Recent research indicates that most of the value created in the pharmaceutical industry comes from marketing compounds that improve on the effectiveness of an established drug, have fewer or less intense side effects, or can be used against a broader range of diseases. But which of these follow-on strategies works best? In many therapeutic classes, the latter two can drive commercial success more often than the former, according to a McKinsey study.

Our study also confirmed the idea that the first drugs to treat an ailment often fail to create more value than their later-to-market competitors do. We found that fast- followers—drugs marketed 2 to 5 years after their first-in-class predecessors—typically produce significantly more revenue. Products that reflect a strategy of differentiation and are launched 5 to 15 years after the original drug, and latecomers that reach market more than 15 years later, have generated present-value sales as high as those of their first-in-class rivals (Exhibit 1).

Moreover, being first isn't essential to achieve blockbuster status (defined as sales greater than $1 billion a year). Of the 32 such drugs introduced over the past decade, only a quarter targeted novel mechanisms of action—the way a compound acts at the molecular level in the body to achieve a desired health outcome. Some of these were "me-too" compounds, but a majority had at least one source of significant clinical differentiation: their efficacy, or comparative effectiveness in treating a specific disease; safety, including issues such as side effects and drug interactions; the breadth of their approved uses; and convenience, including longer-lasting formulas and easier application.

But in a challenge to conventional wisdom, which suggests that efficacy is the best source of clinical differentiation, our study found that other attributes, such as safety, were the winners. In fact, over the past decade, improved efficacy has been the primary driver of commercial success in only two of ten therapeutic classes: cholesterol reducers and antiulcer treatments. The reason, in part, is that many drugs are already highly effective, and attempts to improve on them can be costly and inconclusive.

By contrast, many companies have had more success with efforts to improve the safety of products or to get them approved for use in treating additional diseases. Examples are Pfizer's Neurontin, a successful antiepilepsy medication driven predominantly by perceptions of safety differences, and the company's Celebrex, a Cox-2 inhibitor that reduces pain and inflammation and is thought to be easier on the stomach than aspirin.

The breadth of approved uses for drugs can also be a significant source of differentiation. We found that nearly two-thirds of those in competitive therapeutic areas, such as depression and allergies, went on to receive regulatory approval for a broader range of treatments after they came on the market (Exhibit 2). Eli Lilly's Prozac, for example, was launched in 1987 to treat depression but has since been approved for use in a range of conditions. Today, drugs entering these competitive areas often need to have a similarly broad range of regulatory endorsements at the time of their launch, thereby raising the bar for compounds in development.

While companies often want to prove that their pharmaceuticals are distinctive in several respects, we found that best-in-class drugs were usually standouts in a single attribute. Of the recent blockbusters we analyzed, 59 percent were superior to the competition in only one area, while just 14 percent were superior in two, and no drug was the leader in three or more (Exhibit 3). While some of the remaining 27 percent were distinctive at launch, they are now comparable to their competitors in all significant dimensions but continue to be marketed more effectively. Yet drugs with even a single subpar attribute were generally commercial failures.¹

These findings suggest that pharmaceutical companies might rethink their strategies for clinical trials. The first step is to strike a better balance between research into novel compounds and best-in-class differentiation. Too many companies commit a disproportionate amount of their time and money to discovering first-in-class compounds when best-in-class opportunities might yield higher returns. Second, efficacy shouldn't be pursued at the expense of other sources of differentiation; claims of safety, pharmacokinetics (that is, dosing), and convenience may be as desirable and more clinically attainable.

Third, managers should seek one meaningful point of differentiation and scuttle drugs that have even a single subpar attribute. Last, the clinical-trial strategy for a drug must be infused with a long-term perspective on which diseases it will be approved to treat and when, particularly in primary-care areas such as cardiovascular or arthritic pain. A drug that is too narrowly defined risks being outflanked by competitors, though efforts to get a drug approved for too many diseases can jeopardize the cost-effectiveness and timeliness of the development process.