Featuring: Dr. Chitra Lele, Sciformix Corporation
In 2016 and beyond, analytics-driven strategies will shape the industry’s marketing and sales in a more efficient and cost-effective way. The accessibility of consumer data, combined with tools that can quickly mine that data, will pave the way for more targeted and effective tactics. Former physician profiling and programming buying, big data is the future of pharmaceutical marketing.
Traditionally, pharma has targeted the top-decile physicians on the assumption that the historically highest prescribers also represent those with the highest prescribing potential. This approach doesn’t account for physicians whose prescribing has peaked – where sales calls will yield deminshing returns – nor does it account for physicians in historically low deciles with potential to become high prescribers for the brand. According to Patrick Homer, principal industry consultant, global practice, health and life sciences, SAS, using data to determine predictive behaviors helps more accurately identify high prescribers and can evaluate all variables to create a behavioral profile. This method helps marketers target other physicians who share the same characteristics btu have not yet started prescribing highly. These physicians represent high potential, but have yet untapped value, and, are the most productive targets for promotion.
Having determined the present and/or future value of a healthcare provider at the brand, franchise – group of brands – or enterprise level marketing investments can then be aligned with the appropriate providers. Specific groups are created and receive tailored communications that drive the most incremental scripts for that segment. In the process, pharma companies not only create a lift in scripts, but they understand why these physicians are prescribing and how they respond to sales and marketing activities.
Before being able to accurately identify targets, however, marketers most integrate all of the data sets – sales, prescription, and CRM data; physician and patient marketing data; and patient longitudinal data – that typically sit within different repositories and combine them into a predictive model.