Pharma commercial analytics is the data and software that a drug company uses to forecast a medicine's sales, plan its sales team and measure its marketing. It brings prescription data, sales data and market data together in one place. Much of this work is done in Indian analytics centres, for drug companies in the United States and Europe as well as in India.
The data comes from several sources. Sales data shows how many packs moved from the company to stockists, the distributors, and from stockists to chemists. In India these are called primary sales and secondary sales. Prescription data, bought from market research firms, shows which doctors prescribe which medicines. Visit data comes from the company's customer relationship management (CRM) system. It records every visit by a medical representative (MR), the company employee who visits doctors to explain its medicines. Market data comes from pharmaceutical intelligence databases.
Imagine a drug company in Mumbai launching a new diabetes tablet, with 1,200 MRs across India. Before the launch, the platform builds a forecast with time-series forecasting models. The models use how similar tablets sold in their first two years, the number of doctors who treat diabetes, and the price. Say the forecast is 60 lakh strips in the first year. The forecast decides how many strips the factory makes, and how many MRs the company needs. Next, the platform decides which doctors each MR should visit, through sales territory planning. It sends each MR's list of doctors to the field sales software on the MR's phone.
After the launch, the platform measures what works. Each week, it compares secondary sales in each territory with the MRs' visits there. Say sales in one district are 40% below the forecast, although the MRs visited every planned doctor. The analysts then look closer. Perhaps the chemists there have not stocked the tablet yet. The platform also suggests which doctors to reach online, and through which channel. This is called next best action.
In many of these tasks, a model drafts the answer and an analyst checks it before anyone acts on it. This way of working is called human in the loop. A wrong forecast is costly. Too few strips means empty shelves at chemists, and too many means stock that expires in the warehouse.
Pharma commercial analytics is different from a general customer data platform (CDP). A CDP tracks consumers who buy directly from a company. A drug company's main customers are doctors, who prescribe a medicine but do not buy it. So the data in pharma commercial analytics is built around doctors, prescriptions and visits.