Conjoint analysis is one of the most widely used and powerful quantitative methods in market research. It helps uncover how people make choices and what they really value in products and services. It is commonplace in marketing, advertising and product management, helping test acceptance of new product features, pricing, messaging. These days, it is also frequently used in healthcare, environmental, government and non-profit settings.
The most common type of conjoint analysis, called choice-based conjoint, involves presenting people with choices from several product concepts and then analysing the drivers for those choices. The output from conjoint analysis is measurement of utility or value. It is perfectly suited for answering questions such as:
The utility scores (also known as partworth utilities) are used to build simulators that can forecast market shares for a set of different products offered to the market. Through modelling (simulating) people’s decisions, you can find optimal features and pricing that balance value to the customer against cost to the company and forecast potential demand in a competitive market situation.
Conjoint analysis is a robust methodology that has been developed since the 1970s. It far surpasses its alternatives, such as SIMALTO and self-explicated conjoint, in its predictive power. Thanks to Conjoint.ly, you do not need to dive deep into technicalities of the methodology that desktop software tools require. And you can rest assured in the quality of the analysis and full functionality.
The foundation of conjoint analysis is breaking a product or service down into its components (they are called attributes and levels) and then testing combinations of these components in order to find out what customers prefer. It is then possible to estimate the value (also called “partworth utility”) of each component of the product in terms of its effect on customer decisions.
For example, a smartphone may be described in terms of attributes such as brand, screen display, colour, and price. Each of these attributes is broken down into levels - for instance levels of the attribute for screen display might be 5”, 5.5”, 6”.
Rather than merely asking respondents what they like in a product, or what features they find most important, conjoint analysis employs the more realistic task of asking respondents to choose between potential product concepts, which are combinations of attributes and levels. These product concepts are then carefully assembled into choice sets. Each respondent is typically presented with 8 to 12 choice sets (or questions).
The process of assembly of levels into product concepts and then into choice sets is called experimental design and requires a substantial deal of statistical expertise. Conjoint.ly automates this process, using state-of-the-art methodology. You can specify the number of alternatives (concepts) per choice set, the number of choice sets per respondent, and other settings when you set up an experiment.
Then respondents go through the conjoint survey to complete the choice tasks (typically it takes a couple of hundred responses, but may vary depending on the complexity of the study). You get a report that contains: