The Survey Monsters
Transform product development with inferential survey techniques
Let’s set the stage
Before we dive into more complex measurement systems, let's start with some foundational inferential survey techniques that can transform how we approach product development.
Today, we’ll explore MaxDiff, TURF, and Conjoint analysis —techniques that have proven invaluable in prioritizing customer problems, maximizing reach, and building the right products.
Understanding the fundamentals is crucial as we build towards the more intricate methods that yield deeper insights and greater impact.
Monsters at the UX Virtual Summit
I had just joined a new company, and within my first few weeks, they asked me to present at their UX Virtual Summit happening the following month. The excitement of sharing "Weird survey methods & you" was mixed with a good dose of nerves. The org wasn’t familiar with these unconventional methods, making this my first opportunity to introduce them to MaxDiff, TURF, and Conjoint.
As the presentation day approached, the mix of excitement and nerves reminded me of childhood fears—monsters under the bed. This sparked my idea for the Survey Monsters.
Prioritize with MaxDiff and Marvin the Most
A team can only solve four (of twelve) customer problems. How do they choose?
Marvin recommends a MaxDiff so you’ll know which problems are most important to customers and by how much more.
Outshining traditional methods
Rating, ranking, and allocation tasks can fall short when there are differences in scale interpretation and the item list is longer.
The MaxDiff provides clear, powerful data.
Power up your signals
The data the MaxDiff yields, scored on a 0 to 100 scale, offers insights into the relative importance (aka ratio-scaling) of each feature considered. Ratio-scaling is essential for making meaningful comparisons.
By starting from a true zero point, it allows you to accurately say one feature is twice as important as another.
These comparisons provide a solid foundation for interpretability by technical and non-technical audiences alike.
MaxDiff’s typical applications include:
Let’s recap,
Maximize with TURF and Tina Total-Reach
TURF identifies features that cater to the most important problems.
AND
It reaches the maximum distinct customers.
It shows how each feature contributes to overall market coverage.
By leveraging MaxDiff preference scores, TURF calculates the incremental reach of tested features and identifies the best combinations for maximum market impact.
TURF’s typical applications include:
Let’s recap,
Build with Conjoint and Curtis Combo
Conjoint is essential for testing how to build features,
before committing them to code.
The technique helps define minimally viable and maximally delightful, modeling implications for willingness to pay and market share.
Conjoint helps us understand the value placed on each feature, and model market reactions to feature trade-offs, ensuring strategic planning and competitive analysis are spot on.
It removes the guesswork from product development, analyzing thousands of potential configurations to ensure the final product fully meets customer needs.
Calculating importance
Calculating category & level importance shows the relative importance of each attribute and its impact on the total utility.
This approach is incredibly useful because it allows you to break down a product into its composite parts, identify the key drivers of purchase choice, and create ideal product configurations.
Conjoint’s typical applications include:
Let’s recap,
Embrace monster wisdom
Ask yourself what if before we went to market we could,
By incorporating inferential survey methods like MaxDiff, TURF, and Conjoint analysis, we can. Three and a half years later, these techniques are now a staple of how my prior employer builds products and roadmaps, offering a strategic edge in UX and product development.
Stay tuned for upcoming posts where I dive deeper into each of these methods, sharing tips and tricks I've picked up along the way, case studies that highlight their practical applications, and example visualizations to illustrate their effectiveness.
‘Til next time, I’m Bianca
You can find the BriteNote here
