Learning Objectives

• Understand conjoint analysis for measuring preferences
• Calculate part-worth utilities
• Simulate market share for product configurations
• Regression analysis basics
• Understanding of trade-offs

1. What is Conjoint Analysis?

2. How Conjoint Works

1. Identify attributes and levels (e.g., Price: $10/$15/$20; Brand: Apple/Samsung/LG; Screen: 4K/8K)
2. Create product profiles - combinations of attribute levels. Full factorial (all combinations) = 3x3x2 = 18 profiles. Orthogonal design reduces to 9.
3. Respondents evaluate profiles - rank order or rate each profile
4. Estimate part-worth utilities using dummy variable regression
5. Interpret and simulate market share for any product configuration

3. Part-Worth Utilities

• Higher utility = more preferred
• Sum of utilities = total utility for a product
• Convert utilities to choice probabilities using logit model Attribute Importance: Range of utilities for an attribute / sum of all ranges. Higher range = more important attribute.

4. Applications

• New Product Design: Find optimal combination of attributes
• Pricing Strategy: Determine willingness to pay for features
• Segmentation: Different segments have different part-worths
• Market Simulation: Predict choice share for new products

Q1: What problem does conjoint analysis solve?

Measures how consumers value different product attributes and their willingness to trade off between them. For example: will consumers pay $5 more for better battery life? Q2: What is a part-worth utility?

Numerical value representing preference for a specific attribute level. Example: Apple = +2.5, Samsung = +0.3, LG = -0.8. Higher = more preferred. Q3: What is orthogonal design and why use it?

Statistical design where attribute levels are uncorrelated, allowing independent estimation of each attribute's effect. Reduces number of profiles from full factorial (e.g., 27 to 9) while preserving estimation ability. Q4: How to calculate attribute importance?

Range of part-worths for attribute A / sum of ranges across all attributes. Example: Price range = 3.0, Brand range = 1.5, Screen range = 0.5. Price importance = 3.0/(3.0+1.5+0.5) = 60%. Q5: What is willingness to pay (WTP) in conjoint?

How much utility a feature adds, converted to monetary terms using the price utility. If brand adds +2.0 utility and $1 price reduces -0.5 utility, WTP for brand =$2.0/0.5 = $4. Join Discord PreviousCluster AnalysisNextSegmentation, Targeting & Positioning (STP)