Identifying the Key Drivers of Brand Image
By Rajan Sambandam, PhD, Chief Research Officer
Background and objectives
As part of its brand management strategy, a major energy utility (ABC Company) wanted to measure its brand image and identify specific image attributes that had the greatest impact on overall impressions of the company. In this way it could gauge its current performance in an increasingly competitive market, and prioritize image improvement efforts based on what was most important to consumers.
Notes on methodology
We were commissioned to design and conduct research to serve the above-stated objectives. Telephone research was conducted among 1,207 household decision-makers within the company’s service area. Each respondent was asked to rate the overall image and overall quality of ABC Company, and to evaluate the company on a series of imagerelated statements.
In total, interviews averaged approximately 15-minutes in length. Overall image and overall quality were measured using a five-point scale: excellent, very good, good, fair, or poor. Reactions to image-related statements were captured using a five-point agreement scale: strongly agree, somewhat agree, neither agree nor disagree, somewhat disagree, or strongly disagree.
Two models of brand image were produced: one using traditional multiple regression; one using Satiscan™, TRC's proprietary means of key driver analysis. The bulk of this case study focuses on a comparison of these two models. But first, we present for the reader a brief description of Satiscan™.
A brief description of Satiscan™
Satiscan™ is an analytic method developed by TRC specifically to address questions such as those posed by ABC Company. Using a directed search algorithm, Satiscan™ produces a map of brand image. This map can be used to identify those aspects of service that have a direct impact on image, those that have an indirect impact on image, and those that have no noteworthy impact on image.
This is a significant development in key driver analysis. Traditional methods (typically regression analyses) can only identify direct impacts on satisfaction, and ignore the fact that aspects of brand image inevitably interrelate to each other, and hence have indirect as well as direct effects on a customer’s overall perceptions of a company.
Satiscan™ reviews all possible path models in relation to customer responses and only then determines the ideal model of brand image. This may not seem like a tremendous difference until one realizes there are literally thousands of possible path models to choose from, and the odds of choosing the optimal one in advance are quite slim, unless there is a precise theory to guide model building.
Comparing a Satiscan™ model to a traditional regression model
The difference between Satiscan™ and traditional regression analysis is clearly shown by comparing two maps developed for ABC Company. Figure 1 is the view of brand image put forth by traditional regression. BREAKING POINT As the arrows indicate, this model suggests that seven specific aspects of service significantly influenced ABC’s image directly, and that none of these seven related to one another. The numbers above each arrow denote the relative importance of each aspect. To simplify this discussion, it is safe to assume that the larger this number, the greater that service aspect's impact on brand image.
When a Satiscan™ map was produced, however, a different and far more complex picture of brand image emerged. Looking at it, (Figure 2) you can see that some attributes had a direct impact on image, others had an indirect impact via other aspects, and still others had both direct and indirect impacts.
An apples-to-apples comparison
This comparison can be boiled down to an apples-to-apples comparison. As with traditional regression, the total impact of each service aspect on overall image can be calculated with Satiscan™, and the resulting weights are comparable
|Committed to community||0.31||--|
|Charges reasonable rates||0.22||0.1|
|Earned my loyalty||0.13||0.12|
|Respected in the community||0.07||0.07|
|Values you as a customer||0.07||0.06|
|Useful conservation information||0.07||--|
|Goes the extra mile||0.06||--|
|Believable source of information||0.03||0.05|
|Reliable electric power||0.03||--|
|Concerned about environment||0.01||0.05|
to those derived from regression. To further emphasize the differences between the two methods, both sets of impact weights are listed in the Table 1.
In some ways, the Satiscan™ and regression models produced similar results. Most notably, both methods identified overall quality as a critical determinant of brand image, and in fact each assigned this variable a similar importance weight. The ability of Satiscan™ to model indirect relationships between variables, however, produced a richer and more actionable model for ABC Company. Consider the following:
Committed to the community had no significant influence over image in the regression model, because it's impact was entirely indirect. Satiscan™ was able to show how community commitment impacted perceptions of rates, respect, and overall quality – variables that in turn had direct influence over ABC's brand image.
By accounting for direct and indirect paths of impact, SatiscanÔ showed the attribute charges reasonable rates to be more than twice as important than suggested in the regression model.
Both SatiscanÔ and regression assign similar importance to the attribute earned my loyalty. Only Satiscan™, however, provided ABC Company with the secondtier information it needed to take meaningful action on this component of brand image. Loyalty, it suggested, could be gained through honest and open communication, demonstrated extra effort, reliable service and demonstrated concern for the environment.
The regression model provided seven independent drivers of brand image. Satiscan™ yielded 12 key drivers, and helped explain how these independent variables worked in concert to shape ABC's brand image.
Through its ability to map both direct and indirect impacts on brand image, Satiscan™ gave ABC Company a more complete and accurate picture of how it could improve its image among consumers in its service area. More importantly, it enabled ABC Company to make improvement recommendations that were datadriven, actionable, and intuitive.