“Listening In” to Find and Explore New Combinations of Customer Needs

Ongoing dialogues between customers and web-based virtual advisors (e.g., Kelley Blue Book’s Auto Choice Advisor) are an untapped source of marketing research data for new product opportunities based on new combinations of customer needs. These data provide the scale necessary for complex products (148 trucks and 129 customer needs in our application). The tracking data are available at little incremental cost. The calibration data rely on AIO data (activities, interests, and opinions) that are collected routinely by many firms. Because we embed Bayesian methods, calibration evolves readily and incrementally for new trucks (many enter each year) and for new features (technology evolves rapidly in this industry). There are five components to “listening in.” (1) A Bayesian virtual advisor selects questions adaptively to provide recommendations. (2) An opportunity trigger (with Ρ-matrix flagging) tracks recommendations to identify opportunities based on new combinations of customer needs. (3) A virtual engineer explores the opportunity with directed quantitative and qualitative questions. (4) A design palette explores how customers themselves would design a truck to fulfill the desired combinations of needs. And, (5) a market-sizing mechanism (Ρ-matrix clustering) estimates the share of customers who desire the new combinations of needs. The five modules are flexible. They are designed to work well together and with existing, complementary methodologies. We examine the properties of “listening in” from three perspectives. (1) Formal analysis demonstrates that when recommendation probabilities drop (opportunity trigger), there are opportunities for trucks with new combinations of benefits (needs). The new trucks will fulfill customer needs better than existing trucks. (2) Monte Carlo simulation, calibrated on real data, demonstrates that the methodologies can recover “known” needs’ combinations (and needs’combination segments) from noisy data. The simulation also demonstrates how to calibrate the opportunity trigger. (3) A “proof-of-concept” application demonstrates that “listening in” is feasible in the pickup-truck category (over 1,000 web-based respondents). The application identified opportunities for new truck platforms worth approximately $2.4-3.2 billion and $1-2 billion, respectively. One of these opportunities has been confirmed with (more expensive) existing and on-going market research. The other is still being explored. We close by indicating how the methodology extends to other industries. “Listening In” to Find and Explore New Combinations of Customer Needs Identifying new platform opportunities is one of the most important roles of market intelligence. Monitoring [web-based advisors] provides a rich source of observed in-market customer behavior that complements our current inquiry tools which, by their nature, are forced to ask customers either to state their intentions before they are actually in the market or to remember after the purchase what they did (and why) when shopping for a vehicle. No form of inquiry is perfect, however, whatever its limitations, the currency [of web-based advisors] presents a valuable source of market understanding that is already streaming by and is of great value when used appropriately. – Vince Barabba, General Motors General Manager of Corporate Strategy and Knowledge Development (responsible for overseeing GM’s New Business Development Network) Unmet Combinations of Customer Needs Represent New Opportunities Automobiles and trucks are complex products. The investment for a new automotive platform might require as much as $1-to-2 billion and 1,200 person-years of investment. Such investments are justified by the scale of the market. For example, with approximately 150 trucks on the market, the average truck needs less than one percent of the marketplace to be profitable; each share point is worth $800 million in annual revenue. Most automotive platforms are redesigns to provide known customer benefits (“needs”). However, long-term survival requires that new opportunities be identified. For example, in the late 1980s, through a combination of qualitative focus groups and quantitative perceptual mapping studies, a new opportunity was identified for luxury vehicles that could haul moderate loads. Today this luxury SUV segment is one of the most profitable automotive segments. Another example came from leading-edge users. In the 1960s teenagers and young adults were customizing inexpensive vintage Fords with V8 engines – Ford recognized this opportunity for inexpensive sporty cars with large engines. The first production car in this “pony” segment, the 19641⁄2 Mustang, sold 420,000 units in the first year ($10 billion at today’s prices, classicponycars.com/history.html). The 1983 Chrysler minivans provide a third example. Growing families needed a vehicle that could carry a 4’x 8’ sheet of plywood, fit easily in customers’ garages, drive like a passenger car, have a side door for small children, and incorporate a sedan-like liftgate for shopping. Chrysler sold 210,000 units in the first year and dominated this new segment for years to come (allpar.com/model/m/history.html). These are but some of the many automotive examples where profitable new platforms filled previously unrecognized (by the auto industry) combinations of consumer needs. The firms that identified first these new combinations of