"If only we could see the whole relationship on one screen" is the lament of every executive dedicated to raising the bank's cross-sell ratios.

Yes, having a complete view would vastly enable wise decisions about expanding a customer relationship. But rare is the bank that has a 360 view of a customer or even a 180 or 90. One glimpse at a time is the best view that many ever get of the average bank customer.

Cross-selling can't wait for data nirvana. Undersold customers are a strategic asset, but they won't stay that way. They will find the additional financial services they need somewhere.

Though few banks have a 360 view of their customers, some still do fairly well at cross-selling. They are clearly not letting data gaps constrain them.

Within your own bank, some number of customers holds multiple products. So clearly the data you have is good enough, especially if you're a mid-sized or small bank.

The problem is not lack of data. It's a lack of data usage.

The alternative to the massive IT integration doesn't have to be yesterday's spreadsheets, manually cleaned and reconciled, then more manual work to make them reveal actionable information. Instead, the fastest and easiest way is what might be called the poor man's data mart — a standalone "cross-sales data mart." Banks that go this route have found a pragmatic way of accessing available data and leveraging their current reporting platform, all without significant IT intervention.

Building this sort of cross-sales data mart takes six steps, all leveraging capabilities a bank already has.

Extract the data you have. Based on your cross-sell assumptions and strategies, are you looking for mortgage customers who might need mortgage insurance? Or small business customers who might need investment products? Or student loan holders who might need a debit card? First look for data on your most trusted, accurate systems, and then augment it with other available sources.

Create the rules for loading the data to your cross-sales data mart. This will ensure you are working with a universe of customers with whom you can reasonably expect cross-sale success. You need business rules (e.g., take only those customers who have been active in the last 18 months, or eliminate those that have recently closed accounts with us). And you need technical rules (e.g., records showing Scott Silverman and Scott A. Silverman at the same address are duplicates).

Run analytics on the data. What has been that customer's behavior in the past? What activities is that customer engaging in now? And what can we predict, based on available data, that customer is likely to do in the future?

Some banks make extensive use of third-party information sources to estimate the size of customers' wallets and thus their own share of that wallet. But lacking that, they can still arrive at a reasonable estimate based on the customers' product usage. A customer with a high-balance personal checking account and a low-balance mortgage has apparent potential to buy a CD.

Define the reports that cross-sell campaign managers can use. Some bank's reporting capabilities are limited to static reports, while some offer interactive reports that let you drag and drop columns and manipulate them for deeper insights. Answer such questions as: What should our new cross-sales data mart deliver to users? What dimensions do they need to see - customers, products, and services? What metrics do they need to see - numbers and financials?

Determine who should have access to which data and reports. Federal laws, state laws, and bank policy all play a part in dictating access to your cross-sales data mart. Security is paramount in protecting customers and protecting the bank against unwonted incursions.

The final step is to operationally align your sales and marketing teams with the data you have analyzed. The data should be consumed and results of campaigns should be fed back to the data mart so that it becomes part of the overall business fabric.

Now, after a process that typically takes a mid-sized bank 12-16 weeks (compared to years or more for a full-fledged cross-business IT integration), you are ready to implement targeted cross-sell campaigns.

You have gone beyond good hunches to fact-based estimates that will make the most efficient use of your sales and marketing resources. Not only is your data good enough right now, but making use of it through cross-selling will perfect it. Every time you use it — every interaction you have with the customer — will add to your knowledge about that customer and what that customer needs next.

Scott Silverman is a managing principal in the consulting solutions division of Collabera Inc.