Reduce Risk by Building a Diversified 'Portfolio' of Customers

September 12, 2007

The Nobel Prize-winning economist Harry Max Markowitz revolutionized modern investment practice when he introduced the world to his "portfolio theory." Markowitz's big idea was actually quite simple: He believed investors would be served best if they built a "portfolio" of securities -- a broad-based roster of assets that included stocks with varying degrees of risk and potential return. The key to diversification, according to Markowitz, is to include a mix of assets with returns that do not move in tandem. In a nutshell, investors should select portfolios, not individual securities.


Recently, a research team at the W. P. Carey School of Business began to wonder whether Markowitz's now widely-accepted investment method might be applied in the world of marketing, as well.


Marketing professors Ruth Bolton, Michael Hutt and Beth Walker, and Crina Tarasi, a doctoral student in marketing, have proposed that a company's choice of customers is every bit the "investment decision" that an investors' choice of stocks is. And so, according to the research team, companies would be wise to identify the risk-return characteristics of all of their potential customers, then build a "customer portfolio" that, like a well-designed investment portfolio, can reap returns in both good times and bad -- and shield against tough times in the marketplace.


The researchers, who are now working with a large North American firm to study the real-world application of their idea, believe their method could help companies break the all-too-commonly seen habit of seeking out only those customers who appear to be attractive now, while ignoring those that may well prove to be moneymakers in the long run.


"What firms typically do is try to attract and keep the most desirable customers of the moment, either the ones they think are most profitable or the ones that have the most lifetime value, instead of looking at the overall profitability of all of their customers -- and what kind of customers it makes sense to add in order to minimize risk," says Walker. "It's just enormously tempting for companies to go after only what is profitable at the moment."


The result?


"You might end up with a kind of patchwork of customers," adds Bolton. "But if you had looked at this from a total portfolio sense, that would have given you a different risk-return profile, rather than just looking at the financial returns for the next quarter."


A defense against burst bubbles?


The research team first got the idea for this new approach to customer acquisition while researching IBM in the early 2000s. It was the heyday of the dot-com startup, and IBM was taking heat from Wall Street for not attracting more business from the then-exploding dot-com market.


"They were under pressure for not making more penetration into the dot-com customer pool," Hutt says. "They weren't pursuing the new economy, if you will. They were receiving all of this heavy criticism from analysts even though IBM was still writing very large outsourcing contracts with some customers and maintaining relationships with major enterprise accounts."


That wasn't good enough for analysts, Hutt says, who wanted IBM to follow the lead of the more trend-conscious firms, such as Sun Microsystems, that were going out of their way to be supplier of choice for those young, rich dot-com entrepreneurs. The company even introduced a new ad campaign targeted specifically at the dot-commers. In it, Sun boasted that it was "the dot in dot-com."


Then the dot-com bubble burst. Sun lost a sizeable chunk of its portfolio, and IBM looked a whole lot smarter.


"When the bubble burst, the portfolio that Sun had built basically went away," Hutt says. "Many of those customers had gone public with venture capital, and many of them probably had a lot of Sun boxes sitting in their offices that were never opened. So it was very interesting at that time. What was actually helping stabilize things [for many companies] were those large outsourcing contracts [such as the ones IBM signed], which began behaving like bonds do in a turbulent market. And when things did start to turn around with respect to IT spending, it was not necessarily the large enterprises that [started it], but it was the more established small and medium- sized firms that led the way."


In other words, whether or not IBM had intended to do so, the company had built for itself a pretty smart portfolio.


Despite the company's relative lack of dot-com customers -- customers that, according to the research team's approach, might be seen as high-risk/high-return clients -- those bread-and-butter outsourcing contracts kept cash flow coming in when times were hard. And when some of the firm's smaller customers -- small-cap stocks, in other words -- started doing well, IBM's bottom line rebounded quite nicely.


And that, Hutt says, should be a lesson for other firms.


"If you think of a customer roster as a stock portfolio, where benefits of diversity pay off, you might suggest that an IBM or a Sun could benefit by looking at a portfolio comprised of customers with different challenges and cash flows," he says.


From blue-chips to junk bonds


The research team classifies customers in much the same way that investors classify stocks.


For every blue chip stock and small cap, international stock and junk bond, there is a comparable class of customer. And if companies could simply begin thinking of their own client rosters in this way, those companies would go a long way toward both protecting themselves against market dips and setting themselves up for long-term gains.


The customer classifications break down as follows:


·    "Blue Chip Stocks": These are large, established customers with long-term, often non-contractual relationships. And though these blue-chip companies don't ensure big returns, they offer low risk and stability.

·    "Small and medium caps": Small- and medium-sized customers that have both high risk and high growth potential. Because of their counter-cyclicality, the research team says, these customers represent a great pool for diversification.

·    "International stocks": International customers, especially in emerging economies, have high risk and high growth potential. Like small and medium caps, this class of customer presents a logical diversification pool -- even though, individually, they are inherently risky.

·    "Treasury securities": Examples here might include subscription contracts, including maintenance or service contracts, which are highly predictable and offer the lowest risk of any customer class. In tough times, these customers provide guaranteed, predictable returns.

·    "High Quality Bonds": Long-term contracts with long-established customers. Contracts, as a rule, reduce risk, and contracts with long-term customers are even more valuable. These customers, the research team says, are predictable and less likely to endanger a good working relationship by defaulting on a deal.

·     "Junk Bonds": Short-term contracts with short-term customers. Though contracts provide some protection, the short-term nature of the relationship creates uncertainty. Default risk is high, but a small presence in an overall portfolio may enhance diversification and potential returns.


The researchers are quick to point out that these categories are broad ones, and may well be further broken down into additional categories -- international customers, for instance, may be divided by region, industry or nation, to further identify risk and return factors.


They also note that finding the right customers is only half the battle.


Companies must also recognize that the customer portfolio strategy only works as a long-term strategy. In the securities market, frequent buying and selling rarely produces winning results, so the research suggests that companies adopt a similarly long-term outlook when it comes to their customers, too.


"Like securities, for which the cost of commission can outgrow the return when they are traded too often, the acquisition cost for customers can surpass the return if the company errs in selecting customers or decides to discard them prematurely," the researchers recently wrote. "In other words … the customer acquisition policy should reflect the long term company strategy and not impulse decisions."


Bottom Line:


  • Just as Markowitz's "portfolio theory" suggests investors are best served by an investment in a broad array of assets, companies can benefit by viewing their customer roster as a "portfolio" -- a mix of different customer types with varying degrees of risk and return potential.
  • The bursting of the dot-com bubble reveals the danger of not taking a portfolio approach -- companies who focused specifically on the dot-com market struggled in the burst's wake.
  • Just as investors view blue-chip stocks as safe, reliable investments, companies can view large, long-term customers as their blue-chip customers. But companies should also seek out smaller firms, or international firms, that may offer less certainty but also bring huge growth potential.
  • Just as the frequent buying and selling of stocks rarely brings favorable long-term results, companies should develop an overall customer portfolio strategy and consider the ways that particular types of customers add value.