The promise of lower transaction costs, increased sales productivity, and more convenient service has lured banks into setting up new electronic and product-specific channels. But they have quickly found that their delivery capabilities are outstripping the traditional branch-centered model they use to manage them. As a result, they face stubbornly high efficiency ratios, expected revenues that never materialize, and channel managers at odds with the standards by which they are measured and rewarded.

To resolve these problems, banks must adopt a new approach to the management of multiple channels. In particular, they must address four issues:

• Who owns the customer?
• How are operational issues resolved?
• How are managers measured and rewarded?
• Where does each business t?

Customer ownership

"By all means add the new channels, but book the revenues with the branch network, where they belong."

The notion that customers can be "owned" is championed by branch managers seeking to preserve a branch-centered organization. They maintain that the branch remains the prime point of contact for sales, service, and relationship management. But their argument does not stand up to close scrutiny.

Far from staying loyal to a branch, most customers today use a number of channels (Exhibit 1). They graze from ATM to telephone to branch, depending on their needs at any given time. They are no more "owned" by a banking channel than they are by a TV channel. If a bank designing a multichannel organization persists in assuming that a channel can own customers, it is deluding itself about how customers use banks and the range of choices they enjoy.

Operational issues

"Maybe individual channels don’t own the customer, but there has to be a single point of accountability for coherent product and channel offerings, branding, and resolution of customer service issues. Someone has to look out for the customer in this institution!"

This familiar refrain is sounded by advocates of "customer-centered" management. For them, retail banks should be organized around customer segments, and those who manage these segments should dictate what products and services are available and how they are delivered.

But this approach can be dangerous. Most delivery channels are used by several customer segments, and a channel decision made by the manager of one of them could adversely affect other segments. Say the owner of a mass-market segment directed branch tellers to encourage these customers to use ATMs. If the tellers inadvertently encouraged mass affluent customers to do the same, they might prompt unwanted defections among these customers.

Resolving operational issues such as how products and channels will be offered to target customers must be an explicit part of multichannel management design. It is also likely to require CEO involvement, for two reasons.

First, decisions on these issues help to shape a bank’s customer value proposition. Will customers enjoy universal access to the full portfolio of channels, for example, or will there be incentives (or disincentives) to use certain channels?

Second, a decision that is optimal from a business unit’s perspective may not be optimal for the institution as a whole. The CEO must be involved in decisions such as: can dedicated product delivery platforms, such as investment centers, develop their own brands independently of the bank’s umbrella brand? Can individual channels choose to distribute other providers’ products, such as mutual funds and mortgages, if they see a profit opportunity? Will the various salesforces be allowed to compete directly, or be required to collaborate?

Performance measurement and rewards

"You’re not going to take away my revenues and just give them to some new channel, are you?"

In a multichannel organization, developing performance measures means:

• Allocating accountability for profit and loss to the right people
• Closely tracking the key economic drivers of the business, whether by channel, product, or customer
• Obtaining meaningful channel, product, cost, and revenue data from management information systems so that appropriate standards can be established.

One unfortunate legacy of the branch-centered system is that a customer’s balances and revenues are normally allocated to branches, regardless of the channels actually used by that customer. As a result, the contribution of branches to the network is often overestimated, and that of newer channels underestimated. This means that the new channels have to depend on acquiring new customers for revenue, even when most of their customers are branch based.

Allocating revenue to branches has two consequences. First, only the CEO has a true P&L, and hence bears all the responsibility for managing performance. Other senior managers are unable to find out how profitable their businesses really are. Second, banks lack the tools and measures to make informed decisions about resource allocation.

Structure and reporting

"Please—not another matrix!"

As banks add new channels to their delivery networks, it becomes harder for the branch-centered organization to manage products, segments, and channels, and the connections between them. Decision making slows down, issues get "kicked upstairs," and managers’ view of how their businesses are performing is blinkered by unresponsive IT systems that have failed to keep pace with the evolution of the network.

Within the bank, leaders become more internally focused. Instead of running their businesses, they get caught up in the issues and conflicts that arise because the decision-making framework does not match the complexity of the business. In one bank, all marketing decisions concerning products and delivery were still funneled through a single strategic marketing group, despite the fact that the number of choices offered to consumers had increased five-fold in recent years. The group became a bottleneck and the focus of internal conflict as it tried to prioritize and reconcile the bank’s marketing messages.

