How can a company with multiple marketing initiatives see a 500 percent return on one hand and a 25 percent loss on the other, and not be able to show which is the winner?

In financial services, this is a common problem. Unlike other industries, such as consumer packaged-goods (CPGs), most of the financial-services (FS) industry has not devised the capabilities necessary to fully interpret the range of data it collects. Thus, FS companies frequently cannot analyze the return on investment (ROI) on their marketing spend. According to Booz Allen Hamilton estimates, the industry lays out more than $10 billion annually for marketing; its ad spending alone is approximately $8.5 billion, putting it fourth, behind automotive, retail, and telecom/Internet — and that means a lot of money is going to waste.

The FS industry can boost its marketing effectiveness by 15 to 25 percent, resulting in significant bottom-line savings, by putting in place tools and processes that will measure marketing ROI more accurately than marketers’ intuition. Putting those elements in place requires marketers to start by answering a series of questions:

• How much are we really spending on marketing, and across what types of initiatives?
• Across the entire spend, how much of a return are we getting and where is it coming from?
• Is the marketing strategy aligned with the sales strategy, including the desired customer relationships?

Gathering this information allows FS marketers to base strategic decisions on hard numbers rather than soft estimates. One bank, for instance, discovered that the vehicles it uses to acquire and retain customers yielded wildly different levels of effectiveness. It also found, to its surprise, that results it had previously attributed to marketing were actually more closely related to its number of branches per capita. Another company drew on its ROI findings to slash its annual broadcast television budget from $70 million to $10 million, and shift spending into cable television, online media, and sponsorship channels.

Digging Up the Information
The paradox of FS marketing is that the industry has access to more marketing data than nearly any other industry, due to its in-depth relationships with its customers, but it is much more difficult to capture, analyze, and act on this data than it is in other industries. FS companies face certain inherent structural constraints in financial services that other industries don’t. In CPG, for instance, coupons, point-of-sale data, and trade spend information — materials that FS doesn’t have at its disposal — paint a clear picture that marketers can use in analysis. Further, CPG companies do not contend with complexities such as interest-rate climates or variations in customer profitability due to product mix.

As a result of these constraints, the FS industry has little in the way of standard approaches to market analysis. The ways FS companies gather data vary broadly: Some gather massive quantities of raw data; others rely heavily on anecdotal evidence. Some marketing teams still measure performance armed with only year-over-year comparison methods, whereas others use equally simple forecasting models. Finally, the FS industry still relies on cumbersome “control” markets. This method has long been discredited in most industries because it excludes certain markets in order to use them as a baseline, resulting in forgone sales opportunities or diminished brand presence in those markets.

Building a Toolbox
Although these sector-specific conditions are constraints, they don’t have to be barriers. We have used three ROI methodologies at companies with various levels of data at their disposal; the amount of data available to a particular company will determine how best to employ some combination of these three strategies.

Breakeven Analyses. In cases in which data is difficult to obtain, breakeven analyses add quantitative rigor to certain types of marketing spend that lack true outcome data. Consider the case of a mortgage bank’s $2 million TV show sponsorship, which allows the bank to air several commercials and associate its name and logo with the show. A breakeven analysis estimates what sort of viewer penetration would be needed to justify the spend. The bank estimates that, of the estimated 4 million households that regularly watch the show, the sponsorship reaches one-quarter. From these 1 million households, the bank estimates that 25,000 are in the market for a mortgage, because 2.5 percent of U.S. households took out mortgages in that year.

For this bank, the estimated lifetime value of a mortgage is $4,000. Therefore, the bank would need 500 new customers from the pool of 25,000 to break even. This bank’s national market share is already 6 percent, so the bank could reasonably expect to “automatically” get 1,500 of these customers, and would therefore need to boost its share of this group to 8 percent. Given the strength of viewers’ loyalty to the TV show and the implied loyalty to its sponsors, the bank believes the breakeven target to be achievable and is satisfied with its investment in the sponsorship.

Impact Models. Where good outcome data exists, impact models reduce noise and isolate the linkage between marketing actions and business outcomes. Many FS companies have already captured the data necessary to understand their return on marketing investments; next, they should put tools in place that regularly feed this information to the marketing team in a format they can use. Doing so requires five steps: