Syntex sized its sales force with a model built from managers' own judgment

Pharma sales forces are expensive, and getting the size and deployment wrong wastes money on the wrong calls or misses the product that deserves coverage.

Instead of running a field experiment, Syntex had experienced managers and salespeople estimate each product's sales response to effort through a structured Delphi process, then built decision-calculus models from those estimates.

The models predicted two-years-ahead sales better than the forecasts Syntex already had, and guided the firm to increase its sales force and reallocate effort. The decision produced a continuing $25 million and 8% annual sales increase, and the project won the Franz Edelman Award.

Model the sales response, then size the team

Estimation is faster and cheaper than a field experiment, and the managers had the knowledge to do it.

Predicting better than existing forecasts gave management reason to trust the model.

Deploying effort to products with the best response raises total sales without more reps everywhere.

The model shifted strategic focus to product markets with better long-term potential.

Managers' own judgment, turned into explicit response curves, can size a sales team before you run a test.

The Syntex project became a classic example of the decision-calculus approach to sales-force planning, and its results were documented in an Interfaces paper. The same technique was later applied widely in pharmaceutical and other fields as a way to plan selling effort before expensive testing.