Possible models

Any bank seeking to improve its channel management must explicitly address these four key issues. For its organizational design, it can choose from three broad models on a scale that ranges from tight coordination to unfettered competition (Exhibit 2). The choice of model will ultimately depend on the bank’s strategy (does it emphasize customers or products?), its market position (low market share would argue for more aggressive competition between channels because the bank’s objective is acquiring customers rather than increasing its share of wallet of existing customers), and its view of how the market is likely to evolve (for instance, the speed of product commoditization, changing customer needs, and the costs of complexity) (Exhibit 3).

The coordinated channels model

"I’ve sorted out the issues around the use of brand and pricing across the channels, but cumbersome transfer pricing arrangements and lack of true P&L mean I still get caught up in resource allocation discussions."

Most banks trying to improve their multichannel management exert more coordination rather than less. A coordinated channels model has a clear hierarchy, dominant profit center, defined channel roles, and integrated capacity management (Exhibit 4). To succeed, this highly structured, centralized model needs a carefully designed governance structure, a hands-on CEO, and responsive IT. It is most banks’ default option.

Any bank opting for this model should understand the assumptions on which it is based. First, if investment in the mechanisms that are needed to coordinate channels is to be justified, the benefits of greater cross-selling efficacy, lower customer acquisition costs, and better customer retention must outweigh the drawbacks of slower response times, more costly processes, and lack of broad-based P&L accountability.

Second, the costs of complexity must be more than offset by the value of retaining customers through the deliberate management of their experiences across all the bank’s channels. This means that tellers, call center operators, and salespeople must coordinate their efforts and share information so that they show a single face to customers. Say a customer asks a call center operator about the bank’s mutual fund offerings. Next time that customer visits a branch, a teller should ask whether he or she needs any more information or would like to see a salesperson.

Third, although day-to-day decision making may get slower, this approach can allow senior management to respond quickly and to dedicate resources to new opportunities that might be too risky for an individual profit center to pursue. The Bank of Montreal was able to build its leading-edge telephone bank, mbanx, in just one year—fast work for this highly coordinated multichannel organization. Concerns about cannibalization from the branches were forestalled by the fact that the CEO sponsored, funded, and launched the venture.

Though a comfortable and safe choice, this model fails to isolate the economics of individual products and channels, and thus perpetuates the subsidy of less cost-effective channels. In addition, since it tends to push decision making up the organization, it draws the CEO’s time away from important issues.

The competing channels model

"I’m all for making people accountable for results, but allowing open competition in the network will confuse the customer and raise further channel conflicts."

The competing channels model devolves accountability for results down the organization, exposing channels to the discipline of the market

A free marketplace for channels and products has strong appeal at both intuitive and practical levels. The competing channels model devolves accountability for results down the organization, exposing channels to the discipline of the market (Exhibit 5). Greater transparency in its economics should lead a bank to reduce the resources invested in its mature channels and reinvest in new businesses and channels.

While a more competitive marketplace improves innovation, resource allocation, and accountability, it also means that capabilities in areas such as marketing, product design, and IT may have to be duplicated in each unit. Another drawback is that this model fails to leverage opportunities across the bank, such as the consolidation of the customer relationship in one place. There is also a risk of channel conflict as each channel and product group comes to believe that it deserves to be a standalone business.

The managed competition model

For most multichannel banks, the cost of a coordinated approach may exceed its value. Equally, for a bank to turn itself into a cluster of competing channels can undermine the value propositions and brand equity that it has built during years of serving its customers.

What is needed is the best of both worlds: a model that sets standards across channels based on the bank’s strategy for customer and product offerings, but relies on internal competition to secure efficient resource allocation. Under this managed competition model, product, channel, and service providers independently pursue opportunities to achieve their financial targets within strategic boundaries set by senior management (Exhibit 6).

Banks adopting this model must be willing to take three steps:

Set marketplace guidelines across channels on those few product/channel interfaces that really need it, such as checking and savings accounts, consumer credit, and investment sales. These guidelines ensure that each business unit pursues the corporate strategy even though it may not represent the most profitable approach for that unit. They also recognize that product lines with their own distribution capability (for example, credit cards through the mail and phone) need transfer price and service level agreements specifying the terms under which they provide their delivery capability to other products.

Recognize that, within the limits set, channels will compete and affect one another’s performance. This internal tension, while less comfortable than a coordinated approach, permits clearer insights into the performance of individual distribution channels and the abilities of those managing them.

Make the reinvestment and divestment decisions that are now apparent. Any channel portfolio has both mature and emerging components, and investment must be focused where it can earn the highest returns. When customer revenues are no longer allocated to the branch network, managed competition will expose the economics of marginal branches. Those that cannot be restructured to improve cost-efficiency or sales effectiveness will probably have to be closed.

