OR WAIT null SECS
In-market experimentation is the best way to address the challenge of pitching brand leaders on new ways to reach their target population through digital and other non-personal channels, write Scott Beauchamp and Michael Calamari.
There is an ever-increasing number of vendors pitching brand leaders on new ways to reach their target physician population through digital and other non-personal channels. Understanding how each of these potential levers fits in with broader commercial strategy - at different stages of a drug’s lifecycle - is a complicated and critical challenge. As in other industries like financial services and retail, where digital strategies have been crucial for some time, in-market experimentation is the best way to address this challenge.
Finding the right balance between more expensive personal and potentially more cost-effective non-personal investments is especially germane in a cost-constrained environment. Many leaders’ default reaction to allocating resources is to turn to their promotion response models, which seek to reveal the ROI of each investment option. While these models provide insights, they do not accurately isolate the impact of any individual promotional tactic, providing brand leaders with broad ideas without actionable recommendations.
This shortcoming is rooted in a failure in understanding how prescribers are predisposed to act as opposed to the extent to which they will change their behavior because of a commercial action. Consider a scenario in which a drug receives a new indication. In response, the brand team:
Conducts in-person and remote training for its sales force
Begins in-person details with visual aids
Invites physicians to a new series of speaker events
Sets up a new digital physician portal
Launches a DTC advertising campaign
Buys ad space on EMR portals
Hires a contract sales force to target lower-decile physicians
Within a couple of months, some of these healthcare providers begin prescribing the new therapy. Yet it is unclear which tactic caused that behavior. Which physicians would have prescribed anyway without the increased attention? Even more complicated, how do sales actions affect non-personal actions, and vice-versa? The challenge of isolating both the individual impact and interaction effect lies in the following complications:
• Opt-in (selection) bias: Providers that brands prioritize are very rarely representative of the rest of the provider network – brands are likely to target providers with the most perceived upside. As a result, it is hard to discern the extent to which these commercial actions generate new scripts as compared to the scripts that those physicians would have written anyway.
• You can’t model what hasn’t been done: Regression-based models are backwards-looking. Evaluating strategies taken during a different stage in a drug’s lifecycle will not accurately predict what will happen in the current environment. Further, new actions, such as untested digital strategies, cannot be modeled. This reality applies to both brand new actions, as well as significant increases in investments in existing channels. For example, if existing spend on EMR ads is $5MM, understanding the ROI of doubling that investment could be highly inaccurate.
Increasingly, pharma leaders are realizing that in-market experimentation needs to be part of their commercial decision-making process. This process - pioneered on the R&D side of pharmaceutical organizations -involves trying an idea with some providers or in some markets and comparing the performance of the group receiving the action versus a highly similar group operating business as usual. The resulting analysis reveals the cause-and-effect relationship between the tested action and KPIs.
Recently, a brand leader at a top 20 pharma company wanted to understand which digital strategies she should invest in, and how the outcome of those strategies would vary based on other commercial tactics, such as in-person details. To answer this question, the brand leader tested each of five different digital channels with a select group of HCPs. Resulting test vs. control analysis revealed the incremental impact of each channel: only two of the five tested options met the desired hurdle rate, despite claims from the vendors about enabling the brand to reach its target customer population. This analysis also revealed how digital investment interplayed with in-person details, showing that digital engagements were significantly more effective for physicians who had not received a high frequency of details in the previous months. As a result of this test, the brand leader was able to focus her non-personal spend on the highest-ROI channels while also informing the right level of sales investments. Further, the brand was able to use this same test vs. control approach to evaluate its other levers, such as speaker programs. This process enabled them to develop a perspective on the optimal commercial mix based on real world, cause-and-effect insights.
As reaching providers in-person becomes increasingly difficult, lack of coordination between field and brand teams (e.g., making a change in detail frequency without coordinating with the digital team) can be detrimental to commercial strategy. A key part of the coordination solution is having an agreed-upon, evidence-based measurement methodology that provides an objective lens for understanding the optimal mix of promotional tactics.
Developing the capability to conduct accurate real world test vs. control analysis - and then a joint team to act upon the insights - is the first key for organizations that want to take their commercial strategy to the next level. Organizations can get started today by identifying natural variation in their tactics (e.g., some physicians engaged in a digital communication and others did not; or, some physicians stopped receiving the same frequency of details). With the right analytic capability in place, organizations can measure this natural variation on a test vs. control basis to reveal cause-and-effect insights, despite not having purposefully designed tests. Leaders can then take multi-channel coordination one-step further by developing a commercial testing pipeline that will enable them to rapidly learn which strategies are effective, and importantly, yield insights about the extent to which each channel complements or harms the other. For example, do non-personal follow-ups to a speaker event generate the desired response? How does that vary by physician specialty?
Becoming an experiment-driven commercial organization ultimately empowers brand leaders to manage the business more proactively by enabling them to try more ideas without unnecessary capital risk. Retailers, restaurants, banks, and hotels have been effectively reaching their customers across channels for years and are leading the way in using in-market experiments to inform their marketing mix. It’s time for pharma leaders to climb on board.
Scott Beauchamp is Vice President, and Michael Calamari is Principal, both at Applied Predictive Technologies.