THE forces that propel a book onto the bestsellers list have been analysed mathematically for the first time. The findings could help publishers use initial sales data to decide which books to back further with advertising campaigns, and which books are not worth the bother.
鈥淭here are two types of mechanisms to sell books 鈥� word-of-mouth and advertising campaigns,鈥� says Didier Sornette from the University of California in Los Angeles. Word-of-mouth is known to produce a slow growth in sales, which eventually peak and then decline slowly, whereas advertising or a good review gives sales an instant but short-lived boost.
Using estimated sales data for bestselling books on Amazon.com, the team found that the shape of the peaks could be modelled using the same equations that predict how epidemics spread. One variable in the equation determines the sharpness of the peaks, and publishers could calculate this parameter from early sales data to figure out how easy it would be to persuade more people to buy a book. For instance, a sharp peak indicates transient interest, so further advertising would be unlikely to sway people. But a gradual decline indicates that the book could benefit from more promotion.
Advertisement
