Affinity Analysis »
By MC on Sep 24, 2007 in Personalization | 0 Comments
The technology, Affinity Analysis, typically exceeds an 85% accuracy rate in predicting how well consumers will like specific entertainment, advertising or consumer product offerings.
MediaChoice offers a revolutionary change to the process of target marketing and research. It makes use of a complex, yet transparent process to determine the ranking and relevance of personal ratings. The foundation of this intuitive technology is the ability to determine item affinities with a high degree of accuracy and extract actionable information based on those affinities.
Other research tools and prediction systems stem from characterizing users such as collaborative filtering, factor analysis and vector mapping. None of these systems reliably predict a consumer’s reaction to specific entertainment, commercials, packaged goods, etc., where Affinity Analysis excels.
Media planners and researchers who must base their decision entirely on the blunt instruments of behavioral and demographic studies may find that their top choices are either prohibitively expensive or consume their budgets too quickly. MediaChoice may pinpoint some less expensive alternatives, where consumers are already predisposed to like the product being promoted. Negative affinities may also be significant. A negative affinity between two items indicates that consumers who enjoy one item tend to dislike the other, and vice versa. Even if the demographics appear promising, it obviously makes little sense to expend scarce media dollars to advertise a product to consumers who are statistically inclined to dislike it.
MediaChoice generally uses a simple questionnaire template that can be added to most phone, mail, mall intercept, theater or on-line research studies. This saves time and money for our clients. MediaChoice also retrieves data continuously from Internet users as well as representative samples fielded several times a year.
Technorati Tags: affinity analysis, target marketing, consumer ratings, collaborative filtering, prediction system
