A potential buyer of a product or service needs to consider many factors when deciding what to buy, and potential sellers must decide on the factors to offer and at what price. Buyers typically place their own economic self-interests first: spend the least amount of money, time, and effort to get the product or service they want. Companies want to satisfy customer needs but also have a competing objective: to find the factors that maximize appeal, revenue, and profit.
A general class of research approaches to deal with these types of intersecting (or conjoined) variables in a structured manner is known as "trade-off research", or more precisely "conjoint analysis". Conjoint analysis has many practical, real-world applications in new product development and design. Robert Walker, CEO of Surveys & Forecasts, LLC and a recognized conjoint analysis expert, notes "Any situation where a client needs to determine an optimal combination of variables for marketing or financial purposes is a candidate for one or more conjoint approaches."
Conjoint analysis involves the reaction to, and statistical relationships between, multiple factors in a choice decision. A factor has specific characteristics, such as features or benefits that vary, called “levels”. Simple examples of factors are price, quantity, and size. Some real-world examples might include:
- An industrial paint sprayer might come with multiple hose lengths, paint capacity, nozzle diameter, or spray wand length.
- A credit card may have different levels of annual fee, airline points per dollar spent, levels of travel protection coverage, or access to airport lounges.
- A laptop manufacturer may offer different screen sizes, keyboard layouts, processor speed, graphics cards, memory, hard disk size, or bundled software.
In conjoint designs there can factors with two levels (i.e., "yes" or "no"), or factors with many more (i.e., five price points, six pack sizes, etc.). As complexity increases (factors, levels), the number of possible product combinations grows exponentially. For example, four factors with three levels each produces 34 = 81 unique combinations. Theoretically, we could show all 81 to a single respondent – but would we want to? Poor data quality and exhausted subjects are two good reasons not to!
Conjoint analysis solves for this problem of "too many combinations" by showing each subject a randomized subset, also known as a partial factorial design (i.e., a subset of a complete factorial design). Features or characteristics of a product or service are sequentially exposed using various forms of conceptual description. Concept features and levels are then varied by subject in the overall matrix of combinations. In simplistic terms, this akin to Swiss cheese: we know what the overall shape is (i.e., the complete design), but there are holes inside (partial exposure). As might be expected, more complex conjoint designs require large sample sizes for statistical reliability.
From the research data we calculate demand for the optimal combination of factors - that is, the set of choices that maximizes consumer appeal. We also can isolate the importance of each factor (such as price, quantity, or size), and identify the optimal combination of factors. Additional and essential deliverables also include demand simulation and revenue estimation for the optimal product configuration. The simulator permits scenario testing using different combinations of features, and using demand scores generated from the research.
When applied against an interested target or segment, an estimate of penetration can be developed, and with additional inputs, potential sales can be forecasted. The design and screening criteria, as well as factors and levels chosen, are major factors in making sure that conjoint results ultimately make sense. Each assignment is based on considerations of the business, decisions that need to be made, and the number and complexity of choices to be evaluated by potential customers.
If you have a new product development roadmap ahead of you at your company, conjoint analysis might be something to consider. For more information, visit the Surveys & Forecasts, LLC website or get in touch at info@safllc.com.
