Advanced Data Analysis Techniques & How Your Company Can Benefit

Posted by Op4G Staff on March 30, 2021

When it comes to market research, Op4G does it all. Our services run the gamut from survey consultation and programming, to sample recruitment on our proprietary panel, to survey analysis and reporting. Industry experts are taking notice. In fact, this month, Quirk’s Marketing Research Review named Op4G a Top Full Service Research Company!

In keeping with this latest recognition, we’d like to highlight some of the data analysis techniques we utilize and how other companies can benefit. Specifically, we’re delving into 3 advanced data analysis techniques in our arsenal: Conjoint, MaxDiff, and TURF Analysis. These techniques go beyond the typical tabulation of survey responses. They apply statistical algorithms to the data, gleaning rich market insights for our clients. Here’s how—and when—these techniques work:


  • Conjoint Analysis: Let’s say Apple is developing yet another new laptop. Naturally, the company needs to know what attributes matter most to consumers: the screen size, the memory, the processor speed, the operating system, the price, etc. But asking consumers directly can yield inaccurate results due to various biases and “lack of discrimination." Instead, Apple could turn to conjoint analysis. 

    With conjoint analysis, the researcher presents a set of products - each having a specific price and different levels of attributes. Ideally, the number of attributes is fairly low (~5). The panelists must then rank, rate, or choose from the presented products, making trade-offs between the underlying attributes in the process.

    ConjointQuestionExample_Op4GThis type of analysis reveals the panelists’ “implicit valuation" of these attributes. The company can then incorporate the most valued attributes into the product design. In addition to new product design, common applications of conjoint can include product line extensions, measuring price sensitivity, and branding/packaging. 

  • Maximum Difference (MaxDiff) Analysis: Now imagine that hospitality company, Marriott International, is updating some of their resorts and wants to know which resort attributes/features are preferred by their customers. They have an idea of what attributes or features they want to add, but need to narrow down the top choices. 

    With MaxDiff, the researcher presents survey respondents with 30 different resort attributes (e.g., a hotel gym, all-inclusive package, swimming pool). The respondents must then rate each attribute from least to most important. This process repeats a number of times with the list of attributes varied so that the respondent selects the best/worst features from a number of product characteristic subsets. At that point, a MaxDiff analysis of panelists’ choices can accurately determine how consumers would rank all 30 attributes. 

    The goal of this type of analysis is to rank attributes in terms of their importance to customers on a common scale, so that comparisons and trade-offs between them can be made. The main difference between this type of analysis and Conjoint is that MaxDiff is best used for single-level preferences versus multi-level (e.g., 'reasonable prices' vs. listing specific price points). Common applications of MaxDiff include message testing, product feature studies, or assessing benefits vs. side effects in health care products.  


  • Total Unduplicated Reach and Frequency (TURF) Analysis: Pretend that you own a successful fleet of ice cream trucks (everyone’s fantasy, right?). You want to break into a new market but your truck can only accommodate 2 out of 6 possible flavors. Which do you pick to maximize the odds that a consumer will find a flavor worth buying? TURF analysis to the rescue!

With TURF Analysis, the researcher asks panelists in the target market to rate or score the 6 flavors, thus revealing their first and second favorites. Then, for each of the 12 possible flavor combinations, the researcher can calculate the percentage of panelists that would find their first or second favorite flavor in that combination. This percentage is also known as the “reach."

TURFQuestionExample_Op4GThe next step is to calculate, for each flavor combination, the average number of favorite flavors per respondent. For example, a very popular combination like chocolate and vanilla may include the two favorite flavors for most panelists, resulting in an average close to 2. This number is known as the “frequency." By factoring in both the reach and frequency, TURF analysis can identify the optimal product combination for that market.


These examples merely hint at the power of advanced data analysis. Conjoint analysis, for example, could also assess the appeal of advertisements or service design. MaxDiff could gauge customer satisfactions and brand preferences. And TURF analysis could provide estimates of media potential, helping to shape communications strategies.

Thus, virtually any type of market research (B2B, Consumer, Healthcare, etc.) could benefit from advanced data analysis! To learn more about these techniques, or Op4G’s full service solutions, please contact or download our client one-pager below. 


Download Our Client Information Sheet


Topics: Consumer Research, Growth, Research Methodologies