We have completed six ACBC studies in five months, all for clients in the automotive industry. We feel ACBC is, today, the best methodology to handle complex product (or complex competitive environment) studies. Its adaptive scheme conforms well to each individual's purchase decision funnel, and increases the relevance of the information at the individual level. Treatment of price is very realistic. In our view, ACBC is today the most powerful conjoint method for pricing studies.
My clients like the dynamic structure of the Adaptive CBC survey, and the increased relevance of the questions and tasks. They understand the value of the adaptive approach of ACBC much better than the importance of orthogonality in traditional CBC. All else being equal, ACBC is an easy sell, as the client clearly sees how naturally it mimics the purchase decision-making process.
Even though the interview is longer than CBC, we have seen very high completion rates and minimal drop-out rates (in some cases less than 5%) from the start task (BYO) to the finish (calibration). We feel respondents are really engaged by the variety of tasks, perceive the narrowing focus through the survey, and 'want to see where all this will lead.' The need for cleanup of cheaters and speedsters has dropped dramatically compared to standard CBC studies, reducing the need for make-up oversampling and overall time needed for fielding/data cleaning (estimated down by 20%).
The information provided by BYO, Choice Winner, Must-haves and Unacceptables is of additional value to the client. Simulations based on HB analysis have reproduced market conditions very closely, even when working with subsamples of the order of 100-150 respondents. Given enough experience, we think we will be able to reduce sample size (currently we use a standard of 500 respondents for a 10-attribute study), passing through cost and time savings to the client.
Clustering on the price utilities from ACBC is, in our experience, highly resolved, with stable clusters. Clients have explicitly recognized in these cluster patterns their qualitative understanding of customer response and 'price barriers.' We feel this makes ACBC the strongest methodology available for quantitative price positioning and 'willingness-to-pay' studies.
The additional time required by the survey development is well worth the effort in terms of survey efficiency and quality. Sawtooth Software in our experience has an outstanding record in anticipating our needs for additional features. The software is reasonably – if not very – user-friendly.
Emanuele Leveroni Ph.D.
Ockham Associates, LLC