Comcast's hostile move to buy Disney is the type of adventurous offer that takes investors' breath away—along with their money. Since making an unsolicited, $48 billion bid for the Disney empire last week, Comcast has lost roughly $9.3 billion in market value. The odds against pulling off a successful mega-deal are well known: More than 70% of the big-bang mergers performed since 1995 failed to create significant shareholder value, according to our research.

Yet there is a better way. Executives should make a big deal about doing a lot of little deals. We have taken a close look at the so-called "deal paradox"—why so many deals seem to fail, and yet company executives and boards keep doing them. Our research examined 1,700 public companies and their deal history from 1986 to 2001, and separates the acquisition winners from the losers.

It turns out that frequent acquirers that build skills and experience through a host of small deals come out on top. By turning their back on Comcast-type moves and applying themselves instead to building capabilities to identify and execute small deals, the slow and steady tortoises earn twice the returns of the inconsistent hares.

By contrast, the Comcast-Disney hook-up has several classic traits found in deals that end up destroying meaningful amounts of shareholder value: It is a massive deal (depending on how you establish value, Disney is worth about $60 billion, including long-term debt); it is a merger of equals (Disney shareholders would end up with 42% of the combined company); it is hostile (Disney CEO Michael Eisner has already told Comcast no way): and it represents a new line of business for Comcast (Comcast is a cable company and Disney is a whole bunch of things, but is not a cable company). It also has dubious strategic merit: Comcast CEO Brian Roberts spoke to analysts about restoring growth to the theme-park and animation businesses, activities far afield from Comcast's core cable franchise.

Companies that create significant shareholder value through deals make a different set of choices:

— They make it a point to do lots of deals. In our study of all publicly traded companies from 1986 to 2001, we found that the more deals a company made, the more value it delivered to shareholders. These companies are serial acquirers, doing deals in a systematic fashion, year in and year out.

— They do little deals. The average deal size for the winners in our study was 10% of the acquiring firm's market cap. Most of these deals were so small that they were rarely disclosed other than in SEC filings.

— They do friendly deals. Our interviews with the CEOs of the deal winners pointed out how long and involved the courtship process can be. Many of the acquired companies were private, requiring the buyer to convince the seller of the merits of the move.

— They stay close to the core business. Our work repeatedly pointed to the value of reinforcing a company's core business as opposed to moving into far-flung adjacencies. When the winners we analyzed moved beyond their core, they still built on fundamental strengths in their business. Our studies found that deals motivated by increasing the scale of an existing business outperformed those undertaken to expand scope by a considerable margin.

— The strategy is easy to understand. When a CEO has to explain why a deal makes sense, it usually doesn't. All companies have a fundamental basis of competition on which their success is rooted. Deals that strengthen that basis of competition make sense; those that open up a new competitive basis usually don't.

Comcast's bid for Disney is all the more surprising because it departs sharply from the process that made the company an industry leader. Comcast was built through a series of acquisitions of small, then progressively larger, cable operators. From 1986 through the AT&T Broadband acquisition in 2002, the company methodically built up its core cable company, becoming in our research a top-tier performer and a model for how to use acquisitions to profitably grow. When it made non-cable investments, such as the Golf Channel and the E! Entertainment channel, it stayed close to its core cable business.

Making the Disney deal succeed will require a very different set of capabilities. Indeed, when companies abandon their core and start looking for alternative platforms of growth, it often signals trouble. There is portent in Comcast's announcement, but it probably is not the message that Brian Roberts wants to send.