Karakaya et al 2011 - Overview

Summary

This paper builds an agent-based simulation of a monopolistic market for a technological product to assess the efficiency and profitability of different marketing strategies (price, promotion, quality, opinion leader targeting) in the presence of word-of-mouth effects. The key finding is that product quality is the most significant factor affecting profitability when WOM is active, because WOM amplifies quality signals.

Overview

Citation: Karakaya, C., Badur, B., & Aytekin, C. (2011). Analyzing the Effectiveness of Marketing Strategies in the Presence of Word of Mouth: Agent-Based Modeling Approach. Journal of Marketing Research and Case Studies, Vol. 2011, Article ID 421059.

Research question: How do different marketing strategies (varying price, promotion, quality, and opinion leader targeting) affect firm profitability when consumers are subject to word-of-mouth effects?

Approach: ABM simulation of N=1000 consumers connected through a social network based on preference similarity, with M=200 opinion leaders, over T=20 time steps, replicated 100 times per experiment.

Main Content

Model Components

The model has four key components:

1. Consumer agents with heterogeneous preferences and sensitivities (see Consumer Utility Function Components)
2. Social network formed by preference similarity (see Social Network Formation in Consumer Markets)
3. Opinion leaders with amplified WOM influence (see Opinion Leaders and Social Influence)
4. Purchase decisions via utility threshold + logit function (see Logit Purchase Decision Model)

Consumer Utility Function

The total utility for consumer $i$ is:

$$U_i=U_{i1}+U_{i2}+U_{i3}+U_{i4}$$

where the four components cover quality, promotion, WOM, and price respectively. See Consumer Utility Function Components for full formal specification.

Experimental Design

Five decision variables are manipulated:

• Price of the product
• Promotion intensity
• Two product design parameters (quality attributes $A_1$, $A_2$)
• Number of opinion leaders targeted

Each experiment varies one parameter while holding others constant, and is run with and without WOM to isolate the WOM effect. See Population Initialization and Parameter Sensitivity for parameter tables.

Key Findings

1. Quality dominates under WOM: When WOM is active, product quality is the most significant factor affecting profitability — positive WOM from satisfied customers amplifies quality signals
2. Price sensitivity with WOM: Lowering price increases sales but negative WOM from dissatisfied consumers can offset gains
3. Promotion has diminishing returns: Increasing promotion has limited additional effect when WOM is present because organic WOM substitutes for paid promotion
4. Opinion leader targeting matters: The number of targeted opinion leaders significantly affects diffusion speed but has complex interactions with other variables

Connections

• This paper operationalizes the general ABM framework from ABM Methodology and Principles in a marketing context
• The social network formation mechanism (Social Network Formation in Consumer Markets) is based on preference similarity, connecting to Heterogeneity in Agent Models
• The experimental design provides a template for Population Initialization and Parameter Sensitivity

See Also

• Consumer Utility Function Components — the 4-component utility model
• Logit Purchase Decision Model — the stochastic purchase decision
• Word of Mouth Mechanisms — how WOM spreads through the network
• Social Network Formation in Consumer Markets — preference-similarity network construction
• Opinion Leaders and Social Influence — amplified WOM from targeted opinion leaders
• Population Initialization and Parameter Sensitivity — experimental calibration design with parameter tables
• ABM in Marketing Strategy — broader context of ABM in marketing