The Surprising Inefficiency of Field Deployments

Dipak Bhattacharya

Most of us believe that the number of active prescribers grows continuously after launch, plateauing shortly before patent expiration. This orthodoxy is driven in part by the belief that most of our growth comes from new prescribers. Consequently, it makes sense to cast our nets as broadly as possible at launch. As we move further from launch and prescribers make their decision to either adopt or reject a new technology, this orthodoxy is less defendable. Without new evidence of benefit or a new indication, it is not clear what would trigger adoption. Continued promotional investment beyond this point could be considered wasteful.

So why are so many launch deployments steadfastly maintained over time? The answer can be traced back to a pair of cognitive traps. The first is the confirming-evidence trap.The prevailing wisdom in the pharmaceutical industry is that launch deployments should strive to cover ~80% of the prescribing potential in a given category. The reason for this is that launch deployments that exceed this threshold quickly experience diminishing returns. In essence, they add a large volume of low value prescribers that undermine profitability. When drug launches exceed expectations, companies maintain these footprints and sometimes expand them, and managers take note of this dynamic associating success with broad deployments.

The other dynamic that managers take note of is the tendency of new prescribers to fuel a disproportionate amount of launch growth.On average, 100% of growth in year one and a significant portion of growth in year two comes from new prescribers (see Figure 1), which naturally leads managers to associate success with new prescribers. These dynamics play an important role in shaping beliefs and the confirming-evidence trap leads managers to search for, interpret, and favor information that supports these beliefs.

The second cognitive trap that leads managers to maintain their initial deployments is the status quo bias. The source of this bias lies in our desire to protect ourselves from emotional harm. When we take action, we take responsibility and this exposes us to criticism and regret.In the business context, doing something that fails tends to be punished far more quickly and severely than doing nothing. This explains why so many mergers ultimately founder. Acquiring companies avoid taking swift action and legacy structures and processes become entrenched.

Figure 1

The antidote for these traps lies in reframing deployment decisions as time dependent.If we take manager’s tendency to associate success with new prescribers, we need to widen their lens to see how their contribution evolves over time.As we see in Figure 1, the proportion of new customers peaks in year one and is then overshadowed by a growing base of existing customers. New customers ultimately represent the minority of total customers over time. In addition to these numerical realities, research by Everett Rogers on the diffusion of innovation suggests that later adopters are more likely to discontinue innovations than are earlier adopters. This suggests that growing market share with existing customers can be a far more profitable strategy over time than investing disproportionate resources prospecting for new customers.

Another way to reframe deployment decisions as time dependent is to study the evolution of active prescribers. As a firm specializing in data and analytics, we observed a number of field deployments in which the majority of field resources were deployed against non-prescribers. While this makes sense during the immediate post-launch period, it is less defendable as brands mature. To understand when the base of active prescribers matures, we conducted a comprehensive claims analysis of drugs launched in the U.S. between 2014 and 2023. We excluded drugs with missing data, drugs with fewer than 30 prescribers, and drugs with less than 37 months of data that had not reached an obvious peak. This left a sample of 248 drugs which we further segmented into hematology/oncology, other specialty, and primary care categories. Our analysis confirmed that the base of active prescribers for new drugs peaks 3.5 years after launch and that this pattern is consistent across categories (see Figure 2).

Figure 2

Our research suggests that the prescriber base for most brands plateaus relatively quickly after launch and then begins to decline. Without new evidence of benefit or a new indication, it is not clear what would trigger adoption. Continued promotional investment beyond this point could be considered wasteful and this presents a resource reallocation opportunity for companies willing to think differently. If we consider a specialty brand entering its fourth-year post launch with $20 million in direct selling costs, over half of its calls deployed against non-prescribers and no new indications pending, that brand could reasonably reallocate $10 million in expenses with no material impact on growth. For an industry wrestling with decreasing R&D productivity and the existential threat of government price cuts, this seems like a remedy capable of offsetting these looming threats.