Marketing Secrets: A Conceptual Model and Quasi-Experimental Study: An Abstract

Abstract

Secrets involve deliberately hiding information from other people. They are widely used in marketing: from “secret” recipes and ingredients of Heinz ketchup to the notably vapid trailers of the “Game of Thrones” latest season. In this paper, I set out to develop a conceptual model for the use of secrets in marketing and support the model with a naturally occurring quasi-experiment in the context of the computer games industry.

The very fact of having a secret involves a critical distinction between those in the know—insiders—and those from who the secrets are kept, outsiders (Simmel 1950). I argue that keeping a secret about soon-to-be released products might be a source of value for both insiders and outsiders. From the outsider’s perspective, the effect is based on the positive customer uncertainty and optimism bias (Tanner and Carlson 2009). Brand equity plays a moderating role as brand knowledge decreases perceived risks (Keller 1993), thus performing a positive framing of uncertainty. That is, the stronger the brand, the more value it can create by going secret.

In computer games industry, publishers routinely invite customers to participate in either “open” or “closed” beta testing of their products to get a feedback and fix the software bugs. In the “closed beta,” participants are required to sign nondisclosure agreements that prevent them from sharing any information about the product. I analyze the use of the “open” vs. “closed” beta testing strategies as a naturally occurring experiment. The hypothesis is that the leading brands are more likely to choose the closed beta strategy, and this choice is more beneficial for their new products.

The multivariate ANOVA employs a type of beta-testing and a brand’s TOP-10 status in the industry as independent variables, sales volumes and customer ratings as dependent, and price as a covariate. The results support the hypothesis. The between-subject analysis reveals that there is a significant main effect of beta testing on sales volumes (but not on customer ratings) and an interaction between brand equity and beta testing strategy.

These findings extend the consumer behavior literature on positive uncertainty (Lee and Qui 2009) and contribute to the discussion on new product pre-announcements. They depart from the view that the companies must seek to reveal as much information as possible early (Ofek and Turut 2013) and demonstrate that doing the opposite might be even more effective.