Why are so many on-line banking ventures—start-ups and incumbents alike—foundering? A business that can conduct such a high proportion of its transactions electronically would seem well suited to move on-line rapidly. Yet, perversely, the rate at which customers around the world are adopting Internet banking varies enormously, from a few percent to double digits, both among countries and among banks within the same country.

In an attempt to find out why rates are so variable, McKinsey studied 65 leading banks in ten European countries.¹ The findings show that in bringing customers on-line, "push" factors (the efforts that banks make to attract customers to on-line operations) are almost as important as "pull" factors (demand from Internet-savvy customers for electronic services). The study also indicates which elements of those push and pull factors really matter, both for converting existing customers and for attracting new ones; which banks have the best position for success on-line; and what the others can do to catch up. Finally, the study makes it clear that banks will have to pursue on-line customers, however difficult the environment may seem.

Despite the presence of more than 2,500 Western European banking sites on the World Wide Web—some banks have as many as three, each targeting a different customer segment—only eight million people in the region use the channel either to check their accounts or to make transactions. In other words, about one Western European Internet user in four banks on-line. The study confirms the relatively low adoption rate: by the middle of the year 2000, existing customers who had moved on-line accounted, on average, for only 8 percent of any given bank's customer base. The banking industry was even less successful in acquiring new customers on-line: at the same time, they made up only 2.9 percent of the customer base. The difference between the rates for existing customers converted to the Internet and new ones acquired on it reflects the European banks' low customer churn rate—just under 5 percent a year. So the fact that the on-line acquisition rate is roughly equal to 60 percent of yearly churn means that the Internet could become a fundamental driver of change in the banking industry's market dynamics.

McKinsey's study went on to consider the pull factor of Internet use, country by country. It also compared customer conversion rates with push factors such as bank size (measured by the number of customers or employees or by the size of the asset base), cost-effectiveness (the cost-to-income ratio or cost per asset managed), channel diversification (the number of automated teller machines per customer), and organizational structure (whether banks offer only e-banking or include brokerage services as well and whether on-line operations are spun off or integrated).

The results confirmed anecdotal evidence that most conversions of existing customers arise from their demand for on-line services: pull accounts for 60 percent of conversions. But that still leaves 40 percent of customers whom banks can try to push on-line. And push factors are the only statistically relevant explanation for differences in the banks' ability to attract new on-line customers (Exhibit 1).

Common themes did emerge from the study. Smaller banks convert existing customers and acquire new ones more actively, though the leverage effect of size is especially strong in winning new customers. Smaller banks do twice as well as their larger counterparts among people who switch banks; presumably, they offer the newcomers more in hopes of quickly increasing their share of customers in a business with traditionally low churn off-line. As for the remaining push factors, brokerage is typically a driver for acquiring customers quickly, and their familiarity with other kinds of electronic banking, such as ATMs, does help banks move current customers on-line more quickly, while the potentially huge savings in transaction costs give banks a strong incentive to do so.

One finding was a surprise, given the popular wisdom that Internet operations must be cocooned or spun off and should offer unique benefits. Although the structure and range of the products offered on-line do have a strong bearing on the banks' ability to acquire new customers, such offerings don't appear to have a statistically significant role in converting more of the existing ones to Internet banking. Thus, although attacker banks bent on enlarging the customer base should regard these factors as important, they are of much less interest to banks more concerned with pushing current customers on-line.

An analysis of the on-line penetration of banks and of their overall cost-effectiveness indicates clearly which banks are likely to be leaders—and laggards—in converting customers (Exhibit 2). Some banks are clearly being pulled on-line more than their peers because of demand from their customers (cluster 2). Another group of banks, compensating for relatively low on-line penetration by pushing harder to get customers onto the Internet, has achieved reasonable conversion results (cluster 3). A group of banks located mostly in Scandinavia and Finland (cluster 5) has succeeded in making the most of the region's strong pull factor by simultaneously pushing customers on-line. By contrast, a small number of laggards (cluster 4) that lack Internet-savvy customers and are not particu-larly cost-effective overall have therefore had little success on-line. Finally, cluster 1 represents the majority of banks. For them, neither push nor pull factors are particularly strong, and while they are not necessarily disadvantaged by their low Internet penetration, they would have to make a continuous and active effort to push themselves out of the middle and develop a more aggressive on-line conversion strategy.

In general, the most cost-effective banks are also the most profitable as reckoned by return on assets or return on equity, so the potential for big cost savings on the Internet could widen the performance gap between Europe's strongest and weakest banks. Banks therefore have little choice but to pursue on-line customer conversion: doing so will enable the best to maintain or increase their lead and the laggards at least to stay in the game. And though banks in Internet-ready countries will have an easier time, banks everywhere have a chance to influence their customers and to improve their own performance.