Speaker 1: So what exactly does conjoint analysis do? First of all, conjoint is a portmanteau for considered jointly, and it's the output of a conjoint analysis that really interests a marketer. There's two basic outputs that we utilize to make all sorts of interesting market predictions. First, for each individual person participating in the conjoint study, we'll have a numerical assessment of the relative importance that they attach to the different attributes of a product. So for example, if we were doing a conjoint study about trying to optimally design a pair of running shoes, we might find that there's a certain type of customer that the weight of the running shoe itself accounts for 40% of the differences in their preferences for running shoes. This would indicate to us that the weight of the running shoe is an important feature that we need to take into mind when designing a new running shoe. Another common output of a conjoint analysis is the value, or utility, that's provided to customers by each potential feature or attribute for an offering. Imagine we were designing a GPS sports watch. The results of a conjoint analysis might reveal the following. A given consumer places about 15 additional dollars of value on extending the battery life of a GPS running watch by an additional six hours. This is very important information. Imagine if we knew that we could extend the battery life of our watch at a cost of $10 per watch. If it turns out an overwhelming majority of our potential customers are willing to pay about $15 for that feature, we know that we have a winner. We take the results of a conjoint study and we merge it in with other prediction models, other assumptions, or other market information. When we do so, we're able to simulate a variety of what-if scenarios for a new product design that might be entering the market. This is useful to help us estimate market shares, revenue, sales, profitability. Let's talk about a few more common uses of conjoint analysis to illustrate its function. Conjoint analysis is common in service and retail design. For example, imagine that we were a gym. To increase our number of monthly memberships, should our new fitness centers have an additional dozen elliptical machines or should it add a dozen new treadmills? Notice because of the fixed-constraint retail space of our gym, we can't do both. We have to make a trade-off between these two options. What about designing a membership? Imagine a private golf course package for a membership fee of $1,000 that gives people access to private courses that they otherwise would never be allowed to play at. A question we might ask is should we expand the number of private golf course offerings from 10 to 15 or should we guarantee our members that they get at least one additional weekend tee time for an existing club? Notice that we're assuming that it would cost us more money to offer more private golf courses or it would cost us more money to be able to offer an additional weekend tee time, but we're not going to be changing the price. Product design is perhaps the most common application of conjoint analysis. Should we use an innovative new composite material to build our softball bat or should we focus on an innovative new bat head design or the shaping of the head to maximize people's preference for our line of softball bats? Again, directing our attention for a particular feature at a particular level, probably at the same price, this is a common use of conjoint. We can even use conjoint studies to optimize the design of actual events or experiences, although this is a little less common. For example, there's an outdoor running relay event called Ragnar. To get people excited and involved, they offer a variety of other ancillary activities such as music, dancing, grilling marshmallows, etc. Given that Ragnar doesn't have the resources to offer everything for their events, which combination of features would make people most compelled to want to continue to participate in Ragnar? In these instances, there's trade-offs that have to be made, and the consumer has to consider these things all simultaneously when making a purchase option. Perfect recipe for conjoint analysis. Before we dive into an example of conjoint analysis, let's talk a little bit how it's different from other marketing research techniques. Sometimes people ask, why don't we just ask people directly what attributes are most preferred or most important to them when we're designing new products? The answer to that is without asking consumers to make trade-offs between different features, they tend to say that everything is important to them and the data that we collect really loses meaning, gain meaningful insight. Consider this example. Imagine that we had four different attributes about our gym that we wanted to ask people how important they were to them. How important the quality of the classes are, the atmosphere of the gym itself, the quality of the cardio machines, and the price. And as you may have done in surveys in the past, perhaps we asked that on a 9 point, not at all important to extremely important scale, or 7 point or 5 point. Before we even collect the data, we should be able to anticipate what the results will probably look like. We can already anticipate they will characterize them as being extremely important. There's absolutely no trade-off in the way that we designed the survey, so people simply tend to say that these things are all extremely important to them. This doesn't really provide us with a lot of actionable insight as a marketer because we don't know where to focus our attentions with our limited resources to maximize people's preference or enjoyment of our gym. For more information visit www.FEMA.gov