Musings on Entrepreneurship and Innovation
Monday, October 16, 2006You Can't Schedule Cosmic Events
I recommend In Era of Blockbuster Books, One Publisher Rolls the Dice in today's The Wall Street Journal (registration required). It is the story of a publisher's failed attempt to engineer "the next 'Da Vinci Code'."
Consider the following:
- A book is composed entirely of information. As a consequence, the industry clockspeed is extremely rapid. The lifecycle of a novel on the shelf is most often measured in weeks.
- Books are experience goods. Furthermore, in order to gauge the experience, a consumer must invest a considerable amount of time. Consequently, potential customers rely heavily on word of mouth and information cascades.
- As information products, books can be readily and rapidly duplicated. This helps to exacerbate the impact of small differences between books. In other words, books are highly sensitive to initial conditions and are path dependent.
It seems logical that big marketing budgets, strong early reviews, and broad distribution can cause a book to become a blockbuster. The publisher featured in the newspaper article seems to make that assumption when he asserts, "blockbusters have to begin big." However, consider the following from Arthur De Vany's careful analysis of the movie business in Hollywood Economics:
The evidence rejects the model that a big opening is necessary or sufficient for a movie to gross large box-office revenues. The statistical herding on which this model relies is also clearly rejected...By making strategic choices in booking screens, budgeting and hiring producers, directors and actors with marquee value, a studio can position a movie to improve its chances of success. But, after a movie opens, the audience decides its fate. The exchange of information among a large number of individuals interacting personally unleashes a dynamic that is complex and unpredictable. Even a carefully managed and expensive marketing program cannot direct the information cascade; it is a complex and stochastic process that can go anywhere...Revenue forecasts have zero precision...The audience makes a movie a hit and no amount of "star power" or marketing hype can alter that.
After decades studying the movie business, De Vany echoes screenwriter William Goldman's sentiment that "nobody knows anything." As to the book publishing business, one noted author observes that You Can't Schedule Cosmic Events and the featured publisher notes "book publishing, for all its planning, remains a roll of the dice."
Books and movies may well be extreme examples of information rich, easy to replicate, consumer experience goods. However, as the information, design, and emotional content of context-specific consumer goods increases, and as product development and manufacturing capacity becomes more broadly distributed, I suspect that it's likely that the dynamics of an expanding set of consumer products will behave more like the book and movie businesses. There is already growing evidence that market share distributions across a broad range of consumer products is best described by a power law, which may, at least in part, reflect information cascades among consumers.
The implication for inventors and entrepreneurs is cautionary: we probably don't know as much as we may think we do about about engineering a successful product. We can probably shift the probability distribution, but there is no such thing as a sure thing. De Vany's prescription?
The way to make Hollywood management right more often and to make better movies is to be constantly testing the inside view against the outside view...Pervasive optimistic bias is based on (1) unrealistically positive self-evaluations; (2) unrealistic optimism about future events and plans; (3) an illusion of control...Managers are excessively risk averse in small decisions (low-budget movies) and when they face ambiguous probabilities (unknown or little known artists). The certainty effect means that small projects by unknown or lesser-known individuals are underweighted in comparison to large projects with known individuals...smaller movies are undervalued...The narrow framing of decisions leads to the sure-thing principle. Bold forecasts enable cautious decision makers to take bold risks.
As originators, developers, manufacturers, and marketers of new products, we can improve the chances for success. But, in the end, users determine whether an invention will be adopted and become a successful innovation. And, outside of basic commodities, users must, at least to a degree, discover what they want. All of us, therefore, should be wary of the illusion of control and of strategies based on linear assumptions of cause and effect.
