What can you do with tracking studies and what are their limitations?
Neil Barnard
Partly based on work with Ehrenberg and Barwise, this paper notes that tracking studies (which can cover a wide variety of advertising and brand measures) tend to be very stable over time. Brand attribute scores, particularly for established brands, often do not reflect planned advertising responses, favour dominant brands, and any changes tend to follow rather than lead shifts in buying and share. 'Likely to buy' measures other than for new products are often out of step with actual behaviour. By contrast, advertising measures are responsive...