The Brand Personality of 45 Major U.S. Brands (Part III)

Background

It is essential for brands to find unique ways to gain a competitive advantage by differentiating themselves from their competitors. The Zion & Zion market research team is engaged in an exclusive in-depth series of studies to understand how consumers perceive the personalities of top U.S. brands. This article is the third in our new series on 45 major brands in 15 industries (see tables 1 and 2). In this series, we analyze how top brands are perceived across three key brand personality scales: Aaker’s (1997) classic five dimensions of brand personality, Freling, Crosno, and Henard’s (2011) brand personality appeal dimensions, and Haji’s (2014) negative brand personality dimensions. In the first two articles of our series (see here), we investigated how the brands score on each of the three scales. In this article, we explore how the brand personalities form groups, or clusters, of brands. In future articles, we will highlight the impact various brand personality dimensions have on brand performance, perform deep-dives into the 15 industries and more.

Table 1. Industries Included in this Study

Table 2. Brands Included in this Study

In this article, we perform k-Means cluster analysis on all 45 brands in the sample based on their average personality scores. This technique allows us to better understand how brands group naturally within and across industries. The results of this study show that the leading brands in our study form four distinct clusters. One of these clusters scored high on all five classic brand personality dimensions, high on clarity and originality and low on negative personality traits. Seven of the 11 brands in this cluster hail from the technology industry. Two brands form their own cluster at the opposite end of the scale, Wells Fargo and Facebook, both beset by recent scandals tarnishing their brands.

Executive Summary

The Zion & Zion research team sought to gain a deep understanding of the brand personalities of 45 top brands in the U.S. The brands were selected based on established brand value and industry leadership rankings. We surveyed 9,309 adults, 18 years of age and up. Each respondent was asked about one of the brands. Respondents in our survey were also asked to rate their familiarity with each brand on a scale of 1 to 7, with 1 being “very unfamiliar” and 7 being “very familiar.” Only respondents who answered with at least somewhat familiar (5 or higher) were retained for the study which resulted in a data set of 6,444 responses.

Cluster Analysis of Brands

We used 11 dimensions to evaluate whether the 45 brands formed groups that are similar to each other within the group as a whole, but distinct from other sets of brands (see table 3 for the list of dimensions, and see our previous articles in this series for a description of the details of these dimensions). This approach creates an 11-dimensional analysis space, which cannot be presented on a two-dimensional screen in its entirety. To allow visualizing the k-Means clustering technique, and give a sense of the clustering results, the scatterplot in Graph 1 presents the results for just three of the dimensions.

Table 3. Brand Personality Dimensions

Graph 1. 3-Dimensional Example for Visualizing Brand Personality Clusters

Throughout this research, the number of clusters was optimized based on the results of two techniques – the Akaike Information Criterion (AIC), which is a method of calculating how well a model fits the data, and the Elbow Technique, which allows the researcher to evaluate improvements in fit across a range of potential cluster numbers. For our set of 45 brands and 11 dimensions, these two techniques suggested that the brands form 4 distinct clusters (see table 4 for the average scores in each cluster and table 5 for the distribution of brands to the clusters). Graph 1 gives a sense of what the full 11 dimensional space conveys, a visual separation of the brands into distinct groups and areas. This visual separation is a reflection of the perceptual differences between the 4 groups of brands.

Table 4. Cluster Averages

Table 5. Cluster Membership for each brand

The 4 Clusters of Brands

Taken together, tables 4 and 5 paint a clear picture of 4 distinct groups of brands. In group A, we find brands that have the lowest scores for Negative Brand Personality dimensions. These brands are seen in a positive light, but they are simultaneously seen as 2^nd in terms of the classic brand personality dimensions and also 2^nd in terms of clarity and originality. Group A contains brands like Southwest Airlines, Hilton, and Geico. Group B meanwhile has the highest scores on the classic brand personality dimensions and on clarity and originality, but they come in tied for 2^nd when it comes to not being seen as having negative traits. Group B includes brands such as Disney, Nike, and all Streaming Services. Group C ties Group B for 2^nd on negative traits but is only in 3^rd place for the classic dimensions and clarity and originality. Group C contains brands such as American Airlines, Walmart and Verizon.

Cluster D is made up of only two brands, Facebook and Wells Fargo. This group possessed the highest scores on all Negative Brand Personality dimensions, and the lowest scores on the “Big 5” and the Brand Personality Appeal dimensions. It is interesting to contemplate the recent unethical practices of both brands and their potential contribution to the lingering negativity that shaped the results of the scores by these consumers. For example, it was revealed that Wells Fargo “agreed to pay $3 billion to resolve potential criminal and civil liability stemming from [practices of opening] millions of accounts…to customers under false pretenses or without consent…” In a similarly damaging episode, the FTC fined Facebook $5 billion dollars for user privacy violations in 2019. It is not unreasonable to speculate that these actions will have a lasting impact on how consumers perceive these brands well into the future.

Interpreting the Data

When interpreting the data shown on the charts, it is important to keep in mind that each brand has its own unique profile that needs to be understood in relation to the brand’s positioning and target segment. It would be misleading to see this as a competition where a brand is strong only if it scores high on all appeal facets and low on all negative facets. However, in contrast to the Aaker classic brand personality scale, it is generally true that brands would prefer to be seen as clear and original and would not like to be associated with negative traits. Brands can perform well if they have middling scores on all of these dimensions, so this is not a hard and fast judgement, but brands would generally not like to be seen as scoring both, low on appeal, and high on negative facets. The caveat is that being perceived negatively by some consumers may be acceptable if the brand’s target segment sees the brand in a positive light.

Putting the Data to Work

This article is the third article in our multi-part series where we will look at each of the brands in 15 industries in more detail. Later in this series, we will look at how each brand’s profile matches up with its competitors, what facets are key for market success in each industry, and we will also evaluate a series of industry-specific questions.

The Zion & Zion research team encourages marketers to look at how they want their brands to be viewed by consumers and look for ways to distinguish themselves from the pack. In our series we will show that by picking and choosing specific areas to improve, brands can forge a distinct and notable identity. Our series can thereby show how such differentiation can help brands to gain a competitive advantage.