LUCIDCONTRAST™
Best Suited For:
Testing messages, product attributes/benefits/features, and endpoints
Appropriate For:
All types of respondents in both qualitative and quantitative settings
What Sets LUCIDCONTRAST™ Apart
- Experience and Norms: Lucid Health has extensive experience and norms to contextualize scores.
- Greater differentiation: It is inherently comparative, unlike some other techniques.
- Flexible: It is highly adaptable and can accommodate many types of stimuli: text of various length, graphs, charts, and tables.
- Intuitive: The intent of the questions that drive this approach are clear to respondents.
- Simplicity: Scores are generated directly from respondent input, meaning no error is introduced by having to model or estimate scores, as with choice-modeling type techniques.
- Grounded in Solid Multidisciplinary Research: The technique underpinning LUCIDCONTRAST™ has been validated.
Contact us to learn more about LUCIDCONTRAST™
LUCIDDRIVER™
Best Suited For:
Understand the extent to which product attributes explain (or “drive”) a desired action, such as likelihood to select, satisfaction, etc.
Appropriate For:
All types of respondents in a quantitative settings
What Sets LUCIDDRIVER™ Apart
With driver analysis, one hopes to understand:
- How much likelihood to select (or any other desired action) is explained by all of the attributes included in the model (in statistical terms, R²)
- The extent to which each individual attribute contributes to that overall explanation (arguably, this understanding is more important to the marketer).
However, there are three key shortcomings of traditional driver analysis
- Inability to Estimate the Impact of Individual Attributes: Multicollinearity prevents the model from isolating the impact of one attribute from other correlated attributes.
- Ignoring Attribute Asymmetry: Traditional driver analysis tends to be one dimensional (linear), indicating that performance on a particular attribute is symmetrical; in other words, it inappropriately assumes that gains for performing well are equal to losses for performing poorly.
- Instability of Single-Product Driver Models: Researchers commonly conduct a driver analysis for only the product of interest. Unfortunately, single-product driver models are less stable than aggregate models and often generate questionable or nonsensical results, predominately caused by outliers. LUCIDDRIVER™ solves for these shortcomings. It is computationally superior to traditional driver analysis as it uncovers the impact an individual attribute has on the desired action.