Loading...
  • Customer Experience
    Customer Experience
  • Employee Experience
    Employee Experience
  • Brand Experience
    Brand Experience
  • Product Experience
    Product Experience
  • Core XM
    Core XM
  • Design XM
    Design XM

Brand Drivers Analysis Widget (BX)

What's on This Page:


Was this helpful?


This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

The feedback you submit here is used only to help improve this page.

That’s great! Thank you for your feedback!

Thank you for your feedback!


About the Brand Drivers Analysis Widget

The brand drivers analysis widget measures how independent driving variables (drivers) affect key outcome metrics for a brand. The brand drivers analysis widget shows the importance of each independent variable based on the data collected so far.

This chart looks like a bar chart with brand imagery broken out along the left and different colored bars representing different brands

For example, brand drivers analysis can help answer questions such as, “Which of these factors have a bigger impact on the overall brand rating – ‘ease of use’, ‘value for money’ or ‘relevant content’?”

Data Requirements for Drivers

The drivers in this widget are your independent variables. These can be either brand imagery questions or product questions written in a scale format. Product questions can be about reputation, performance, reliability, and more.

Drivers generally come in two different data formats, depending on how your brand tracker survey was set up.

Driver Questions

In one format, the brands are the answers to the driver questions.

Example: When asking respondents which of these brands offers a good value for the money, they can select from a list of brands.
Question that says "Based on your experience, which of the folowing has a good offer for the money?" then a list of brands to choose from. multiple ca be chosen. There's a "none" option and a green widget implies these answers are carried from a previous question
When mapped to your dashboard, these fields should be the Multi-Answer Text Set field type.
Fields like "value for the money" mapped as multi answer text sets in a dashboard's data mapper

Mapped Field

In the second format, the survey’s been set up so that each individual survey response corresponds to the evaluation of only one brand. This means there will be a separate field that identifies which brand is being evaluated.

Example: One question, formatted as a scale, asking whether the brand gives a good value for the money or not.
A one-answer multiple choice question asking, "Based on your experience, how well would you say this brand could be described as offering a good value for the money?" Then the answers are in a scale format from very well to not well at all

If you’re using this format, there should only be one of each imagery question for all brands. Notice how in the example above, the brand name is piped in to allow the same question to be used for different brands.

You will also need to have some sort of binary embedded data field prepared before data collection to indicate that a brand either does meet the expectations of the outcome metric (1) or does not (0). This would mean using branch logic to specify which scale points correspond to which values. For more detailed steps, see the Customizing your Survey with Scales subsection.

Two branches. One designates an embedded data field named "good value for the money" should be set equal to 0, and the other sets it equal to 1

Qtip: Some fields configured for conversion funnel reporting, such as consideration, may already be formatted this way in your survey.

When mapped to your dashboard, the binary embedded data fields should be mapped as the Numeric Value or Number Set field type.

Imagery questions mapped in the dashboard

Data Requirements for Outcome Metrics

The outcome metric is the dependent variable in your analysis. Outcome metrics should represent some important measure for your brand. This can often include NPS questions, likelihood to recommend, or consideration questions.

An NPS question, rating liklihood to recommend from 0 to 10

The goal is to determine how the drivers affect this metric to give insight into possibilities for improvement.

Qtip: If you want to look at the impact of drivers on multiple outcome metrics, make a separate widget for each outcome metric.

If using Driver questions, outcome metrics need to be mapped to the dashboard as the Number Set field type, because there’s a discrete list of numeric values that can be selected in the question. You should have a separate field for each brand.

NPS quesions mapped in the dashboard

Qtip: Your data set may already have NPS fields mapped together under a field group – however, field groups are not compatible with this widget. You will need to make sure these fields are individually mapped.

If using Mapped fields, the outcome metric needs to be mapped to the dashboard as the Number Set or Numeric field type. You will have only one outcome metric field for all brands. You also need to map the field where you identify what brand’s being evaluated. This can be any field type, although we recommend Text Set.
one NPS field mapped to dashboard
Brand field mapped to dashboard

Calculation

Driver analysis is conducted using Relative Weights Analysis to establish the relationship between the dependent and independent variables. Relative Weights Analysis not only handles high multicollinearity among predictors, but also calculates importance ranking and overall impact for each attribute.

