Promotion-Response Models to Optimize your Portfolio (download presentation)

SRI’s 3rd Rx & Biotech Portfolio Management Conference, Hyatt Hotel in Penn Landing, Philadelphia, Mar 25-27, 2002.

Abstract: What is the promotional investment each drug of your portfolio deserves? In the event of a 25% budget cut, how should that cut be apportioned? From a promotional mix standpoint, how should the investment be broken out by channel (detailing, sampling, meetings and events, journal ads, DTC, other A&P)? In a WW setting, how should the spend be apportioned by geographic area or country? Whatever the variation may be, we are looking at the same fundamental and emotionally-charged budget allocation question that lies at the heart of corporate planning (annual budgeting, long range planning, development of marketing programs, definition of company strategy). This allocation question is all the more heated since it may mean, with the overall investment left untouched, millions of windfall or forgone dollars, not to mention made or broken careers.

The basic idea hinges upon the notion of marginal return. Invest in each drug until the marginal return from that drug is equal to your opportunity cost. If this leads to a total portfolio investment that exceeds your budget (and borrowing is not an option), cut down on the investment while making sure the marginal returns from all drugs are equal. Indeed, if the marginal return from drug A is lower than that from B, transfer investment from A to B (steal from the poor and give to the rich). The theory is quite straightforward. It is its implementation that requires skill and finesse. First, what does return mean? Return may be regarded as profit or sales, or any combination of the two. Also, sales and profit can be current-year, short-term or long-term. What expenses need to be included or excluded, or what needs to be considered as fixed or variable may have to be clarified. Second, this approach requires that we establish response models for the drugs of the portfolio (a response model describes how sales of a drug vary with promotional investment). This paper explains how to construct response models and discusses several techniques in that regard with special emphasis on keeping GIGO (garbage in – garbage out) at bay: Delphi sessions, scenario planning, competitive SOV/MS benchmarking, and aggregation of physician-level response models. 

 If you would like to download this presentation, give us a call at (847) 920-1000 or email us at bayser@bayser.com.