So far, in this course, we have discussed a number of methods that you can use to help you start building your story and understand how and where you fit into the market. In this lesson, we're going to go over an example that pulls many of these concepts together and provides real world context. In this example, we're going to look at some real data from an actual survey that was run on consumers in the US pertaining to perceptions of different beer brands. Yes, beer. Now, as you can see in this table, the brands are listed in columns. We're looking at Amstel Light, Bass, Becks, Corona, Dos Equis, Heineken, Kirin, Molson, Moosehead, Sapporo and St. Pauli. What we're looking at here falls in line with the five to 10 competitive brands we would normally do for such analysis. For each of these beers, we have collected information on the attributes that consumers tend to value most. Examples include: a rich full-bodied taste, good taste, no aftertaste, refreshing, light, lower price, good value for money, whether a beer is prestigious or popular, whether it's considered masculine, whether it's from a country with a brewing tradition, whether it's for young people, whether it gives a good buzz, whether you drink it at picnics and outings or at the bar with friends or whether it's for home or after work or to serve guests at home. So you have these various attributes that have been found to be important for consumers and you have the set of brands. And now, for each of these, the idea is to collect information on what the consumer perceives this brand as being for each of the factors. So for example, the consumers could give a rating on full-bodied taste for Amstel Light and for Bass and for Becks and so on. Once you collect these ratings at the individual level, you need to aggregate them so that you can build one matrix with the brands as the columns and the various attributes as the rules. Usually, it ends up looking something like what you see here, where for each attribute, you have an average rating or mean, as it is also sometimes referred to, for the particular brand that you are reviewing. Once you have this information, you proceed to the next step which we discussed earlier, where you run your fact analysis and do your preference regression, where you regress the overall preference data we saw in the data set on newly created factors, so that you can identify the factors that are really important for the market that you're considering. While we do not cover aggression in this course, please see the quantitative research course in our specialization, where my colleague Olivia covers how you can conduct a regression. In this example, we looked at several attributes that consumers value with respect to beers. On the left hand side, you see what is known as a component matrix. In the component matrix, you see that there are four factors that have been identified by the scree plot as being important. Once we run the preference regression which is on the right hand side, you'll see that only the first two factors are actually beneficial to us describing how consumers view beers. And the way that we understand that only the first two factors are beneficial is because if you look at the last column in that table which is labeled Sig, any Sig value less than .05 is what you would consider an important factor. Conversely, any Sig value that is greater than .05 indicates that it isn't really an important factor. Now that we know that we don't need all four components, we can then go back to the rotated component matrix table and we can identify the attributes or variables that are probably most important for the first two components in the raw data component metrics. We do this by examining the magnitude of the value corresponding to the attribute in a particular factor column. You see that the first component has a rich full-bodied taste, good taste, masculine, is for drinking at a bar, drinking with friends or for home after work as its most important attributes. And component two has no aftertaste, prestigious or popular, for young people and you drink at picnics or outing as its most important attributes. So you could say that component two really represents as market for individuals who like beers and have a strong image and are easy to drink. On the other hand, in component one, you see that these individuals have a liking for more complex flavors of beers. They like it to taste really good. They like it to have a stronger flavor compared to those who are probably positioned around component two. So now that we have the information for component one and component two, and we have done a regression and can predict our factors, we then plot these two components along the two axes of a perceptual map, where axis one is component one, and axis two is component two. Now, lets talk about how to interpret this plot. So lets say we are working for Kirin beer. We see that Kirin is somewhere around zero for factor two and on the positive side for factor one. So Kirin's beer tends to be in a position more aligned with factor one than factor two. But more importantly, what we learn is that Heineken and Becks, two other big brands, are closer to Kirin than beer brands like the Dos Equis, St. Pauli, Bass or Corona and so on. And so this tells us that in the consumers mind, Kirin, Heineken and Becks are probably closer competitors than Kirin, St. Pauli and Molson. And this is why perceptual maps help. They help you not only identify how consumers view your brand, where in this case Kirin is stronger on factor one, which is the rich full-bodied taste factor as opposed two, which was the prestigious brand factor. It also tells you about which competitors in the market place are probably closer to you than others. Now you see here that there are lots of empty spaces all round. For example, I've circled one of them here. These empty spaces on a perceptual map are usually known as white spaces. And white spaces show you spaces of opportunity that potentially exist in the marketplace. So for example, if you were to introduce a new beer, theoretically, you might want to consider introducing it in any of these segments. However, it's important to note that not all spaces are equal. If there are large white spaces, it might be because there are no consumers there, or consumers do not appreciate any brand that might enter in such a space. So you need to pay careful attention to exactly what those white spaces imply. For example, if the white spaces imply the opposite of good taste, you don't want to position your brand in that white space. Now that we have seen this data for Kirin beer, and we looked at its perceptual map a little bit, and we worked on integrating it, it might be useful to try and understand what story you might be able to tell for Kirin beer. If we look at where Kirin beer is, you can say that Kirin beer is stronger in factor one and in factor two. So it more likely appeals to individuals who like rich full-bodied taste beers, who like to drink it at home after work and so on. And also, that Heineken and Becks are closer competitors to Kirin beer than other brands that exist in the marketplace. Hopefully, I have clearly illustrated how useful this type of tool can be in building your story and supporting it. By developing a perceptual map, you can clearly walk your audience through where your organization is and even perhaps, where you wanted to go in a clean and easy to understand visualization.
