The consumer testing of food package graphic design

This purpose of this paper is to present a technique for measuring the impact of package graphic designs on consumer choice and purchase intentions. The technique allows several proposed graphic designs for the same product to be compared, both with one another and with their immediate competitors up to a maximum of eight designs.

The approach is based upon a fractionally replicated Latin square design with two sets of four package graphic designs and five consumer groups as the independent variables. The dependent variable is consumer choice.

Three fully worked examples of the test are presented. The results show that graphic design has a massive and statistically significant impact on consumer choice, ceteris paribus. The test is able to discriminate strongly between package graphic designs that only apparently differ in detail.

The results show that graphic design has a significant impact upon consumer choice, but it also shows that this impact is highly product and situation specific. This may have considerable implications for any claimed wider “empirical generalisations” relating to the impact of graphic designs, or components thereof, on food consumer choice.

The methodology presented allows graphic designs to be tested before they enter the marketplace. It replaces opinion with a reliable and quantifiable output that can be easily interfaced with econometric and market share models that have been developed within the choice modelling literature.

This appears to be the first paper to appear within the literature that presents a reliable, simple and easily administered test that is capable of quantifying the performance of alternative graphic designs on food packaging.