Development and evaluation of measurement tools for conceptual profiling of unbranded products

• We compare best–worst scaling with a direct scaling method for conceptual profiling.
• Study on six unbranded orange juices.
• Best–worst scaling discriminates better amongst juices.
• A moderated training session greatly enhances discrimination amongst juices.
• Precision of conceptual profile unaffected by measurement technique.

All objects including brands, products and packaging have conceptual (implicit) associations and consequently a conceptual profile. Along with its sensory profile this defines and characterises the object. Together these influence our attitudes, our affective reactions (feelings and pleasure) and our behaviour towards it. The conceptual content of a brand is usually clear and readily accessible via established brand personality measures. However, until recently, the conceptual content of product per se has received scant attention.The conceptual content of an unbranded product derives from two sources: the fundamental nature of the product category (‘category effect’) and within-category sensory differences amongst related products (‘sensory specific effect’). The latter leads to subtle variations in conceptual content that are important in the context of product optimisation but may be difficult to measure.Best–worst scaling (BWS) is an indirect method of scaling that has been found to be particularly useful for measurement of ‘soft’ or abstract attributes that are not easily quantified. It has been applied previously to conceptual profiling of brands and products. This study compares the utility of BWS versus a direct rating method (an online technique known as bullseye) for accessing the conceptual content of six unbranded orange juices. Degree of familiarisation with the research process and the juices (intensive familiarisation versus a simple warm-up) was added as a second variable, thereby creating four methodological cells to compare.Irrespective of familiarisation protocol, BWS proved to be more effective than bullseye in eliciting the ‘sensory specific effect’. However the biggest effect was associated with the familiarisation protocol, where intensive familiarisation engendered more effective discrimination amongst the juices than the simple warm-up, irrespective of scaling methodology. We conclude that intensive familiarisation is of great benefit to the conceptual profiling of unbranded products for product development.Within a product’s conceptual profile, BWS and bullseye discriminated amongst the conceptual terms to a similar degree. Previous studies have tended to find that BWS gives greater discrimination amongst the choice items than rating scales. We suggest that the nature of the choice items and the number of decisions required from participants both affect the relative discrimination of the two methods.