The rules of the road

At the heart of managed competition are the "rules of the road" within which business units pursue their own strategies. These rules will vary from one institution to another. However, most banks will want to consider the issues raised in the following questions:

Who initiates contact with the customer? One bank worried that competing channels might give its customers conflicting advice. It stipulated that only the assigned relationship manager could initiate contact with high-value customers. For other customers, channels are free to launch sales and marketing programs at their own discretion.

What roles will each channel fulfill, and what levels of support will be required? When every channel has its own P&L, creative managers will seek to manage their channels’ revenues and costs to the best effect. But certain channel roles will need to be prescribed. A manager seeking to make branches profitable by charging for individual transactions, for example, should not charge (at least directly) high-value customers who use branches as a convenience. "Fee for service" arrangements for channels, segments, and products are needed. Without clearly defined cross-channel roles, channels might withdraw service from some customers of strategic value to the bank (for instance, those who have high projected lifetime value even though their current profitability is low), or fall into destructive competition with other channels.

Who determines minimum service levels? Customer segment managers have traditionally been the conscience of the bank, ensuring that line businesses deliver on their promises. Under managed competition, they are more than a conscience: they are P&L owners directly responsible for deciding which products and services are offered to customers and making necessary cost and service tradeoffs to retain their most attractive customers. The managers of product-based channels are free to craft value propositions to make their offerings competitive both internally and externally. The manager of a channel that is used as a convenience by separately managed customer segments—an ATM by an affluent customer, for example—should be compensated for providing the service. Minimum cross-channel service levels are central to multichannel management, and must be defined explicitly.

Two drawbacks of a competing channels structure are duplication in infrastructure and the failure fully to leverage customer information

What infrastructure investments are required? As we saw, two drawbacks of a competing channels structure are duplication in infrastructure and the failure fully to leverage customer information. It is the CEO who must decide where to spend on infrastructure to ensure that overall capacity is adequate and that growth options are realized. Universal banks clearly need to invest in providing online services to customers, for example. But very few standalone business units would build these services for themselves, because they require heavy investment and produce poor economic returns. The decision to commit resources to such services must be based on their strategic importance.

Which channels have the right to revise the institution’s umbrella brand image, and to what degree? It would probably be unwise to allow branch managers to alter a bank’s brand for their traditional channels: control of the brand might be lost, or multiple modifications might dilute the brand image. However, a co-brand or sub-brand might be appropriate for, say, branches that specialize in the sale of investment products, or supermarket branches that carry a restricted product range. The same principle applies to electronic channels.

In general, when channels are tightly coordinated, the brand should be used in a uniform way; when they are competing with one another, they should have some latitude to modify the brand.

Wells Fargo’s supermarket distribution employs the same brand as its core channel, reinforcing the company’s strategy of providing multiple points of access to a standard array of products that require little advice. The Toronto-Dominion Bank of Canada’s telephone banking, on the other hand, uses a different brand from its core branch network, and the two channels have attracted very different types of customer.

Where do you place top talent, and how do you reward them? Under a managed competition approach where each business has its own P&L, the compensation of individual managers is tied to the attainment of specific profit and growth objectives.

Competition for high-performing managers will be both internal and external. Although managers lower in the organization should be free to try to attract the best talent, staffing decisions for the top 15 to 30 positions in the organization need to be made by the CEO. The success of managed competition depends largely on the abilities of the channel leaders—not just the current ones, but the next generation too. Populating these ranks with "promotable" leaders is critical to realizing the potential of the current network and capitalizing on future opportunities: witness former PepsiCo CEO Wayne Calloway’s role in shaping the careers of the company’s top 200 or so managers, ensuring that waves of leaders with crossfunctional experience were being groomed for the future.

A managed competition approach can be appropriate for any bank pursuing a relationship-oriented strategy aimed at deep penetration of target customer segments, such as the affluent. Paradoxically, perhaps the most compelling reason to adopt this model is the decisions the bank’s top managers don’t have to make. The managed competition model offers banks a way of keeping their options open: they do not have to commit themselves immediately to the challenges of a competitive model even if that is their ultimate objective. Neither need they prescribe individual channel conduct (other than through the rules of the road), as they would have to if they adopted a coordinated model. The "marketplace" aspects of managed competition determine most elements of channel conduct.

Banks are far more complex institutions than they were ten or even five years ago. They sell more types of product and deliver them through more types of channel. In the process, they have outgrown their traditional branch-centered approach to bank management. Managed competition offers them a solution. Instead of simply adding incremental channel capacity and hoping for improved performance, banks pursuing this strategy have the opportunity to realize the full potential of the multichannel capability they have built.