New Product Development
Choice Modelling & Conjoint Analysis
Choice-based conjoint analysis helps businesses identify the combination of product or service features that customers value most. It works by asking people to choose between realistic alternatives, allowing the model to estimate how much value they place on each feature, service element, and price point. This makes it possible to design offerings that maximise customer utility while also improving commercial performance, pricing decisions, and likely market uptake. Choice-based conjoint can be widely used for product design, pricing, and market-share simulation because it reflects how people make trade-offs in real purchase situations.
Our Methodology
Choice-based conjoint analysis can be applied in two practical ways, depending on the business model, data availability, and research objective.
The first approach uses primary data collection through experimental design. In this method, we design structured choice experiments in which participants are presented with different combinations of product or service features, prices, and benefits, and are asked to select their preferred option. By observing how respondents make trade-offs across multiple scenarios, the model estimate the utility they assign to each attribute and identify the combinations that generate the greatest customer value. This approach is especially useful when testing new concepts, designing future offers, or evaluating products and services before launch.
The second approach uses observed customer behaviour from digital journeys, such as the choices customers make when visiting a website, selecting certain options, and not choosing others. Rather than asking customers directly in a survey, this method analyses revealed preferences from real interactions, including clicks, views, purchases, basket additions, and product comparisons. By examining these behavioural patterns, we infer which features, bundles, and price points are most attractive in real market conditions. This approach is particularly valuable when businesses already have rich customer interaction data and want to optimise existing offers based on actual behaviour.
Both methods help identify the most effective combination of product and service features, but they serve different purposes. Experimental choice design is ideal for testing possibilities before market launch, while behavioural choice analysis is powerful for refining live offers using real-world customer decisions.
Case Studies / Applications
Primary data collection through experimental design
Healthcare: A healthcare provider developing a new service package may want to understand how patients value appointment speed, digital access, specialist availability, aftercare support, and pricing. Using a choice-based conjoint survey, participants are shown different package combinations and asked which they would choose. The results reveal which elements matter most, which trade-offs patients are willing to make, and which package design is most likely to maximise patient utility and commercial viability. The same approach can be applied in insurance, hospitality, technology, and entertainment when organisations need to test new product or service concepts before launch.
Insurance: Choice modelling is highly effective when insurers need to package multiple benefits and cost elements into attractive plans. In a Medicare case study, conjoint analysis can be used to estimate the value customers placed on benefits, out-of-pocket costs, network access, and supplemental services, then simulate which plan structures would generate the strongest share of preference.
Hospitality: Hospitality businesses can use conjoint analysis to determine which combination of room, service, amenity, and pricing features creates the strongest guest appeal. A classic example is a hotel, where conjoint analysis can be used to help select target segments, shape service positioning, and guide facility design decisions.
Technology and Entertainment: The same approach is highly valuable for SaaS, digital platforms, and entertainment services where customers evaluate bundles rather than single features. It can be used to decide which features belong in Basic, Pro, and Enterprise plans, which premium add-ons deserve separate pricing, and how to balance content access, convenience, service experience, and price in subscription packages. Real-world pricing projects in software have used conjoint trade-off scenarios to identify preferred feature bundles and support role-based pricing decisions.
Secondary data collection Observed website choice behaviour
An insurance, technology, or hospitality business may already have website data showing which plans, features, or bundles customers explore, compare, and ultimately select, while also capturing which alternatives they do not choose. By analysing these real choice patterns, it is possible to estimate the relative value customers place on different attributes and identify the combinations most likely to drive engagement and conversion. This method is especially useful for optimising live products, service bundles, and pricing structures based on actual customer behaviour rather than stated survey responses.
Combined application for stronger decisions
In many cases, the strongest solution comes from combining both approaches. Experimental conjoint analysis can be used to test new ideas or future package designs, while website behaviour data can validate and refine those findings in real market conditions. Together, they provide a more complete understanding of customer utility, helping businesses design offers that are both strategically sound and commercially effective.