Very few companies seem willing to put their targeting programs to a test. Fortunately one did.

The adjacent charts show the result of testing the response of the Magellan target group (Test Group) against the total population of GP's represented by a sample of over 5,000 validated responses provided by a third party market research organisation (Control Group) in 2004.

One graph shows the volume of recommendations and/or prescriptions for two forms of the same brand of a product that is available both as a PBS and an OTC item. Note that one form was relatively new to the market and started out with effectively zero market share.

The test group called almost exclusively on the 5,500 to 6,000 odd GP's that were ranked as high potential for initiation by the Magellan algorithm. Such exclusivity was possible through the use of a contract or "project" team of representatives.

Whilst IMC is unable to disclose if the calls were first, second, third position, sample drop or whatever; it is clear that the test group of Magellan doctors were more responsive to promotion whilst the results for the control group, based on a sample of over 5,000 GP's is less responsive.

Volume Per GP Writing - Brand A

Note that market research results like these can experience an error tolerance of +/- 4% at 95% confidence intervals. Such an error tolerance is equal or superior to assumptions made for example, about the voting intentions of a population in the broadsheet press in capital cities around Australia. It is thus a commonly accepted method of auditing a segmentation and targeting activity.

Volume Per GP Writing - Brand B

Studies done in the UK show that contract or project sales teams tend to have higher performance indices than in-house sales teams. Whilst there are many extrinsic factors that may influence this as well, no responsibility is taken by IMC for any assumption made from the use of this data. It is the result of an actual research project and IMC asserts that the targeted activity represented by the test group was highly likely to be the result of a well populated target list, rich in concentrated and objective third party doctor specific information.