New Pre-Launch Test Calculates a Movie's Box-Office Success

Model could help movie industry fine-tune ad strategies

THE secret to Hollywood's blockbuster box office sales so far in 1993 is simple: The movies are good.

But in an industry where ``gut instinct'' is still the order of the day, a new mathematical model, developed to predict a movie's sales prior to its advertising and promotional leadup, could become the crystal ball of the motion picture industry.

``This whole notion of understanding better consumers' enjoyment - trying to predict what kind of entertainment experience they're likely to enjoy more or less - is a very hot and timely topic,'' says Jehoshua Eliashberg, a professor of marketing at the Wharton School of the University of Pennsylvania. This new model, he says, could save the movie and other entertainment industries a lot of money.

Professor Eliashberg developed the model over three years with Mohanbir Sawhney, assistant professor of marketing at the Kellogg School of Northwestern University in Evanston, Ill.

The movie industry has shown interest, Mr. Sawhney says. One studio, which he declines to name, plans a test of the model within the next six months, he says.

The model assumes that consumers go through various mental stages in deciding to see a movie. In measuring a movie's prospects, the model uses pre-release advertising, such as newspaper and television ads, simulated positive and negative word-of-mouth conversations derived from movie clips or sneak previews, and the reaction of a test audience to the movie.

These factors are then converted into a mathematical formula that translates consumer attitudes into movie attendance. This in turn is translated into dollars.

The model could be used for any new entertainment product such as compact discs and TV shows, where word of mouth and advertising have a big impact on the consumer's decision, Eliashberg says.

The researchers tested their model earlier this year on two quite different movies, ``Groundhog Day'' and ``The Cemetery Club,'' both released nationally on Feb. 12. Using 140 students at the University of Pennsylvania, the researchers projected movie attendance on the university campus.

Although the professors only tested their model on college students, their national box-office projections (determined mathematically from the university attendance) were considered fairly reliable because 16- to 24-year-olds constitute 70 to 80 percent of moviegoers.

The results: The model projected gross sales of $67.2 million for ``Groundhog Day'' and $8 million for ``The Cemetery Club.'' By July 8, ``Groundhog Day'' had made $70.6 million and was near the end of its run, while ``The Cemetery Club'' had grossed just over $6 million.

Along with predicting a movie's box-office performance, the model also provides useful information on the ``likely consumer response'' to the movie's pre-launch promotion. This would allow studio executives to fine-tune their advertising and distribution strategies before the movie is released.

For example, the study should help a movie marketing manager select scenes for a television ad; edit the movie to appeal to a targeted audience; reposition the movie's promotion to target a different audience (``Ghost'' was variously positioned as a movie about reincarnation or a romance); and plan the media strategy and the intensity of the distribution.

After analyzing the marketing strategies of the two movies, the researchers found that increasing advertising would not have helped ``The Cemetery Club;'' but doubling the advertising budget for ``Groundhog Day'' could have increased its box-office profits by 46 percent.

Other ways to assess a movie's likely performance include test markets, which describe how well a movie performs in a small town. Sneak previews are also used, mainly to assess consumer likes and dislikes of various scenes. But none of these methods provide dollar projections at the box-office.

One major advantage to using this new model is cost, Sawhney says. Current tests run between several hundred thousand dollars and $1 million. It would cost between $25,000 and $50,000 to test their model, he says.

The model also departs from earlier tests by including the effect positive and negative word-of-mouth conversations have on the consumer's decision to see a movie.

``Every film is a new brand to introduce and studios have only one chance to do it,'' says Hollywood analyst A.D. Murphy in a supplement to Adweek Magazine. ``After opening day, it's strictly word of mouth that sells a movie,'' he says. ``The only function of advertising after that is to remind people, `Hey folks, we're still here.' It was word of mouth that made `Jurassic Park.' ''

In its present form, however, this new model would not work with children's movies, such as ``Aladdin,'' where often both the child and the parent are involved in the decision to see the movie.