Learn more about the different types of calculations you can perform on this widget below.

Importance Score

Chart set to importance score. This chart looks like a bar chart with brand imagery broken out along the left and different colored bars representing different brands

Importance score represents the relative weight of a particular driver on the key outcome metric. This shows how important each driver is to the key metric in question in relation to one another. The value will be between 0 and 1. The closer to 1 (the higher the percentage), the more that driver affected the key metric in your study.

See more about Relative Importance.

Pearson Correlation

Chart set to pearson correlation. Data is visualized as bars coming from a center axis

The correlation score is the Pearson correlation coefficient between each driver and the key outcome metric. This is a value between -1 and +1, where -1 means there is a negative linear correlation and +1 means there is a positive linear correlation.

Current Score

Current core set. This chart looks like a bar chart with brand imagery broken out along the left and different colored bars representing different brands

The current score represents the current value of the driver over the key outcome metric’s data set. This is computed by performing an average calculation of the driver field over the responses that answered the key metric question. Since driver fields are binary (0 or 1), this value will be between 0 and 1.

Widget Setup

  1. Select a calculation type. To learn more about these three options, see Calculation.
    Calculation settings in widget editing pane
  2. Determine whether to get brand names from Driver questions or a Mapped field. See Data Requirements for Drivers for more information on the requirements for each.
  3. If you selected Mapped field, click Select field to specify the brand where you store the brand evaluated in each response (usually just “brand,” but naming might vary).
    Qtip: See Data Requirements for Outcome Metrics and look under “Mapped Fields” for more details.
  4. Click Select driver to add each driver.
    Adding drivers to a widget

    Qtip: Drivers are usually questions related to brand imagery and product-specific concerns like reputation, performance, reliability, and so on.
  5. Click Configure outcome metric.

Configuring the Outcome Metric for Driver Questions

Read these steps if you indicated that you would “Get band names from” “Driver questions.”

In new menu that opens, Configuring the Outcome Metric for Driver Questions

  1. Label the outcome metric.
  2. The “field” column is where you choose the field you created for the outcome metric.
  3. The series is where you specify the brand that field measures.
  4. Click the plus sign ( + ) to map another field. You should have a separate field for each brand.
  5. If you would like to perform category analysis, select Evaluate the brands from the above list as a representation of the entire Category. This means that the brands will be aggregated together and a single calculation will be performed. The results will tell you how the industry in general scores on the chosen calculation type. This should only be done if the brands chosen represent a meaningful portion (> 80%) of the industry being studied. Otherwise the data will likely not be useful.
  6. Click Save.
Qtip: Look carefully at the screenshot. Notice how for each field added (e.g., “MusiQ NPS”) the corresponding series (“MusiQ”) has the same brand name listed.

Configuring the Outcome Metric for a Mapped Field

Read these steps if you indicated that you would “Get band names from” a “Mapped field.”

In new menu that opens, Configuring the Outcome Metric for a Mapped Field

  1. Label the outcome metric.
  2. The “field” column is where you choose the field you created for the outcome metrics.
  3. Select the brands you want to analyze in the widget.
  4. If you would like to perform category analysis, select Evaluate the brands from the above list as a representation of the entire Category. This means that the brands will be aggregated together and a single calculation will be performed. The results will tell you how the industry in general scores on the chosen calculation type. This should only be done if the brands chosen represent a meaningful portion (> 80%) of the industry being studied. Otherwise the data will likely not be useful.
  5. Click Save.
Qtip: Your widget will tell you that your results may be insignificant if your coefficient of determination (R2) is under 0.2 and the sample size is under 300 responses. Keep this in mind before making decisions based off the widget’s results. This warning applies regardless of the way you configured your widget (calculation chosen, drivers format, etc.).
Warning window opens to say your results may be insignificant if your coefficient of determination (R2) is under 0.2 and the sample size is under 300 responses.

Widget Customization

Changing Brand Colors and Names

You can adjust the color and name associated with each brand under the Series list.

As described, the colorful "series" settings in the widget editing pane