Standardized survey-based concept test
In an ever-evolving competitive landscape, new product development (NPD) plays a vital role to maintain a company’s competitive position in the marketplace.
Brands are continuously pressured by consumers to develop new products and services that meet their changing preferences, combat increasing competition, and diversify risks. This is particularly true in fast-pacing sectors such as Fast-Moving Consumer Goods (FMCG).
FMCG marketers are most responsible for identifying opportunities within their product portfolio and conduct research to ensure they push forward products with the highest success rate. FMCG marketers use a wide range of research techniques to test new concepts. Traditionally, standardized survey-based concept tests have been widely adopted by marketers as they allow for measuring new concepts against existing and competitor products and services.
Marketers work alongside market researchers to identify what products should be tested in combination with new concepts, and to establish key performance indicators (KPIs) that will determine a winning concept.
KPIs often include metrics such as product or service appeal, innovation, relevancy, purchase intent and overall preference. By measuring scores across these KPIs, marketers and market researchers can determine whether an NPD is perceived as better or worse than existing ones.
These KPIs are typically measured using scale (i.e. Likert scales) and rank questions, or by asking respondents to select their preferred choice among a set of options. These traditional standardized survey-based techniques present numerous benefits, most notably, they provide quantifiable, comparable, and cost-effective results.
Exhibit 1: 5-Point Likert Scale
In addition, marketers are increasingly commissioning concept test studies to research agencies that have built benchmark databases. These databases contain hundreds of past tests that are combined to provide average scores across KPIs and ultimately provide recommendations on whether a new concept should be pushed forward.
However, these traditional techniques carry inherent limitations. Even when respondents are presented with multiple concepts at the same time, the use of scale questions for each concept mean they are measured independently.
Another limitation is the lack of real-life context when testing products. In real life, consumers are presented with multiple products that include different features such as brand, pack size, flavor, price, etc. and are therefore forced to make trade-offs when making their choices. Marketers often combine standardized survey techniques with virtual-shelf survey tests to measure product preference in a closer to real-life context.
Despite the benefits of combining these two techniques to measure product preference among consumers, they still present a significant drawback. None of these methods allow marketers and researchers identify the importance of each of the product features in driving the consumer choice.
There is an alternative research methodology that delivers the benefits of the abovementioned concept test techniques while it also measures the importance of each feature that makes up a product. Conjoint analysis.
What is conjoint analysis?
Conjoint analysis is a survey-based research technique that allows to measure product features independently of each other. There are different types of conjoint analysis, this article focuses on the most popular type of conjoint analysis called Discrete-Choice Conjoint Analysis.
In Discrete-Choice Conjoint Analysis, respondents are presented with different concepts, detailing their different features and are asked to select their preferred option.
Exhibit 2: Discrete Choice Conjoint Question
Each feature, or attribute, (e.g., screen size) consists of different levels (e.g., 75”, 60”, etc.). Answer options or concepts are created by combining all levels across all attributes.
Unlike traditional survey-based techniques, Discrete-Choice Conjoint prompts respondents to make a choice, and implicitly a trade-off among several features.
Every time a respondent selects an option, they are by default selecting a specific level. Therefore, when combining all respondents’ answers, Discrete-Choice Conjoint can then identify the importance of each attribute and level (also called utility value) based on the number of times an option with that respective level has been selected – this requires the use of advanced statistical analysis not covered in this article.
Exhibit 3: Utility Value
As a result, this technique can not only predict the optimal combination of levels that maximize consumer preference, but also subsequent optimal combinations. This is critical to marketers who, in the real world, are often limited by what they can offer to consumers due to lack of resources or investment. Discrete-Choice Conjoint therefore allows marketers to identify alternative options to bring to the market based on different product combinations.
Due to its benefits, the use Discrete-Choice for concept test among marketers has accelerated in recent years. Such growth in adoption has been further supported by developments in survey and conjoint analysis software from companies such as Sawtooth and Conjointly and QuestionPro. These platforms not only allow users to conduct Discrete-Choice Conjoint studies with minimum effort and knowledge, but also offer fully automated analytical tools to identify optimal options in real-time.
Regardless of the methodology, the success of new product development is also driven by external (and often unknown) factors, whose effects cannot be measured or factored in the research such as the distribution of the new product, new competitor products to the market or changing socio-economic conditions.
And while Discrete-Choice Conjoint Analysis cannot mitigate those unknowns, it provides marketers with diagnostics at the lowest product, hence helping marketers to optimize the product makeup and maximize chance of success before it is launched to the market.