goes to the banks most skilled with customer analytics. Smaller players tend to fall at the losing end of this equation, less able to justify expensive talent and systems.
For the many institutions that fall into this category, a key recommendation is to sharpen the strategy for capabilities, and refine the balance between internally developed applications and those that will be outsourced. Each bank needs to develop an outsourcing strategy based on a clear-eyed view of its business strategy. This is the key to identifying “must-own” analytics versus ones that realistically should be outsourced.
Mid-sized and smaller players also need an exceptionally deliberate review of required resources and performance drivers in customer analytics. While big banks have scale advantages, they do not necessarily have a lock on talent development, the kind of systems development needed to fluidly support applications, or on the creativity needed to bring concepts and resources together to win in the marketplace.
For all players, the race is on to engage customers and prospects whenever and wherever they want, offering services and products relevant and compelling to them at that moment. This is the emerging multi-channel paradigm for marketing, sales, relationship expansion and retention. Adroit use of data will be essential. Fortunately, in their role as providers of core customers’ primary checking accounts, banks sit at the center of their customers’ payment activities and have access to a unique set of powerful and valuable data. Monetizing this data advantage is the essential retail banking challenge of our time.
Sherief Meleis is a Managing Director, Alan Schiffres is a Senior Adviser, and Grace Lee is a Principal at Novantas, Inc., a management consultancy based in New York City. They can be reached respectively at firstname.lastname@example.org, email@example.com, and firstname.lastname@example.org.