Nº 0 · Advanced Analytics
Conjoint Analysis.
Predict real-world choices by understanding the trade-offs customers make.
Nº 1 · Benefits
Choices, not opinions.
Reveals real trade-offs
People can’t accurately describe their own decision rules. Conjoint forces them to choose, and infers the rules from what they pick.
Models the whole market
Predict share, price elasticity and cannibalisation across the full competitive set under any scenario you can specify.
Optimises the package
Find the feature, price and positioning combination that maximises share, revenue or margin for any target audience.
Nº 2 · The method
What is conjoint analysis?
What do your customers actually value? Conjoint analysis tells you.
Respondents are placed in realistic choice scenarios and asked to select between options that vary across attributes such as price, features, and brand. From their repeated choices, statistical modelling extracts the underlying value they place on each element, producing a quantified map of customer preference grounded in behaviour rather than opinion.
People don’t make decisions in a vacuum. They weigh price against features, brand against convenience. Those trade-offs reveal what they genuinely value when it matters most. Conjoint analysis captures exactly that.
The result is the ability to predict share, optimise profit, and anticipate competitive response before committing to a direction.
When to use it
When a decision depends on understanding how customers weigh competing priorities, conjoint is the tool you need.
01
New product development and feature prioritisation
02
Pricing strategy optimisation and willingness-to-pay analysis
03
Product portfolio optimisation
04
Market share prediction and forecasting
05
Package and bundle configuration
06
Product positioning and differentiation strategy
07
Range rationalisation decisions
08
Competitive response modelling
09
Value proposition development
10
Feature value quantification for business case development
Nº 3 · Decisions
Decisions it supports.
Conjoint hands back a simulator that lives beyond the study, letting you re-test decisions for years as market conditions change.
01
Proposition
- Conjoint tells you which features drive uptake, which configuration maximises profit, and what the optimal price point looks like.
- Analysis provides a clear basis for positioning against competitors without leaving share on the table.
02
Brand
- Understand the monetary value your brand carries and the price premium you can credibly command so that you can grow with confidence rather than guesswork.
03
Customer
- Breaking down preferences by segment reveals which groups value which features.
- That means you can customise offerings to serve different target audiences
04
Communications
- Make communication more effective by identifying which benefit claims drive purchase intent.
Nº 4 · How it works
Five steps from design to a working simulator.
Conjoint is design-led. We over-invest in the front half so the back half is decisive.
STEP · 01
Define attributes
Identify the product features, service elements, price points and brand variables to test. Typically, 4–8 attributes with 2–5 levels each, delivering actionable results without overloading respondents.
STEP · 02
Design choice tasks
Respondents are shown realistic product profiles, usually three to four options, and asked to select their preference, mirroring how decisions are made in practice.
STEP · 03
Collect trade-off data
Respondents complete 12–20 choice tasks. Each one requires a trade-off. The accumulated data captures how different attribute combinations are valued across the sample.
STEP · 04
Model preferences
The modelling tells you exactly what each feature is worth to your customers and what they’re willing to pay for it.
STEP · 05
Simulate scenarios
Interactive market simulators predict share, revenue, and profit under different product configurations, pricing strategies, and competitive scenarios, giving you a reliable basis for decisions.
Nº 5 · Getting started
What you need to begin
- Product or service features and attribute levels to test
- Price points or price ranges for testing
- Segmentation variables for subgroup analysis
- Current market share data (if available for calibration)
Nº 6 · FAQ
Frequently asked questions.
Fixed choice-based conjoint presents all respondents with the same set of pre-determined choice tasks. Adaptive CBC adjusts questions in real time based on earlier answers, homing in on individual preferences more efficiently. Adaptive CBC typically requires fewer tasks (8–12 versus 12–20) and produces more precise individual-level estimates. Adaptive requires more sophisticated programming. Fixed is simpler to implement and remains the standard approach for most studies.
The practical range is 4–8 attributes, each with 2–5 levels. Fewer attributes may not justify the research investment. Any more than 8 risks overloading respondents and reduces data quality. For complex products, hierarchical approaches can first identify which feature categories matter most, then drill into specifics within those categories.
When properly designed and fielded, conjoint predictions typically achieve 90% accuracy in predicting actual market share and choice behaviour. Accuracy depends on realistic task design, appropriate attribute selection, a representative sample, and rigorous model validation. We validate using hold-out tasks and, where possible, real purchase data.
MaxDiff measures the importance of individual features in isolation, which is useful for establishing priorities. Conjoint tests combinations of features together, mirroring the way real purchase decisions are made when multiple factors (including price) vary at once. The two techniques complement each other well: max-diff for prioritisation, conjoint for optimisation and prediction.
Yes, and it’s one of the most valuable attributes to include. By testing price alongside features, conjoint analysis quantifies willingness to pay and price sensitivity, identifies optimal price points, and shows how demand shifts across price levels. That enables profit optimisation, not just share maximisation.
Nº 7 · Get in touch
Ready to run a conjoint study?
If you’re facing a product, pricing or portfolio decision and need data you can act on, we can help. Get in touch to discuss your brief.
Nº 8 · Related
Related methods.
MaxDiff
Force real trade-offs to reveal what truly matters to customers.
Read moreKano analysis
Prioritise features based on what drives satisfaction, not just stated importance.
Read moreSegmentation
Define customer groups that drive decisions, not just descriptions.
Read moreKey driver analysis
Identify what actually drives outcomes, not just what correlates with them.
Read more