Editor’s note: Juan A. Tello is senior vice president, Americas, at marketing research firm SKIM, Costa Rica.
Most brands rely on a brand tracker to monitor brand health and competitors over time. While these trackers can lead to insights about how trends affect your brand and your competitors, they usually explain very little about causality: how and why a brand shifts positions over time. To better understand brand health, marketing researchers can turn to brand driver analysis to extract more strategic insight from tracking data that your company has already collected. This article will cover a framework marketing researchers can apply to better understand why and how a brand gains or loses ground in the marketplace; define what should be strategic points of difference and points of parity for your brand; and identify potential white-space brand identity opportunities in the market.
Structuring the framework
The brand driver framework includes six main steps:
- Define the key brand equity statements or attributes.
- Define an overall brand health metric correlated with sales, which will be explained by the rest of the brand equity attributes.
- Quantify relationships between attributes and the key brand health metric.
- Quantify brand equity ownerships vis-a-vis competition.
- Build visual ladders combining the information above.
- Simulate potential gains in brand positioning as a function of improving brand performance on key equity attributes.
The first step in this framework is to identify the brand equity attributes that collectively describe the dynamics of the category and the positioning of all brands in the market. To capture a broad spectrum of the market, a set of attributes that may number as high as 80 are tested in a brand tracker (more would be a burden for respondents). Ordinarily, many of these attributes are highly correlated and can be clustered into 15 to 25 consumer underlying dimensions, which can then be assembled into a hierarchical model. Constructs at the bottom level tend to include more functional performance-related variables such as reliability, portability, product quality, ease of use and fair price. These subsequently drive other constructs found at a higher level which are more emotional and/or holistic in nature, such as customer service, product performance and value for money.
The initial list should include functional and emotional items since both are involved in decision-making and the clustering process and may reveal interconnections between them. It is important to list attributes that are also relevant to your brand’s competitive set. This method can produce an accurate picture of an entire market as long as the appropriate data representing all market players are included. For example, a company may decide not to use its existing brand tracker data because it does not fully capture novel and disruptive benefits of newcomer online competitors. Instead, it may field a new study with an enhanced and broader list of attributes covering different market spaces.
This modeling requires two types of metrics: an overall brand health metric and an association metric between brands and brand equity attributes. The first one describes positioning in the marketplace. An important step of the analysis is to define which variable to use as a proxy for brand health for a given industry. There are many options – purchase intent, overall performance rating, likeability, uniqueness, loyalty or likelihood to recommend – and the right choice is often not obvious. I recommend conducting an external and internal validation of these metrics. External validation helps quantify how strongly these metrics are correlated with the business, as measured by sales or market share. This requires numerous waves of brand tracking studies and the corresponding sales data over time. It also should align with top management strategy on what the company wants to drive and communicate to the market. Internal validation looks at the variance and distribution of the metrics to identify which provide a better spread and differentiation between brands. By looking at both assessments, it is usually possible to select a unique metric that will sit atop the model, while all other variables will converge toward it. An alternative preferred by some is to not choose one but rather have a meta-metric that combines all related brand health metrics into one. While this is possible, it becomes more difficult to clearly identify and communicate what is ultimately being driven as a dependent variable.
The second is a set of associations between brands and the attributes on the list described above. This takes the form of a multi-punch matrix in which binary variables describe whether a given brand is associated with a given attribute. The next step is to use multivariate statistical methods such as Bayesian belief networks, structural equation modeling and partial least squares to uncover relationships and interactions between factors. While I recommend the latter method, the results from these are often similar so the choice depends on the objective and the type of data available.
Finding white-space opportunities
To make this model actionable, an additional dimension of analysis is required: brand ownership. Brand ownership quantifies the strength of an association between a brand and an attribute relative to the brand’s overall positioning in the marketplace. For example, the dominant brand in a market might be mostly associated with product performance, while a brand with a smaller market share might be able to capitalize on a particularly strong association with e-commerce. To eliminate the effect of the big brand we analyze normalized values, allowing us to compare all brands on a level playing field.
Brand ownership analysis can also reveal white-space opportunities, attributes that are not associated with any particular brand. It is much easier for a brand to move into these spaces than to shift consumer perception of a factor that is already owned by a competitor. In this example market, customer service might not be strongly associated with any brand, representing a space that a brand could claim.
By overlaying brand drivers and attributes owned by brands, marketers can easily identify relevant strengths of their brands, threats from competitors and opportunities to expand their brands’ identities. Brands can also define strategic points of parity and difference vs. competition.
Once the model is complete, it can be used in multiple ways. I typically look at differences by segment to determine whether differentiated strategies might be appropriate for different groups. One obvious choice is to split by age cohort or any other demographic variable. Heavy users of a product might be compared to light users. Consumers can be divided by frequency of shopping trips or preferred shopping destinations.
These segments can then be compared on the basis of other variables. For example, segments can be mapped across a malleability index, a measure of how difficult it is to change brand perceptions within each segment. Brands can then identify opportunities for growth if a group is found to lack a strong attachment to any particular brand within a category, or if another group of consumers has not fully formed a set of associations with your brand.
Three deliverables
Marketing researchers who turn to this method should look at concluding with three deliverables. The first deliverable is a brand ladder. It shows a full profile, with both strengths and weaknesses, for one brand at a time. This ladder visualizes how the brand is perceived in the minds of consumers. The brand ladder can be used to see which important attributes are owned by the brand, as well as the attributes in which the brand shows a weak association. At a glance, the brand ladder can inform us about the attributes that are driving the brand’s performance or the lack of it.
The second is the market opportunity ladder. This ladder visualizes available opportunities in the market. It combines all brands in a market but it only displays ownerships. Weaknesses are not incorporated but white-space opportunities are visible. It can be used to identify which attributes are solely owned, co-owned and not owned in the overall market and how impactful these are toward brand health. It helps to guide brand strategy toward points of parity and points of difference for the brand to own. At a glance the market opportunity ladder can be used to conduct an opportunities and threats analysis for the brand.
Finally, a contribution bar chart offers a traditional ranking of brand drivers in terms of how impactful they are on brand health.
All of the visualizations discussed up to this point are descriptive but this method also has predictive ability through market simulations. If the analysis identifies customer service as an unclaimed white space, a scenario can be simulated to assess the gains in overall brand health of moving into that space vs. the risks of a competitor claiming it first.